Downloaded from www.jayaram.com.

np
Computer graphics.
Book: computer Graphics Motivation of CG:
Author: Donald Heam M Pauline baker. - Appealing picture produce.
- Humans respond better to pictorial information.
Date: 2065/4/12 - Human brain recognizes visual patterns.
- “If it looks right , it is right”, Jim Blinn (CG pioneer)
What is computer Graphics? Reasons to study CG:
- information presentation.
- Use a computer to create pictures. - Job in CG ( Games, movies etc)
- Started early 60s . Ivan Sutherland (MIT) - New medium for artistic expression.
- CG has many concept. - Communicate ideas better.
- Computer scientists create libraries, tools that artist /non - Take animation course.
techines can use to create pretty pictures.
- Artists use CG tools to creates pretty Pictures. Use of CG:
CG tools: - Art. Entertainment, publishing
- CG tools include hardware and software tools. - Movies , TV, books, magazines, game
Hardware Tools: - Image processing
- Output devices video monitor , printer - Alter images, remove noise
- Input devices keyboard , mouse , touch panel, - Processing monitoring
- Graphics cards/ Accelerator - Large systems or plants
Software tools: - Display simulations:
- Flight simulators, virtual worlds.
- Operating system - Computer- aided design, architecture, electric circuit
- Complier design
- Editor - Scientific analysis and visualization.
- Debuggers - Molecular biology(human brain), matlab,
- Graphics library
CG Example:
CG libraries: - Biggest CG consumers today are:
- Functions/ routine to draw line or circle etc. - Movies (Hollywood)
- Ellaborate : pull -down menu, 3D coordinate system etc. - Computer Games: eg Mdden NFL football 2004.
1
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
- Animated movies: E.g Toy story, e.g Finding Nemo - The beginning of modern interactive graphics, however
- Special effects: e.g spider man 2,Spider man 3,Matrix were found in IVAN SUTHER LAND’S seminal doctoral
Reloaded. work on the sketchpad drawing systems. He introduced
Elements of CG: data structures for storing symbol hierarchies built up via
- Poly lines connected st lines( edger, vertices) replication of standard components ( used for drawing
- Text; font, typeface circuit symbols). He also developed interaction technique
- Filled regions; colors , patterns. that use the keyboard ad light pen for making choices,
- Raster images: pixels have values (pixmax) pointing and drawing and formulating many other ides and
technique still in use today.
- By the mid- sixties, a number of research projects and
Date:2065/4/13 commercial products had appeared as the potentially of
CAD activities in computer, automobile and aerospace
1.1 HISTROY OF COMPUTER GRAPHICS: The grew enormously for automating drafting-insensitive
fundamental principal and technique derive in the past are activities. The general motor system for automobile design
still applicable in today’s computer graphics technology and and Intek Digitek system for lens design were pioneers in
generally will be applicable in the future too. showing the efforts utilizing graphics interaction in the
The history of the computer graphics can be study interactive cycles common in engineering.
as a chronical development of hardware and software. The - Due to the high cost of graphics hardware , expensive
evolution of graphics under various terms are expect on computing resources, difficulty in wring large interactive
following points. program and due to many other reasons the human-
computer interaction was still done primarily in both mode
- Crude plotting on hardcopy devices such as teletypes and using punched cards. After the advent of graphics based
line printers dates from the early days of computing. personal computers such as Apple Macintosh and IBM
- The whirlwind computer developed in 1950 at PC, the costs of both hardware and software droven down.
Massachusetts institute of Technology ( MIT) had Millions of graphics computer were sold exclusively for
computer-drive CRT displays for output, both for operator offices and home. Thus the interactive graphics (GUI) as
use and for cameras producing hard copy. “the window on the computer” became an integral part of
- The SAGE air-defence system developed in the middle PC featuring graphical interaction.
1950’s was the first to use command and control CRT The reason for making interactive graphics
displays console on which operators identified targets affordable was the advent of direct-view storage tube (DVST)
with the light pens. which replace buffer and refresh process and eliminated all
2
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
flicker in the system. Before this, buffer memory and
processors were enough only to refresh at 30HZ and only few Date: 2065/4/14
thousand lines could be drawn without noticeable flicker.
Application of computer graphics:
- Another major hardware advance of late sixties was Computer graphics started with the display of
attaching a display is a minicomputer, relieving heavy data or hard copy plotter and CRT screens had grown include the
demands of refreshed display devices (like user- creation, storage and manipulation of mode is of images of
interaction handling and updating image on the screen) objects. These models come from a diverse set of fields and
with the central-time sharing computer. include physical mathematical, engineering , architectural,
- In 1968, another such devices was invented. The refresh natural phenomena and so on. There is virtually no area in which
display hardware for geometric transformations could graphical displays cannot be used to some advantages.
scale, rotate and translate points and lines on the screen at Today al most all application
real time; perform the 2D and 3D clipping and could programs even for manipulating text or numerical data uses
produce parallel and perspective projections. graphics extensively in user interface for visualizing and
- The development of inexpensive raster graphics, based on manipulating the application specific objects. Graphical
television technology in early seventies contributed more interaction has replaced textual interaction with alphanumeric
to the growth of the field. The Raster displays stores terminal. Even people who do not use computer encounter
displays stores display primitives (lines, characters or computer graphics in TV commercial or special effects. It is an
areas) in a refresh buffer. integral part of all computer user interfaces and is indepensible
The development of graphics cannot be for visualizing 2D, 3D and higher- dimensional objects. We find
stand-off without the study of graphics input technology. The computer graphics used in a diverse areas as science, engineering
clumsy, fragile light pen has been replaced by mouse, the tablet, , medicine business, industry, government, art, entertainment ,
touch panel, digitizers and other devices. Interaction to computer education and others.
using devices require no knowledge of programming and only a
little keyboard use; the use makes choices by selecting menus, COMPUTER AIDED DESING (CAD):
icons , check options, places predefined symbols on screen and In CAD, interactive graphics is used to design components and
draws by indicating consecutive end points to be connected by systems of mechanical, electrical, electromechanical and
lines or interpolated by smooth curves and fills closed areas electronic devices including structures such as buildings,
bounded by polygons or point contours with shades of gray, automobile bodies, airplane, VLSI chips, optical systems and
colors on various patterns. Now computer graphics have become telephone and computer networks. The emphasis is on
integral and infamous technology in the computers system. interacting with a computer-based model of the component or
3
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
system being designed in order to test, for example its structural,
electrical or thermal properties. The model is interpreted by a Date: 2065/4/15
simulator that feeds back the behavior of system to the user for Computer Art:
further interactive design and test cycles. Computer graphics is used to generate arts. They are widely
Some mechanical parts are manufactured by describing how used in both fine art and commercial art applications. Fine arts is
the surfaces are to be formed with machine tools. Numerically drawn by artist hand and this kind of art is perfect to the artist
controlled machine tools are then set up to manufacture the parts skill. Artist use a variety of computer methods including special-
according to these construction layouts. purpose hardware, artists paints brush program, other paint
Architects are interactive graphics method to layout floor packages, specially developed software. Mathematics packages,
plans that shows positioning of rooms, doors, windows, stairs, CAD packages, desktop publishing software and animation
shelves and other building features. An electrical designer then packages providing facilities.
try out arrangements for wiring, electrical outlets and other Moreover, artists uses a touchpad or a stylus or digitizer to
system to determine space utilization on a building. The realistic draw pictures. The movement of object is captured by some
displays then allows architects and their clients to study input hardware. These arts are usually generated by using
appearance of building and even go for a simulated “walk” mathematical functions or algorithms.
through rooms or around building. Computer art is not as realistic as fine arts. Mostly the
commercial art is used to produce animations to demonstrate or
PRESENTATION GRAPHICS: present commercial products to the public. Find artists uses a
Another major application areas of computer graphics is the variety of computer techniques to produce images. These images
“Presentation Graphics”. Presentation Graphics are used to are created using a combination of 3D modeling package, texture
provide illustrations for reports or to generate transparencies for mapping, drawing programs and CAD software.
use with graphics. These technique for generating electronic images are
Presentation Graphics is commonly used to summarize also applied in commercial art for logos and other design, page
financial, statistical, mathematical, scientific and economic data layouts combining text and graphics, TV advertising sports, and
for research reports, managerial reports and other types of other areas. Animations are also used frequently in advertising
reports. Typical examples are bar charts, line graphs, surface and TV commercial and produce frame by frame, where each
graphs, pie charts and other displays showing relationship frame of the motion is rendered and saved as an image file.
between multiple variables. A common graphics method exployed in many
The 3D graphics are usually used simply for effects, they can commercials is morphing, where one object is transformed into
provide a more digramatic or more attractive presentation of data another.
relationship.
4
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Education and training: to visualize large amount of information graphical computer
Computer graphics is used in education and systems are used.
training for making it more effective and more illustrative. E.g if
a teacher is to teach bonding of molecules or electron jump form Image Processing: Image can be created using simple point
higher energy state to lower energy state or the structure of gene. program or can be fed into computer by scanning the image.
Then he can demonstrate these concepts using computer graphics These picture/ images need to be changed to improve the quality.
software or presentations. Form image/pattern recognition systems, images need to be
Another example could be taken for surgery. A changed in specified format so that the system can recognize the
student can learn surgery using data gloves and realistic meaning of the picture. For example scanners with OCR features
computer graphics. This way the cost of education will be low must have letters similar to standard font set.
and risk of human life as well. Other examples could be flight
simulator and driving simulator for pilot and driving training. Graphical user Interface: GUIs have become key factors for
Modes of physical systems, physiological systems, the success of the software or operating system. GUI uses
population trends or equipments such as the color coded diagram graphical objects called gizmos to represent certain objects or
helping trainees to understand the operation of the system. process involved in human computer communication for virtual
purpose. Lots of aesthetics (colors) and psychological analysis
Entertainment: Computer graphics methods are new have been done to create user friendly GUI. The most popular
commonly used in making motion pictures, music videos and TV GUI is windows based GUI. The GUI creators are putting having
shows. Images are drawn in wire-frame form and will be shaded emphasis on 3D GUI creation.
with rendering methods to produce solid surfaces. Music videos
use graphics in several ways. Graphics objects can be combined 2.0 HARDWARE CONCEPT: Input device are used to feed
with the line action. data or information into a computer system. They are usually
Computer graphics are also used to introduce virtual used to provide input to the computer upon which reaction,
characters to movies like character in “Lord of the Rings”. outputs are generated. Data input devices like keyboards are used
to provide additional data to the computers whereas pointing and
Visualization: Some methods generate very large amount of selection devices like mouse , light pens, touch panels are used to
data/information for example a survey of one lakh people’s provide visual and indication-input to the application.
choice for using different toothpaste generates large amount of
data. Analysing the property of the whole amount of data. 2.1 Keyboard, Mouse , Light pen, Touch screen and Tablet
Analysing the property of the whole amount of data is very input hardware:
difficult. If we try to locate each and every data value. Therefore
5
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Keyboard: Keyboard is a primary serial input device. For each Mechanical mouse contains two rollers. One rollers count
key press, a keyboard senses specific codes(American standard horizontal displacement and another roller counts vertical
code for Information Interchange, ASCII) to the computes. displacement. The motion of the roller in the base of mechanical
Keyboards are also capable of sending/coding combinations of mouse is converted to digital values that are used to determine
key press and special keys like function keys. The coding for the the direction and magnetic of movement.
combination of key pressing is called “chording”. It is useful for
expert users in giving exact commands to the computer. Cursor- Optical mouse uses a special mouse pad with a matrix of Black
controlled keys and function keys are used for doing operations and white grid of lines. The matrix reflects light from the LED
in a single keystroke. on the bottom of the mouse. The intensity/quality of reflected
Other types of cursor-pointer devices are trackball or joystick. light varies as Black and white lines reflects it alternatively. The
A numeric keypad is after included on the keyboard for fast entry alternate reflections are used to calculate the relative position
of numeric data. change.
Several different technologies are used to detect a key
depression including mechanical contact closure, change in Advantages:
capacitance and magnetic coupling. Chording is not possible • None of the part of computer screen is obscure (hidden
with standard “coded keyboard”, which returns only an ASCII from head).
code per keystrokes and returns nothing if two keys are pressed • Movement becomes easier as had rests at a desk for
simultaneously. An unencoded keyboard can identify of all key movement.
press simultaneously allowing chording. • Direct-point- and –click.
• Easy to use and understand.
Mouse (Mechanical and optical): Mouse is a serial input
device used to select object movement in graphics system. It Light Pens: Light pens are pencil shaped devices used to select
converts horizontal movement into vertical movement. Primary screen positions by detecting the light coming from points on the
principle of curser movement is to detect the amount of CRT screen. They are sensitive to the short burst of light emitted
displacement of mouse in horizontal plane. This displacement is from the phosphor coating as the instant electron beam strikes a
mapped into screen. Generally the mapping ratio is 1:8 . particular point. Other light sources , such as background light in
According to their working mechanism tow types of mouse are the room are usually not detected by a light pen.
there. They are: An activated light pen pointed at a spot on the screen as the
• Mechanical mouse and electron beam light. Some of that spot generates an electrical
• Optical Mouse. pulse that causes the co-ordinate position of the electron beam to
be recorded.
6
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Disadvantages: are interrupted and reflected back from their mid wave. The
• When used for long time, leads to an arm pain. sensors on the X-axis and Y-axis sides measured the time,
• Some part of the screen will be obscure by hand. the sound waves took to get reflected back and estimate the
• Cannot read black pixel position. touch position.
• Reading might be errorous if background light has very - An optical touch panel uses Infra red (IR) emitters and
high intensity. sensor on X and Y-axis sides, emitter are placed and on the
apposite sides sensors are placed when a user touches the
screen, the point will interrupt two IR beam from X and y-
Date:2065/4/16 axis. This is detected by IR sensors and the touch position
is calculated.
Touch-Panel: Touch panels are a sensitive surface that is used Advantages:
to point directly. The panel can be touched by finger or any other • Touch panel can be mounted an display screen leaving
object like stylus Transparent touch panels are integrated with more space on desktop but mouse, joystick etc. take some
computer monitor for the manipulation of information display. space.
A basic touch panel senses voltage drop when a user
touches the panel. It knows where the voltage has dropped and Tablets (Electronic, Sonic and Resistive ): A common device
accordingly calculates the touch position. for drawing, painting or interactively selecting co-ordinate
- Resistant based touch panel uses two substances of glass position on an object is a digitizer. Typically a digitizer is used to
or plastic separates by insulators. When a user touches the scan over a drawing on object and to input a set of discrete co-
panel these two substances are connected and the resistance ordinate position which can be joined with straight line segments
at that point will drop down. The resistance drop is sensed to approximate the curve surface shape.
and the touch position is calculated. One type of digitizer is Graphics tablets (data tablets),
- In capacitance based touch panel, some charge is spread which is used to input two dimensional (2D) co-ordinate by
on the screen. When user touches the panel the charge is activity a hand curser or stylus as selected position, flat
drawn by finger form each side proportionally. The sensor surface. A had curser contains cross hairs for sighting position,
calculates the charge in frequency of charge/voltage and while a stylus is a pencil shaped device that is pointed to the
find out the touch position. position of the tablet. Many graphics tablets are constructed with
- In Acoustic(Sound) based touch panel uses very high a rectangular grid of wires embedded on the tablet surface.
frequency (5 Mhz). The sound waves are generated from Electromagnetic pulses are generated in sequence along the
X-axis and Y-axis sides are reflected from opposite side of wires and an electric signal is induced in a wire coil. In an
either axis. When user touches the panel, the sound waves activated stylus or hand curser to record tablet position.
7
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Acoustic (Sonic) tablets use sound waves to detect a stylus
System Display
position. Either strip microphones or point microphones are used CPU Memory Processor
to detect the sound emitted by an electrical spark from a stylus (Graphic
controller)
tip. The position of the stylus is calculated by calculating the
arrival time of generated sound at different microphone position.

Application:
- Digitizing a drawing in a book. System bus

Date: 2065/4/19

2.2 Raster and vector display architecture: I/O Device

Fig. Architecture of a simple random scna system
Random Scan (or vector display system or Stoke writing or
Calligraphic systems): Display Other
Commman state

- In a random scan display, an image on the screen is drawn

with one line at time and for this reason it is also called
vector display. Display
Controller
- A typical organization a vector system is shown below.
- An application program is input and stored in the system System bus
memory along with the graphic package.
-
JAYA
RAM

Mouse

Keyboard

8
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
- It sends digital and points co-ordinates to a vector
generator that converts digital co-ordinates values to
analog voltages for beam deflection circuits that display an
electron beam writing on CRT phosphor coating.
- The main principal of the vector system is that the beam is
deflected form end point to end point as detected by
arbitrary order of the display commands term as random
scan.
- Since the light out put of phosphor decays in tens or
hundreds of microseconds, the display processor must
cycle through the display list to refresh the phosphor at
least 30 times per seconds (Hz) to have flicker hence the
buffer holding display list is usually called a refresh
buffer. (Note: jump instruction in the figure above in
Fig. Random scan system drawing through the component refresh buffer is to provide cyclic refresh.
lines of an object in any order specified. - A CRT beam in this system is adjusted in such a way that
electron beam only hits the spot where the graphics is to
- Graphics commands in the application program are be drawn.
translated by the graphic package in to a display file stored - Thus the refresh rate in this system depends upon the
in the system memory. number of lines to be displayed.
- This display file is then accessed by the display processor - Random scan display are design to draw all the component
to refresh the screen. lines of pictures 30 to 60 times per seconds.
- Sometimes the display processor in a random scan is - Vector display system are mostly used for line drawing
referred to as a display processing unit or a graphics applications and can not display realistic shaded scenes.
controller. - It produces smooth line drawing because the CRT beam
- The buffer stores the computer produce display list or directly follows the line path definitions that are stored in
display program which contains points and line plotting the form of line drawing commands.
commands with (x,y) and end point co-ordinates as well as Disadvantages:
character plotting commands. - When the number of command in the buffer goes high the
- The commands for plotting points lines and characters are system take long time to process and draw pictures.
interpreted by the display processor. - Cannot apply shading features.
9
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
- Stored intensity values are retrieved form the refresh
Date: 2065/4/20 buffer and painted on the screen on scan line (one row at a
Raster Display: time).
- It is suited for realistic scenes containing shading and
color pattern.
- In monochrome monitor frame buffer consists of one bit
per each pixel and for color monitor frame buffer consists
of 24 bits for each pixel.
- The refresh rate for this system is at the rate of 60 to 80
frame per second. i.e refresh rate are described in units of
cycle per second.

Raster Scan Display:

Diplay
System Processor
CPU Memory (Graphic
controlled)
Fig. A raster scan system displays an object as a set of discrete Moniter

points across each scan line.

- The raster graphics developed in early 70’s. System bus

- It is common CRT monitor.
- It is based on TV technology.
- In this system, electron beam swaps across the screen one
row at a time form top to bottom.
I/O Device
- Beam intensity vary (on or off) by the movement of
Fig: Architecture of simple random-scan system
electron beam across each room.
- Picture definitions are stored in refresh buffer or frame
buffer which holds the setup intensity values for all the
screen point also known as pixel. (pel)

10
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
- Scan line are labeled from Ymax at the top of screen to zero
system Frame Video
CPU
memroy Buffer Controller at the bottom.
- Scan line are labeled from 0 to Xmax .

Date: 2065/4/21
Video Controller:
System bus Raster-scan
Generatro

I/O Device x y
Registor Registor
Fig. Architecture of a raster system with a fixed positon of
the system memory requred for the frame buffer

- Interactive raster graphic generally employ several Memory Pixel

Intensity
processing units. Address Registor

- In addition to the CPU a special buffer processor called

video controller or display controller is used to control the
operations of display device. Frame buffer
- Frame buffer can be any where in the system memory, and
the video controller accesses the frame buffer to refresh Fig. Basic video controller Refresh operation.
the screen.
- A fixed area of system memory is reserved for the frame
buffer. - Two register are used to store the co-ordinates of the
- Video controller direct access the frame buffer. screen pixel.
- Frame buffer location and corresponding screen position - Initially the X register is set 0 and Y register is set to ymax .
are reference in Cartesian co-ordinate. - The value stored in the frame buffer for this pixel position
- Generally the co-ordinate origin is defined at the lower left is then retrieved and then used to set the intensity of CRT
screen corner. beam.
- Positive X value increases to the right and positive Y
value increases from bottom to top. 11
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
- Then the x register is increased by one and process - Video controller also contain a look up table instead of
repeated for the next pixel in the top scan line. controlling CRT beam intensity directly. This provides a
- This process is repeated for each pixel along line. fast mechanism for changing screen intensity values.
- When the last pixel of the top scan line has been processed
the x register is reset to zero and y register is decremented Date: 2065/4/22
by 1 pixels on the screen lines are processed against Random Scan display Processor:
inturn; and the process is repeated for each successive scan - Figure given below shows the architecture of raster
line. graphic system with a display processor.
- After cycling though each pixel along the bottom scan line - It contains a separate display processor the purpose of the
(y =0) the video controller resets the register the first pixel display processor is to free the CPU form graphics loads.
poison on the top scan line and refresh process starts over. It is also called graphic controller or display co-processor.
- Screen must be refresh at least at the rate of 60 frames per Display
Frame Video
processor
second. memory
Buffer Controller

- To speed of the pixel processing video controller can Monitor

retrieve multiple pixel values from the refresh buffer on
each pass. The multiple pixel intensity are then stored in a CPU Display System
processor Memory
separate register and used to controlled the CRT bit
intensity for a group of adjacent pixel. When that group of
pixel has been processes the next block of pixel values is
retrieved from the frame buffer.
- Besides these refresh operation video controller also
System bus
performs different operation video controller retrieved
pixel intensity from different memory area on different
refresh cycle.
- In high quality system, for example , two frame buffers are I/O Device
often provided so that gun buffer can be used for Fig. Architecture of raster graphics system with a display
refreshing while the other is being filled with intensity processor
values. The system memory holds the data plus those program that
- This provides a fast mechanism for generating real time execute on CPU: the application program, graphics package and
animations seans different views of moving object can be operation system. Similarly display processor holds data plus the
successively loaded in the refresh buffer. programs that perform scan conversion and the raster operation.
12
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
The frame buffer contain the displayable image created by the 1. Since the picture definition is stored as the set of line-drawing
scan conversion and raster operation. instructions and not as a set of intensity values foe all screen
- It has also a separate display processor or memory area in points, vector displays generally have higher resolution than
addition to system memory. raster systems. For these reasons, random-scan systems are
- The major task of display processor is digitizing a picture designed for line-drawing applications and cannot display
definition given in an application program into a set of realistic shaded scenes. Moreover, vector displays produce
pixel intensity values for storage in the frame buffer. smooth line drawings because CRT beam directly follows a line
- This process is called scan conversion. path but raster system in contrast produces jagged lines that are
- Graphic commands specifying straight line and another plotted as a discrete point sets.
geometric objects are scan converted into a set of discrete Figure below illustrates this:
intensity point.
- Can converting a straight line sequence means that we
have to locate the pixel position close to the line path and
store the intensity for each position in a frame buffer.
- Similar methods are used for scan converting curved lines
and polygon outlines.
- Character can be defined with rectangular grids or they
can be defined as curved outlines.
- The array size of character grids can vary from about 5 by
7 to 9 by 12 or more for high quality display.
- Character grid is display by superimposing the
rectangular grid pattern into the frame buffer at a
specified co-ordinate position.
- Display processor is also design to perform the number of
additional operation.
- It is used for various liens style (Dashed, dotted or solid),
displaying color areas and perform certain transformation Fig.: Various Scan Methods
and manipulation on displayed objects.
2. Raster scan is easier and less expensive to implement than in
Comparison between raster and vector graphics:

13
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
random scan system, whose vector generators must be highly 5. Another drawback of raster system arises from the nature of
accurate to provide linearity and repeatability of beam’s raster itself. A vector system can draw continuous, smooth line
deflection. (curves) from any point on the CRT face to other, the raster
system displays these lines (curves) mathematically using
3. One major advantages of raster graphics over vector graphics various algorithms causing problems of “jaggering” or
includes the ability to display areas filled with solid colors or “staircasing”.
patterns. Moreover, the refresh process is independent of
complexity (i.e. no. of lines) of image and each pixel in buffer
can be read out on each refresh cycles, allowing flicker free
display.

4. Major disadvantage of raster system compared to vector

system is the discrete nature of the pixel representation. This
happens due to scan-conversion of end points (vertices) into their
component pixels in frame buffer. This can be shown in figure
below: Fig. (a): Actual Line
Fig. (b): Jagged Raster Line

Date:2065/4/23

Architecture of Graphical display terminals:

Colored CRT monitor
The CRT displays color picture by using the
combination of phosphorous that emits different color light . By
Fig (b): Scan – Converted
combining the emitted light from the different phosphorous
range of color can be generate. Two basic technique for
Fig (a): Actual thick producing color display with CRT are:
li 1. Beam penetration method:
Fig. Discrete Pixel Representation 2. Shadow-mask Method. (Roster).
Beam penetration method: This is a random scan method in this
method a line is created. CRT beam is adjusted in such a way
14
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
that electron beam only hits the spots where picture is to be screen shadow mask grid is placed whose number is same as a
drawn. Two phosphorus layer (Red and green) are adjusted such number of pixel. Three electron gun is adjusted for each
that outer layer should red and inner layer should be green. The phosphorous. The electron beams is detected and focus on the
color picture depends on how far the electron beam penetrates. shadow mask hole. The Passed from the hole activate the dot
The slow electron beam only excite outer red layer. The high triangle and produce the color spot on the screen. The color of
speed electron beam excite inner green layer. At the intermediate pixel is controlled by light of intensity. By combing three color,
speed, the combination of red and green lights are emitted to 17 million color can be obtained. For true color each pixel has 24
show two additional colors orange or yellow. bits in the frame buffer.

Electron gun
Shadow mask grid Date:2065/5/2
Chapter : 3

Line Drawing Algorithm:

Scan line Algorithm: The Cartesian slope equation of a

straight line as, y = mx+b ………………….(i)
When m represents the slope of the line and b as the y –intercept.

Magnified phosphorous
dot trigger

(x2,y2)
y2
Figure: Operation of delta shadow mask method
(x1,y1)
y1

Operation of delta-delta mask CRT. These electron guns aligned

with the triangular color dot patterns on the screen, are directed 0
x
x1 x2
to each do triangle by a shadow mask.
Suppose two end points of a lien segment at positions (x1,y1) and
It is used in raster scan system object is the set of discrete point (x2,y2) are shown in fig.
y 2 − y1
across each scan line. Three types of phosphorus dots is coated m= ……………….(ii)
in each pixel that is red, green and blue. Just behind the CRT 15 x 2 − x1
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
b = y – mx …………….(iii)
For any given x-interval △x along the line, we compute the DDA ( Digital differential Analyzer):
This algorithm samples the line at unit interval in one-coordinate
corresponding y- intercept △y form equation (ii). and determines corresponding integer values nearest the line path
△y = m △x …………………(iv) for other co-ordinates.
△x =
m The equation of the line is,
∆y Y = mx+b………………….(i)
These equations form the basis of determining deflection voltage m = y2-y1/x2-x1 ………………..(ii)
in analog devices. For any interval △x , corresponding interval is given by △y = m
△x.
Case: I
For m < 1,
Case I :
Then △x can be proportional to a small horizontal deflection m < 1 , we sample at unit x interval i.e △x = 1.
voltage and the corresponding vertical deflection is set to △y as Xk+1 = xk+1 ……………..(iii)
calculated from equation (iv). Then we compute each successive y-values,
△y = m
Case: II Yk+1 = yk + m ……………………..(iv)
For m >1
Then, △y can be set proportional to a small vertical deflection Case: II
voltage with the corresponding horizontal voltage set m > 1, we sample at unit y-interval i.e △y = 1 and compute each
proportional to △x calculated from equation (V) successive x-values.
Therefore, 1 = m △x
Case III
△x = 1/m
When m = 1, then △x = △y and then horizontal and vertical Xk+1- xk = 1/m……………..(v)
voltages equal. Yk+1+ = yk + 1…………….(vi)
Comments: Above equation hold for the lines processed from left end to
1. It uses multiplication right end. For the reverse process i. e if the line is to be
2. It uses float operation. processed form right to left then,
16
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
For m <1, △x = -1 y

Xk+1 = xk-1
Yk+1 = yk +m 18
For m >1, △y = -1 16
14
(11 ,17)
12
Yk+1 =y=k +1 ……………………(ix) 11 (10 ,15)
Xk+1= xk +1/m…………….(x) 10
8
6
Therefore , in general, 4
2
Yk+1 = y ± m x
xk+1 = xk ± 1 for m < 0 0
2 4 6 8 10 11 12 14 16 18 202224 26

yk+1= yk ± 1 S.N x y
xk+1 = xk ± 1/m for m >0 1 10 15
2 10 16
Eg. Example: Digitize a line with end points (10,15) and (15, 3 11 17
30). 4 11 18
Solution:
The slope of line is Comments:
y 2 − y1 30 − 15 15 1. It is faster to calculate pixel position.
m= = = =3 2. Due to propagation of round off errors due to successive
x 2 − x1 15 − 10 3
m >1
addition the calculated pixel may shift among from the
true line path.
So we sample at y interval. The formula is given by xk+1 = Algorithm:
xk+1/m. 1. Declare the variables, x1,y1 and x2 , y2 dx, dy ,del x, del y
as real and k as integer.
2. Perform
dx = x2-x1
dy = y2 – y1
3. Test if /dy/</dx/ then
Steps = /dx/
17
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Else steps = /dy/

4. set del x = dx/steps/ yk+1 d2

d1
del y = dy/steps yk
x= x1
y = y1
5. Plot (x, y)
6. Do for k = 1 to steps xk+1
x = x+ delx Bresenhan’s Algorithm selects the best pixel co-ordinate by
y = y +del y testing the sing of an integer parameter whose value is
proportional to the difference between the separation of two
Plot (x,y) pixel actual line path.
End do. Case I:
Date: 2065/5/3 m < 1 m >0
3.Bresenhan’s line Algorithm (BLA) : Let (xk, yk) be the pixel position determined then the next pixel
DDA includes calculation related to m and 1/m which is little to be plotted is either (xk+1,yk) or (xk+1,yk+1)
complicated. Bresenhan’s improves DDA algorithm by only
involving integer calculation. .

13 Let d1 and d2 be the seperatio of pixel position (xk+1, yk) and

12
(xk+1, yk+1) from the actual line path.
11
10 Y = mx +b
9 Then, at smapling position (xk+1)
8 Y = m(xk+1) +b
7 From fig.
8 9 10 11 12 13 14 15
D1 = y – yk
D2 = yk+1- y
D1 - d2 = (y –yk) – (yk+1-y)
Let us define a decision parameter Pk for the kth step by
18
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Pk = △x (d1- d2) Therefore, yk+1 = yk+1…………..(v)
△x > 0 , Pk+1 = pk + 2 △y – 2 △x …………(vi)
Therefore, Pk <0 if d1 < d2 Therefore, Initial decision parameter.
Pk ≥ 0 , if d1 > d2 P0 = 2 △y xo – 2 △x yo + c [ from (i)]
= 2 △y xo- 2 △xyo +( 2m+ 2b-1)△x
Pk = △x{ y-yk –(yk+1 – y )} = 2△y xo – 2 △x yo + 2 m △x + 2b △x – △x
= △x{ y-yk- yk-1+y} since, yk+1 = yk+1 = 2 △y xo- 2△x yo + 2. (△y/△x).△x + 2 ( yo – mxo)△x – △x
= △x{ 2{ m(xk+1)+b} – 2 yk – 1} = 2 △y xo - 2 △x yo + 2 △y + 2 △xyo+ 2 △y/△x. xo △x – △x
Px = △x{2mxk + 2m + 2b-2yk-1} since, xk+1= xk+1
Px = 2mxk △x – 2yk △x +(2m+2b-1) △x Po = 2 △y – △x
Px = 2. (△y/△x).xk △x -2yk △x +( 2m+2b-1)
#. Digitize a line with end points ( 15, 18) and (10,15) using
Px = 2 △y xk – 2 △x yk +c BLA
Solution:
△y = /15-18/ = 3
Where, C = (2m+2b-1) △x is a constant.
Now, for next step, △x = /10-15/ = 5
m < 1, we sample at x direction.
Pk+1 = 2 △x xk+1 – 2 △x yk+1 +c………….(ii)
From (i) and (ii) First pixel ,(10, 15)

Pk+1 – Pk = 2 △y (xk+1- xk) – 2 △x ( yk+1- yk) P0 = 2 △y – △x

= 2 × 3- 5
i.e pk+1 = pk +2 △y – 2 △x ( yk+1 –yk) =1
Where, We know that ,
Yk+1 – yk = ‘o’ or ‘1’ If pk < 0
If pk< 0 , then we plot lower pixel
Yk+1 = yk ………………(iii) Pk+1 = pk+2△y
yk+1 = yk
Pk+1 = pk +2 △y ………….(iv) if p ≥ 0

If pk ≥ 0 then we plot upper pixel. Pk+1 = pk + 2 △y – 2 △x

19
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Yk+1 = yk+1

We have pk ≥ 0,
So, p1 = p0 + 2 △y – 2 △x 20

=1+2x3–2x5 19
= -3 18

17

P2 = p1+ 2 △y 16
11 12 13 14 15
= - 3+2 x 3
=3 # Digitize a line with end points ( 15, 15 ) and (10, 18 ) use BLE
and DDA.
P3 = p2 + 2△y – 2 △x Solution:
= 3+ 2 x 3 – 2 x 5
= 9 – 10 △y = /15-18/ = 3
= -1 △y = / 10-15/ = 5
m = 3/ 5 = 0.6
K Pk xk+1 , yk+1 m < 1, we sample at x direction,
0 1 11, 16 First pixel , ( 10, 15) Note:
1 -3 12, 16
P0 = 2 △y– △x For, Pk ≥ 0 ,
2 3 13, 17
=2×3–5 Pk+1 = pk+ 2△y – 2△x
3 -1 14, 17
=1 yk+1 = yk-1 ( when the slope is –ve )
4 5 15, 18
We know that, Pk < 0
If pk ≥ 0
Pk+1 = pk + 2△y
P1 = po + 2 △y – 2 △x Yk+1 = yk
=1+2x3–2x5
= -3
Pk < 0
P2 = pk + 2 △y.
= -3 + 2 x 3
20 =3
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
P3 = p2 + 2 △y – 2 △x 2. Load (x0, y0) into the frame buffer; that is plot the first
= 3 + 2×3 – 2× 5 point.
= -1
Pk < 0 3. Calculate constants △x, △y , 2 △y and 2 △y – 2 △x and
obtained the starting value for the decision parameter as ;
P4 = p3 + 2 △y
= -1+ 2×3 po = 2 △y – △x
=5
4. At each xk along the line, starting at k= 0 , perform the
following test:
K pk xk+1 yk+1
0 1 11 17
1 -3 12 17 If pk <0 the next point to plot is ( xk+1, yk) and pk+1 = pk +2 △y.
2 3 13 16 Otherwise , the next point to plot is ( xk+1, yk+1) and pk+1= pk+ 2
3 -1 14 16
4 5 15 15 △y – 2 △x

5. Repeat step 4 △x times.

20
# Digitize a line with end point s( 20, 10) and (30, 18). The slope
19
of the line is 0.8 using BLE.
18

17
16
15
Case:II m > 1
10 11 12 13 14 15 16

Date: 2065/5/4 d1 d2
Bresenham’s line algorithm: yk+1
1. Input the tow line end points and store the end point in (x0,
y0)

xk x xk+1
21
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
= △y ( x- xk+1+x)
= △y ( 2x – xk+1 + x)
= △y ( 2x – 2xk + 1) [ since, xk+1 = xk+1]
⎛ y k +1 − b ⎞
yk+3
= △y{2 ⎜ ⎟ − 2xk − 1 }
y k+2 ⎝ m ⎠
y k+1
= △y/m{2(yk+1-b) – 2xkm – m}
yk
xk xk+ 1 xk+ 2 = △x { 2 ( yk+1 –b) – 2 xk (△y/△x) – △y/ △x}
= 2 △x( yk+1-b)- 2 △y . xk – △y
Let (xk, yk) be the pixel position determined then the next pixel = 2 △x yk – 2 △y xk + 2(1-b) △x – △y
to the plotted is either ( xk+1, yk+1) or ( xk, yk+1) Px = 2 △x yk – 2△y xk + c …………………..(ii)
Let d1 and d2 be the separation of the pixel positons ( xk, yk+1) Let C = 2 ( 1-b) △x – △y
and ( xk+1, yk+1) for the actual line path. Now for next step,
Y = mx +b
The actual value of x is given by Pk+1 = 2 △x yk+1 – 2 △y xk+1 + c………………….(iii)
x = (y-b)/m From equation (i) and (ii)
Sampling position at yk+1 Pk+1 – pk = 2 △x ( yk+1- yk) – 2 △y ( xk+1- xk)
Pk+1 = pk + 2 △x ( yk+1- yk) – 2 △y ( xk+1 – xk)
Form figure,
d1 = x - xk Where, , xk+1- xk = ‘0’ or ‘1’
d2 = xk+1 – x If pk ≥ 0
Let us define a decision parameter pk and Pk is defined by,
Pk+1 = pk + 2 △x – 2 △y
Pk = △y ( d1- d2) We plot , xk+1 = xk+1, yk+1 = yk+1
△y > 0 P< 0
Pk < 0 if d1 < d2 Pk+1 = Pk+ 2 △x
Pk ≥ 0, if d1 ≥ d2 Xk+1 = xk

Pk = △y (d1 – d2) For initial parameter,

22
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Po = 2 △x yo – 2 △y xo +c i. Check if (y1>y2) then,
x = x1; x1 = x2 ; x2 = t
= 2 △x yo – 2 △y xo + 2 ( 1-b) △x – △y y = y1 , y1 = y2 ; y2 = t.
= 2 △x yo – 2 △y xo + 2 △x – 2b △x – △y ii. Find initial decision parameters. P = 2dx – dy
= 2 △x yo – 2 △yxo+2△x – 2 (yo – mxo) △x – △y iii. Plot the first point (x1,y1)
iv. Repeat the following till y1 < y2
= 2 △x yo – 2 △y xo + 2 △x – 2△x yo + 2. (△y/△x).xo a. If P< 0 then , P = P+2dx.
. △x - △y Else, P = P+ 2dx-2dy
Po = 2 △x – △y X1 = x1 +a
b. Increase y1 by 1. i.e y1 = y1+1
Note: m < 1, x-direction sample. xk+1 = xk+1 c. Plot (x1,y1)
m > 1, y –direction sample. yk+1 = yk +
Circle: A cicle is defined as a set of points that ar all at a
1 Date: 2065/5/10 given distance ‘r’ from the centre position (xc, yc). This
Bresenham’s Complete algorithm: distance relationship is expressed by the pythogorean theorem
1. Input the two end points (x1,y1) and (x2,y2) in cartesian co-ordinate as (x – x1)2 + (y-y1) 2 = r2
2. Compute dx /x2 –x1/ and dy = /y2 –y1<0
3. If x2 – x1 < 0 and y2 –y1 > 0 or x2-x1>0 and y2 – y1 < 0. Methods to draw circle:
Then set a = -1, else a = 1 a. Direct method
4. If dy < dx then, b. Trigonometric method
i. If x1>x2 then, t = x1 ; x1 = x2 ; x2= t c. Midpoint Circle method.
t = y1 ; y 1 = y 2 ; y 2 = t
ii. Find initial decision parameter P = 2dy - dx Direct method:
iii. Plot the first pixel (x1, y1) X2 +y2 = ‘r2’ ; y = ± r 2 − x 2
iv. Repeat the following till /x1/ < /x2/
Trigonometric method:
a. If P< 0 then,
x = xcosθ , y = ysinθ
P = P+ 2dy
Else, P = P+ 2dy – 2dx
Bresenham’s mid point algorithm :
y = y1+a
This is also a scan conveting algorithm . The equation of the
b. Increase x1 by 1 i.e x1 = x1+1
circle with the centre (h,k) is given by , (x-h)2 +(y-k)2 = r2
c. Plot (x1>y1)
……….(i)
5. Else /m/>1 23
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
When, h = 0, x = 0 then the equation of the circle at origin
x2+y2 = r2 ……………(ii) The cicle function tests are performed for the mid position
Differentiating both sides, between pixels near the circle path at each sampling step.
2x+2y dy/dx = 0
Or dy/dx = -x/y , where , dy/dx = slope yk
Now , if dy/dx = 0, then x = 0. y k -1
If, dy/dx = -1, then x = y
x 2+ y2 = r2
Mid
(0,r) Point
(-y,x) (y,x)

xk xk+ 1 xk+ 2
(-x,y) (x,y)
Fig. Midpoint between candidate pixels at sampling position
(r,0) xk+1 along a circular path.
(-x,-y)
(x,-y) Here assume that we have just plotted the pixel (xk,yk ). We next
need to determine whether the pixel at position (xk+1,yk+1) on
(-y,-x) (y,-x) one at the position (xk+1,yk-1) is closer to the circle.

The decision parameter is the circle function which we have

Fig. Symmetry of circle evaluated in equation (iii) at the mid point between these two
pixel.
Consider circle section for x = 0, x = y0 where slope of the curve
varies from 0 to1.Calculation of circle point (x,y) in one octant I,e Pk = fcicle ( xk+1, yk-1/2)
gives the circle point shown for the other seven octants. To apply
the mid point we define a circle function as, = (xk +1)2 +(y -1/2)2 – r2 …………..(iv)
fcicle (x,y) = x2+y2 – r2 …………..(iii)
Suppose,
If pk< 0 , this mid pixel is inside the circle and the pixel on the
f(x,y) = < 0 if (x,y ) is inside the circle boundary. scan line yk is closer to the circle boundary . Otherwise the mid
= 0 if (x,y) is on the circle boundary. point is outside or on the circle boundary , and we select the
>0 , if (x,y ) is outside the circle boundary. 24 pixel yk -1, successive decision parameters can be obtained
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
using incremental calculation. We obtained a recursive For initial decision parameter.
expression for the next decision parameter by evaluation the (x0,y0) = (0,r)
circle function at sampling position. Hence , P0 = f(x0+1, yo- ½)
= f( 1, r-1/2)
Xk+1 +1 = xk +2 = 1+ (r-1/2)2 –r2
= 1 + r2 – r + 1/4 – r2
Pk+1 = fcircle (xk+1+1 ,yk+1 – ½)

= (xk+1+1)2 + ( yk+1-1/2)2 – r2 ………………(v) Po = 5/4 – r

Po = 1-r
Substituting equation (iv) from (v). P = 1-r ……….(ix) since 5 and 4 are integer values so can be
rendered off.
Pk+1 – pk = (xk+1 + 1 )2 + (yk+1- ½)2 – (xk +1)2- (yk- ½ )2 ….(VI)
Date: 2065/5/12
If pk ≥ 0 , the mid position lies on outside the circle hence, xk+1 = 2 2
# Digitize x +y = 100 in first octant.
xk+1, yk+1 = yk -1;
Given,
Now from equation (vi)
Centre = (0, 0)
xk+1 = xk +1
Radius (r ) = 10
Pk+1 = pk + (xk + 1+1)2 + (yk-1-1/2)2 – (xk+1)2 – (yk – ½ )2
We demonstrate the mid point circle algorithm by determining
= pk+(xk + 1)+ 2(xk+1). 1 + 1 + (yk -1/2)2 -2 ( yk -1/2).1 + 1 – (xk
positions along the circle octant in the first quadrant from x = 0,
+1)2 – (yk-1/2)2
to x = y.
The initial value of the decision parameter is
Pk+1 = pk + 2(xk+1- yk+1)+1………….(vii)
Po = 1-r
= 1 – 10
Where xk+1 = xk +1, yk+1 = yk – 1;
= -9
If pk < 0, the midpoint lies inside the circle,
Successive decision parameter values are positions along the
Xk+1 = xk + 1, yk+1 = yk
circle path are calculated using mid-point method as,
Substituting the value in equation (vi)
Pk+1 = pk + 2xk+1 + 1 ( Pk<0)
Pk+1 = pk + (xk+1 +1 )2 + (yk – ½)2 – (xk+1)2 – (yk – ½)2
Set , xk+1 = xk +1 , yk+1 = yk
= Pk + (xk+1)2 + 2(xk+1).1. + 1 – (xk +1)2
Pk+1 = pk + 2xk+1 – 2yk+1 + 1 ( Pk > 0)
= Pk + 2(xk +1)+1
Set xk+1 = xk + 1, yk+1 = yk – 1
= Pk + 2(xk+1)+1
Therefore, Pk+1 = pk + 2xk+1 +1 ……….(viii) 25
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
K Pk xk+1 yk+1 2xk+1 2yk+1
1 -9 (1,10) 2 20 Yk+1 = yk
1 -6 (2,10) 4 20
2 -1 (3,10) 6 20 Pk+1 = pk + 2xk+1 – 2yk+1 + 1 (pk ≥ 0)
3 6 (4,9) 8 18 Yk+1 = yk -1
4 -3 (5,9) 10 18
5 8 (6,8) 12 16 K pk xk+1 yk+1 2xk+1 2yk+1
6 5 (7,7) 14 14 0 -4 (1,5) 2 10
1 -1 (2,5) 4 10
10 2 4 (3,4) 6 8
9 3 3 (4,3) 8 6
8
7 Other pixel Actual pixel
6
(1,5) (1,5) (3,8) (x,y)
5
(-1,5) (1,8) (-x,y)
4
(-1,-5) (1,-2) (-x,-y)
3
(1,-5) (3,-2) (x,-y)
2 (5,1) (7,4) (y,x)
1
(5,-1) (7,2) (y,-x)
0
(-5,-1) (-3,2) (-y,-x)
0 1 2 3 4 5 6
(-5,1) (-3,4) (-y,x)
7

(2,5) (2,5) (4,8)

# Digitize a circle (x-2)2 + (y-3)2 = 25
(-2,5) (0,8)
Solution:
(-2,-5) (0,-2)
Center C (h,k) = ( 2,3)
(2,-5) (4,-2)
Radius (r ) = 5
(5,2) (7,5)
1st pixel (0,5)
(5,-2) (7,1)
Po = 1-1
(-5,-2) (-3,1)
= 1-5
(-5,2) (-3,5)
= -4
Pk+1 = pk + 2xk+1 + 1 (pk< 0) 26
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
(3,4) (3,4) (5,7) 8
(-3,4) (-1,7) 7
(-3,-4) (-1,-1) 6
(3,-4) (5,-1) 5
(4,3) (6,6) 4 Centre
(4,-3) (6,0)
3
(-4,-3) (-2,0)
2
(-4,3) (-2,6)
1
0
(4,3) (4,3) (6,6)
(-4,3) (-2,6) -1

(-4,-3) (-2,0) -2

(4,-3) (6,0) -3
(3,4) (5,7) -4
(3,-4) (5,-1) -5
(-3,-4) (-1,-1) -6
(-3,4) (-1,7) -7

-8

-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8

Complete गनर् बाँक छ।

Algorithm:
1. Get the input centre (xc, yc) and radius of the circle.
2. Set x = 0, and y = r.
3. Plot the first point and its symmetry at appropriate
positions by x = x+xc , y = y + yc
4. Compute the initial decision parameter. P = 1- r
5. Repeat the following till x < y.
i. Set x = x+1.
ii. Check if P< 0 then.
27
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
P = P+2x + 1 Let us define an ellipse function centered at origin by
Else, f(x, y)= b2x2 + a2y2 – a2b2 .
y = y-1 If f(x,y) < 0 , the point lies inside the ellipse.
p = p+2x – 2y +1 = 0, the point lies on the ellipse.
iii. Shift the point (x, y) and its symmetry points by x = > 0, the point lies outside the ellipse.
x+xc , y = y+yc and plot. Starting at ( 0,b) we take unit steps in x-direction until we
iv. sdf reach the boundary between region 1 region 2 ( R1 , R2 ) . Then
Ellipse: Ellipse is an elongated circle. Therefore elliptical we switch to unit steps in y-direction over the remainder of curve
curves can be generated by modifying circle –drawing in first quadrant. At each step, we need to test the value of the
procedures to take into account the different dimensions of an slope of the curve .The ellipse slope is calculated from equation :
ellipse along the major and minor axes. x2/a2 +y2/b2 = 1
The equation of ellipse in standard form is given by Differentiating both sides w.r to x
, 2x/a2 + 2y/b2. dy/dx = 0
x2/a2 + y2/b2 = 1 Dy/dx = -2b2x/2a2y
Therefore, y = +- b √(1-x2/a2) At the boundary region 1 and region 2 , dy/dx = -1 and
y 2b2x= 2a2y..
Therefore , we move out of region 1 (R1) when 2b2x ≥ 2a2y. For
R1
(0,b) region R1.
(-x,y)
(x,y) R2 b2x2+a2y2-a 2b2 = 0

2b
x
2a (a,0) yk

yk-1
(-x,-y) (x,-y)

(mid point)

This algorithm is used to display the ellipse in standard form.

The algorithm is applied throughout the 1st quadrant according xk x
k+1
to the slope of the ellipse. For the region (R1) where the slope Fig. Mid point between candidate pixel at sampling position
of the curve is less then one. We process by taking unit steps in xk+1 along in elliptical path.
x-direction and for the region (R2) we take unit steps in y-
direction. 28
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
If (xk, yk) is the pixel first plotted then the candidate pixel are = b2+a2(b2-b+1/4)- a2b2
(xk+1,yk) and (xk+1, yk-1). Let us define decision parameter at = b2+ a2b2 – a2b+1/4a2 – a2b2
the midpoint (yk-1/2) at sampling position xk +1 by Po = b2 – a2b +a2/4 …………….(v)
Pk = f (xk +1, yk-1/2)
= b2 (xk +1)2 +a2 (yk-1/2)2 – a2b2 ……..(i) b2x2+a2y2-a 2b2 = 0
Pk+1 = b2(xk+1+1)2 +a2 (yk+1-1/2)2 – a2b2 …….(ii)
Substituting (i) from (ii) yk
Pk+1 – pk = b2 { (xk+1 +1)2 – (xk+1)2 } +a2 {(yk+1-1/2)2 – (yk – ½)2 yk-1
} (mid point)

If Pk< 0, mid point lies inside the ellipse, So we plot,

(xk+1, yk) i.e xk+1 = xk +1. yk+1 = yk
Therefore, Pk+1 = pk + b2 { (xk +1)2 + 2xk+1+1 – (xk + 1)2} +
a2{(yk – ½)2 – (yk – ½)2 } xk x
k+1
Fig. Mid point between candidate pixel at sampling position
pk+1 = pk + 2b2xk+1 + b2 …………….(iii) Where,
yk-1 along an elliptical path.
xk+1 = xk + 1 , yk+1 = yk
If Pk ≥ 0, the mid point lies outside the ellipse so we plot
For region R2 ( /m/>1)
(xk+1, yk-1). i.e xk+1 = xk +1 , yk+1 = yk -1.
If (xk,yk) be the pixel just plotted then the candidate pixel at (
Therefore, Pk+1 = pk + b2 {(xk +1)2 +2 (xk +1)+1}+a2 {(yk-1-
xk,yk-1) and (xk+1, yk -1). So let us define the decision parameter
1/2)2 – (yk -1/2)2 }
at the midpoint xk+1/2 at sampling position yk-1 by
= pk + b2 {2xk+1+1} +a2 {(yk-1/2)2 – 2(yk – ½) +1 – (yk – ½)2}
Pk = f(xk+1/2, yk-1)
= pk + 2b2xk + b2- 2a2yk+1
= b2(xk +1/2)2 +a2 ( yk -1)2 – a2b2 …………(i)
Therefore, Pk+1 = pk + 2b2xk+1 – 2a2yk+1 + b2
Pk+1 = b2(xk+1+1/2)2 + a2 (yk+1-1)2 – a2b2 ………….(ii)
Where,
Substituting (i) from (ii)
Xk+1 = xk +1
Pk+1 – pk = b2 { ( xk+1+1/2)2 – (xk+1/2)2} + a2 {(yk+1-1)2 – (yk -
Yk+1 = yk -1
1)2 } …………..(iii)
Now initial decision parameter is given by is given by ,
If pk > 0 , the mid point lies outside the ellipse, so we plot
P0 = f(x0+1, y0 -1/2) (xk,yk-1)
xk+1 =xk ; yk+1 = yk-1
= f(0+1, b-1/2) Pk+1 = pk + b2{(xk+1/2)2 – (xk+1/2)2}+a2 { (yk -1 – 1)2 – (yk -1)2}
= b212+a2(b-1/2)2 – a2b2 29
= pk + a2{(yk+1)2 – 2(yk-1)+1 – (yk-1)2}
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
2
= pk+a {-2yk+1 +1} 4. Compute initial decision parameter for R1 , P1 = b2-
Pk+1 = pk - 2a2yk+1+a2 ……………(iv) Where xk+1 = xk, yk+1 = a2b2+a2/4
yk -1 5. Repeat the following till 2b2x < 2a2y
i. x = x+1
If pk ≤ 0, the midpoint lies on/inside the ellipse, so we plot (xk+1, ii. Test if P1 < 0 then
yk -1) i. e P1 = p1 +2b2x+b2
Xk+1 = xk +1 Else,
Yk+1 = yk -1 y = y-1
P1 = p1+2b2x- 2a2y = b2
Pk+1 = pk + b2 { xk +1 + ½)2 – (xk+1/2)2}+a2 {(yk-1-1)2 – (yk -1)2 }
= pk + b2{(xk+1/2)2 +2(xk +1/2)1+1-(xk +1/2)2} +a2 {(yk -1)2 – iii. Plot the points (x,y) and its symmetry points at
2(yk -1).1 + 1.-(yk -1)2 } approximate position by
= pk + b2 {2xk + 1+1}+a2 {-2y +2 +1} x = x+xc y = y+yc
Therefore, Pk+1 = pk + 2b2xk+1 – 2a2yk+1 + a2 ………...(v) 6. Compute initial decision parameter for R2
Where, xk+1 = xk+1 P2 = b2(x+0.5)2 + a2 (y-1)2 – a2b2
yk+1 = yk -1 7. Repeat the following till y>0
The initial value of initial decision parameter for region 2 is i. y = y-1
P0 = f(x0+1/2, yo-1) ii. Test if P2 > 0 then
P0 = b2 (x0+1/2)2 + a2 (yo -1)2 –a2b2 ……….(vi) P2 = p2 – 2a2y +a2
Else,
[Note: (xo,yo) for region 2 in the last point for region 1. The X = x+1
pixel for other quadrant are determined by symmetry] P2 = p2 +2b2x – 2a2y + a2
iii. Plot the points (x,y) and its symmetry points at
Date: 2065/5/16 approximate points by x = x+xc , y = y +yc
Algorithm:
1. Obtain the centre xc , yc semi-major and semi-minor axis # Digitize an ellipse (x-2)2/64 + (y+5)2/36 = 1. Using mid point
length as a and b. algorithm.
2. Set x = 0, y = b
Solution:
3. plot the point (x,y) and its symmetry points at appropriate
Center = (2,-5)
positions by x = x+xc , y = y+yc
a=8
30
b= 6
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
For region R1 Pk+1 = pk – 2a2yk+1 +a2 Where, xk+1 = xk; yk+1 = yk -1
If pk ≤ 0
First pixel is (0,6) Pk+1 = pk + 2b2xk+1 – 2a2yk+1 + a2 Where, xk+1 = xk+1 ; yk+1 = yk
Po = b2 –a2b +a2/4 –1
= 36 - 64 × 6 + 64/4
= -332 K pk (xk+1,yk+1 ) 2b2xk+1 2a2yk+1
Successive decision parameter values ad positions long the 0 -151 (8,2) 576 256
ellipse path are calculated using the midpoint method as 1 233 (8,1) 576 128
If pk < 0, pk+1 = pk + 2b2xk+1 + b2 2 745 (8,0) - -
If p ≥ 0,
Pk+1 = pk +2b2xk+1 + b2 where xk+1 = xk +1 ; yk+1 = yk -1 A plot of the selected positions around the ellipse boundary
within the first quadrant is shown in fig.

k pk (xk+1, yk+1) 2b2xk+1 2a2yk+1 7

0 -332 (1,6) 72 768 6

1 -224 (2,6) 144 768 5

2 -44 (3,6) 216 768 4

3 -208 (4,5) 288 640 3

4 -108 (5,5) 360 640 2

5 288 (6,4) 432 512 1

6 244 (7,3) 504 384 0
We now move out of region 1, since 2b x >2a2y
2 0 1 2 3 4 5 6 7 8
For region 2, the initial point is (x0,y0) = (7,3) and the initial
decision parameter is Fig. Positions along an elliptical path centerted on the origin with
P0 = b2(x0+1/2) +a2(y-1)2 –a2b2 a=8 and b = 6 using the midpoint algorithm to calculate pixel
= 36(7+1/2)2 + 64(3-1)2 – 36× 64 addresses in first quadrant.
= 36× 225/4 + 64×4 – 36×64 In the following procedure, the mid point algorithm is
= - 151 used to display an ellipse with input parameters a,b , x center and
The remaining positions along the ellipse path in the first y center. Position along the curve in the first quadrant are
quadrant are then calculated as generated and then shifted to their proper screen positions.
If pk > 0 Intensities for these positions and the symmetry positions in the
31
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
other three quadrants are loaded into the frame buffer using the
P’(x’,y’)
set pixel routine.
ty
p (x,y)

tx
Transformation: Changing co-ordinate description of an object
is called transformation.
Types:
- Rigid body transformation ( transformation without
# Translate the given points (2,5) by the translating value (3,3)
defomation in shape. )
Solution:
- Non rigid body transformation (transformation with
(x,y) = (2,5)
change in shape)
Tx = 3
Ty = 3
Basic Transformations: Other Transformation
X’ = x+ tx
1. Translation 1.reflection
Y’= y+ty
2.Reflection 2.Shearing
⎡ x'⎤ ⎡2⎤ ⎡3⎤ ⎡5⎤
3.Scaling ⎢ y '⎥ = ⎢5⎥ + ⎢3⎥ = ⎢8⎥
⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦
Two dimensional transformation:
1. Translation: Changing the co-ordinate position along a st.
line is called translation. P’(x’,y’)
If P(x,y) is taranslated to a position p’(x’,y’) by tx units parallel
to x-axis and ty units parallel to y axis then, p (x,y)
X’ = x+ tx
Y’ = y+ ty
In matrix form: Scaling: The co-ordinate position of an object by multiplying a
⎡ x ' ⎤ ⎡ x ⎤ ⎡t x ⎤ constant factor ( scaling factor) is called scaling.
⎢ y '⎥ = ⎢ y ⎥ + ⎢t ⎥
⎣ ⎦ ⎣ ⎦ ⎣ y⎦ Scaling Techniques:
i.e p’ = p+T where T is transformation matrix. a. About origin
b. About fixed point

32
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
About origin: If p(x,y) be scaled to a point p’(x’,y’) by sx times y

in x-unit and sy times in y-units then,

x’ = x. sx
y’ = y.sx
P’(x’,y’)
In matrix form:
⎡ x'⎤ ⎡ s x 0 ⎤⎡ x⎤ p (x,y)
⎢ y '⎥ = ⎢ 0 s y ⎥⎦ ⎢⎣ y ⎥⎦ (xf , yf)
⎣ ⎦ ⎣ x
i.e In matrix form,
p’ = S (sx, sy)
……….
Scaling:
P’(x’,y’)
# Scale the given triangle whose co-ordinate values are (2,3),
(1,2) , (3,4) where the scaling factor is (2,3) about (i) origin (ii)
p (x,y) about fixed point (1,2)

Rotation: Changing the co-ordinate position along a circular

Where path is called rotation.
y
S (sx, sy) is a scaling matrix
P’(x’,y’)
If sx = sy scaling type is uniform
If sx is not equal to sy scaling type is differential.
p (x,y)

c. About fixed point: If p(x,y) be scaled to a point p’(x’,y’) by

θ
sx times in x-unit. and sy time in y-unit about
(xf,yf) then α
x
o
x’ – xf = sx(x-xf)
y’ – yf = sy ( y-yf)
Types of rotation:
a. About origin
b. About any point
c. About origin.
33
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Figure: y

P’(x’,y’)
About origin: If p(x,y) is rotated to a new position p’(x’,y’) in
anti-clockwise direction by an angle θ , then (x’,y’) can be p (x,y)
calculated as following.
Let the angle made by line OP with x-axis is α and the radius of θ
circulation path is r then, α
x = r cosα (xr , yr)
y = r sinα x
Also, o x- xr
x’ = r cos(θ +α )
y’ = r sin( θ + α ) Do yourself .

x’ = r(cosθ. Cosα – sinθ. Sinα ) # Rotate the triangle having co-ordinates (1,2) , (2,3) (4,5) by 60
x’ = xcosθ – y sinθ
˚ about origin.
y’ = r( sin θ. Cosα + cos θ .sinα)
= xsinθ + y cosθ
In matrix form: Date: 2065/5/18
Matrix Representation and Homogeneous Co-ordinates:
⎡ x'⎤ ⎡cos θ − sin θ ⎤ ⎡ x ⎤
To treat all the transformation in the same manner, homogeneous
⎢ y '⎥ = ⎢ Sinθ cos θ ⎥ ⎢ y ⎥ co-ordinate are used for reapresentation. Each points (x,y) is
⎣ ⎦ ⎣ ⎦⎣ ⎦ represented by triple (x,y,h).
(x,y,h) and (x’,y’,h’) represent same point if on one is multiple
Note: θ +ve ( anticlock wise) of other.
θ –ve ( clock wise ) Ie x’/x = y’/y = h’/h
b. About any point: if (x,y) is rotated to a new position So, (x,y,h) and (x/h, y/h,1) also represent same point, for
p’(x’,y’) by an angle θ, then (x’, y’) can be calculated as simplicity we use h = 1.
following. Therefore, (x,y) in Cartesian system is represented as (x,y,1) in
Now x –xr = r cosα homogeneous co-ordinate system.
Y – yr = r sin ( θ +α ) Now (2×2) matrix is changed into 3× 3 matrix,
34 For translation:
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡ x ' ⎤ ⎡1 0 t x ⎤ ⎡ x ⎤ Let p be the point translated by T(tx1,ty1) to point p’ then,
⎢ y '⎥ = ⎢0 1 t ⎥ ⎢ y ⎥ P’ = T (tx1,ty1). P
⎢ ⎥ ⎢ y ⎥⎢ ⎥

⎢⎣ z ' ⎥⎦ ⎢⎣0 0 1 ⎥⎦ ⎢⎣ z ⎥⎦ Let p’ be the point translated by T(tx2,ty2) to point p’’ then,
P’’ = T (tx2, ty2). P’
Therefore, p’ = T(tx,ty ).p
= T (tx2,ty2).T (tx1,ty1).p
For Rotation (about origin)
Net transformation = T(tx2,ty2). T (tx1,ty1)
⎡ x'⎤ ⎡cos θ − Sinθ 0⎤ ⎡ x ⎤
⎢ y '⎥ = ⎢ Sinθ ⎡1 0 t x1 ⎤ ⎡1 0 t x1 ⎤
cos θ 0⎥⎥ ⎢⎢ y ⎥⎥
⎢ ⎥ ⎢ = ⎢⎢0 1 t y1 ⎥⎥ ⎢⎢0 1 t y1 ⎥⎥
⎢⎣ z ' ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ z ⎥⎦
⎢⎣0 0 1 ⎥⎦ ⎢⎣0 0 1 ⎥⎦
P’ = R(θ ). P
⎡1 0 fx1 + fx 2 ⎤
For Scaling : (about origin) = ⎢⎢0 1 fy1 + fy 2 ⎥⎥
⎡ x'⎤ ⎡ s x 0 0⎤ ⎡ x ⎤ ⎢⎣0 0 1 ⎥⎦
⎢ y '⎥ = ⎢ 0 sy 0⎥⎥ ⎢⎢ y ⎥⎥ = T(tx1+ tx2, ty1+ty2)
⎢ ⎥ ⎢
⎢⎣ z ' ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ z ⎥⎦ ∴ T(tx1,ty2). T(tx1,ty1) = T(tx1+ tx2, ty1+ty2)
Which demonstrates that two successive translations are
Inverse Transformation: additive.
For inverse translation:
T-1 (tx,ty )= T ( -tx , -ty )
# Prove that two successive rotations are additative. i.e R(θ2) .
For inverse Rotation: R(θ1) = R(θ1 +θ2)
R-1 (θ = R( -θ ) Solution:
Let p be the point rotated by R(θ1) to point p’ then p’ = R(θ1 )
For inverse scaling: .p
s-1(sx,sy) = S(1/sx, 1/sy) Let p’ the point rotated by R(θ2) to point p’’ then
P’’ = R(θ2). P’
# Prove that two successive translations are additive. Net transformation = R(θ2). R(θ1)
i.e T(tx, ty). T(tx2,ty2) = T(tx1 +tx2, ty1 +ty2). P’ = R(θ1+ θ2 ) p

Solution: # Prove that two successive scalings are multiplicative.

35
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
i.e S(sx2 , sy2). S(sx1,sy1) = S (sx1, sx2, sy1, sy2) ⎡ 2 0 0⎤ ⎡ 1 / 2 1/ 2 0⎤
⎢ ⎥
Solution: = ⎢⎢0 3 0⎥⎥ ⎢− 1 / 2 1/ 2 0⎥
Concatenating transformation matrices for two successive ⎢⎣0 0 1⎥⎦ ⎢ 0 0 1⎥⎦
scaling operations produces the following composite scaling ⎣
matrix. ⎡ 2/ 2 2/ 2 0⎤
⎢ ⎥
= ⎢− 3 / 2 3 / 2 0⎥
⎡s x 2 0 0⎤ ⎡ s x1 0 0⎤ ⎡ s x1 s x 2 0 0⎤ ⎢ 0 0 1⎥⎦
⎢0 ⎣
⎢ s y2 0⎥⎥ ⎢⎢ 0 s y1 0⎥⎥ = ⎢⎢ 0 s y1 s x 2 0⎥⎥ The transformation points are ,
⎢⎣ 0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦ A’ = TA
Or s(sx2,sy2) S(sx,sy) = S(sx1,sx2,Sy1,sy2) ⎡ 2/ 2 2 / 2 0⎤ ⎡ 7 ⎤
⎢ ⎥
⎢− 3 / 2 3 / 2 0⎥ ⎢⎢15⎥⎥ =
# Rotate the △ABC by 45˚ clock wise about origin and scale it ⎢ 0
⎣ 0 1⎥⎦ ⎣⎢ 1 ⎦⎥
by (2,3) about origin. B’=TB
⎡ 2/ 2 2/ 2 0⎤ ⎡5⎤
Solution: ⎢ ⎥
⎢− 3 / 2 3/ 2 0⎥ ⎢⎢8⎥⎥ =
⎢ 0 0 1⎥⎦ ⎣⎢1⎦⎥
A(7,15)
⎣
C’ = TC

B(5,8)
C (10,10) ⎡ 2/ 2 2/ 2 0⎤ ⎡10⎤ ⎡ 20 / 2 + 20 / 2 + 0 ⎤ ⎡40 / 2 ⎤
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢− 3 / 2 3/ 2 0⎥ ⎢⎢10⎥⎥ = ⎢− 30 / 2 + 30 / 2 + 0⎥ = ⎢ 0 ⎥
⎢ 0 0 1⎥⎦ ⎢⎣ 1 ⎥⎦ ⎢ 0 + 0 + 0 +1 ⎥ ⎢ 1 ⎥
Step 1: Rotation by 45˚ (clock wise) ⎣ ⎣ ⎦ ⎣ ⎦
Step-II: Scaling by (2,3) ¾
S(2,3) Note from miss:
Net transformation: = S(2,3). R(-45)
⎡2 0 0⎤ ⎡ cos 45 − Sin 45 0⎤ General pivot point rotation: We can rotate any selected point
= ⎢⎢0 3 0⎥⎥ ⎢⎢− Sin45 cos 45 0⎥⎥ (xr,yr) by following the sequence of translate – rotate –translate
⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦ operations.
36
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Step 1: Translate the object such that fixed pont moves to origin Net Transformation (T) = T(xr,yr). R(θ). T(-xr,-yr)
i.e T(-xr,-yr) . ⎡1 0 x r ⎤ ⎡cos θ − sin θ 0 ⎤ ⎡1 0 − x r ⎤
Step 2: Rotate the object coordinate origin i.e R(θ) . = ⎢⎢0 1 y r ⎥⎥ ⎢⎢ sin θ cos θ 0⎥⎥ ⎢⎢0 1 − y r ⎥⎥
⎢⎣0 0 1 ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1 ⎥⎦
Step 3: Translate back the object so that the pivot point is
returned to its original position T(xr,yr).
General Fixed point Scaling:
This transformation sequence is illustrated in fig below.
A transformation sequence to produce scaling with respect
to a selected fixed position (xf, yf) using a scaling function is
illustrated below:
(xr,yr) 1. Translate the object so that the fixed point coincides with the
co-ordinate origin.
2. Scale the object with respect to the co-ordinate origin.
3. Use the inverse translation of step 1 to return the object to its
original position.
(a) Original position of object (b) Translation of object so
And pivot point that povit point (xr,yr) The transformation sequence is illustrated in fig below:
is at origin.

(xr,yr)
(xr,yr)

(c) Rotation about origin (d) Translation of object so (a) Original position of object (b) Translate object
That the pivot point is And fixed point so that fixed point
Returned to position (xr, yr) is at origin.
(xr, yr)

37
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
y
A(7,15)

x y
( r, r)

C(10,10)
B(5,8)
x

(c) Scale object with respect (d) Translate object so Step:1

To origin the fixed point is T(-5,-8)
Returned to position Step: 2
(xf,yf) R(90˚)
Step: 3
T(5,8)
⎡1 0 x f ⎤ ⎡s x 0 0 ⎤ ⎡1 0 − x r ⎤ Step: 4
Net Transformation T = ⎢⎢0 1 y f ⎥⎥ ⎢⎢ 0 sy 0⎥⎥ ⎢⎢0 1 − y r ⎥⎥ T(-10,-10)
⎢⎣0 0 1 ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1 ⎥⎦ Step:5
S(2,2)
⎡s x 0 x f (1 − s x ) ⎤
Step: 6
= ⎢⎢ 0 sy y f (1 − s y )⎥⎥ T(10,10)
⎢⎣ 0 0 1 ⎥⎦ Net transformation:
T(10,10). S(2,2). T(-10,-10).T(5,8).R(90) . T(-5,-8)
⎡1 0 10⎤ ⎡2 0 0⎤ ⎡1 0 − 10⎤ ⎡1 0 5⎤ ⎡cos 90 − sin 90 0⎤
Date: 2065/6/5 = ⎢⎢0 1 10⎥⎥ ⎢⎢0 2 0⎥⎥ ⎢⎢0 1 − 10⎥⎥ ⎢⎢0 1 8⎥⎥ ⎢ sin 90 cos 90 0⎥
⎢ ⎥
# Rotate the △ABC by 90˚ anticlock wise about ( 5,8) and scale ⎣⎢0 01 ⎥⎦ ⎢⎣0 0 1⎦⎥ ⎣⎢0 0 1 ⎦⎥ ⎢⎣0 0 1⎦⎥ ⎣⎢ 0 0 1⎥⎦
it by (2,2) about (10,10) ⎡ 1 0 − 5⎤
Solution: ⎢0 1 − 8 ⎥
⎢ ⎥
⎣⎢0 0 1 ⎦⎥

Other transformation:
38
1. Reflection
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
2. Shearing. P’ = Rfy.P
Reflection: Providing a mirror image about an axis of an object Rfy = Reflection matrix about y-axis
is called reflection. y

(i) about x-axis (y =0) 2’

2
x’ = x Original position
Reflected object
y’ = -y 1’ 1

In matrix from, 3’ 3
⎡ x ' ⎤ ⎡1 0 0 ⎤ ⎡ x ⎤ x
⎢ y '⎥ = ⎢0 − 1 0⎥ ⎢ y ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎢⎣ 1 ⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 1 ⎥⎦
P’ = Rfx.P
Rfx = Reflection matrix about x-axis
y
1 Date: 2065/6/6
Alternative way for reflection:
Original position
Step:1
2 3 Rotate the object so that y axis concise x-axis. i.e R(-90)
x Step: 2
2’ 3’ Reflection about x-axis. Rfx
Step: 3
Reflected object
Rotate back the object so that y-axis move to original position .
1’
R(-90)
Net transformation:
(ii) about y-axis (x=0) i.e Rfy = R(90) . Rfx. R(-90)
x’ = x ⎡cos 90 − sin 90 0⎤ ⎡1 00⎤ ⎡ cos 90 sin 90 0⎤
y’ = y = ⎢⎢ sin 90 cos 90 0⎥⎥ ⎢⎢0 − 1 0⎥⎥ ⎢− sin 90 cos 90 0⎥
⎢ ⎥
In matrix from, ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦
⎡ x '⎤ ⎡ − 1 0 0⎤ ⎡ x ⎤ ⎡0 − 1 0⎤ ⎡1 0 0⎤ ⎡ 1 1 0⎤
⎢ y '⎥ = ⎢ 0 1 0⎥ ⎢ y ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ = ⎢⎢1 0 0⎥⎥ ⎢⎢0 − 1 0⎥⎥ ⎢⎢− 1 0 0⎥⎥
⎢⎣ 1 ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ 1 ⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦
39
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡0 − 1 0⎤ ⎡0 1 0 ⎤ y’ = x
= ⎢⎢1 0 0⎥⎥ ⎢1 0 0 ⎥
⎢ ⎥
y

⎢⎣0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦

⎡− 1 0 0⎤
= ⎢⎢ 0 1 0⎥⎥ p(x,y)

⎢⎣ 0 0 1⎥⎦
p(x,y) p’(x’,y’)

45
x

# Determine transformation matrix responsible for reflection of

about the line y+x = 0 .
Solution:
(i) R(45)
(ii) Rfx
# Determine transformation matrix responsible for refection of
(iii) R(-45)
object about the line y = x
(1) R(-45)
Net transformation,
(2) Rfx
R(-45) Rfx . R(+45)
(3) R(45 ˚) ⎡ cos 45 sin 45 0⎤ ⎡1 0 0⎤ ⎡cos 45 − sin 45 0⎤
Net transformation , = ⎢⎢− sin 45 cos 45 0⎥⎥ ⎢0 − 1 0⎥ ⎢ sin 45 cos 45 0⎥
⎢ ⎥ ⎢ ⎥
R(45). Rfx R(-45) ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦
⎡cos 45 − sin 45 0⎤ ⎡1 0 0⎤ ⎡ cos 45 − sin 45 0⎤
= ⎢⎢ sin 45 cos 45 0⎥⎥ ⎢0 − 1 0⎥ ⎢− sin 45 cos 45 0⎥
⎢ ⎥ ⎢ ⎥ ⎡ 1 1 ⎤ ⎡ 1 1 ⎤
⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢ 2 0⎥ ⎢ − 0⎥
⎢ 2 ⎥ ⎡1 0 0⎤ ⎢ 2 2 ⎥
⎡0 1 0⎤ 1 1 ⎢0 − 1 0 ⎥ 1 1
= ⎢− 0⎥ ⎢ 0⎥
= ⎢⎢1 0 0⎥⎥ ⎢ 2 2 ⎥ ⎢
⎢⎣0 0 1⎥⎦
⎥ ⎢ 2 2 ⎥
⎢ 0 0 1⎥ ⎢ 0 0 1⎥
⎢⎣0 0 1⎥⎦ ⎢ ⎥ ⎢ ⎥
⎣ ⎦ ⎣ ⎦
.i.e x’ = y
40
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡ 1 1 ⎤ ⎡ 1 1 ⎤ (5) R(0,2)
⎢ 0⎥ ⎢ − 0⎥
⎢ 2 2 ⎥ ⎢ 2 2 ⎥ ⎡ 0 − 1 0⎤
1 1 ⎢− 1 1 ⎢ − 1 0 0⎥ y
= ⎢− 0⎥ − 0⎥ = ⎢ ⎥
⎢ 2 2 ⎥ ⎢ 2 2 ⎥
⎢ 0 0 1⎥ ⎢ 0 0 1⎥ ⎢⎣ 0 0 1⎥⎦ (0,2)
⎢ ⎥ ⎢ ⎥ y+x = 2
⎣ ⎦ ⎣ ⎦
i. e x’ = -y
y’ = -x (2,0)

# Determine transformation matrix responsible for reflection of

x object about the line y = mx+b
(i) T(0,-b)
(ii) R(-θ )
(iii) Rfx
(iv) R(θ )
# Determine transformation responsible for reflection of object (v) T(0,b)
about the line y +x = 2.
(i) T(0,-2) Net transformation = T(0,b). R(θ ) . Rfx.R(-θ ). T(0,-b)
(ii) R(45 ˚ ) ⎡1 0 0⎤ ⎡cos θ − sin θ 0⎤ ⎡1 0 0⎤
(iii) Rfx = ⎢0 1 0⎥⎥ ⎢ sin θ cos θ 0⎥⎥ ⎢0 − 1 0 ⎥
⎢ ⎢ ⎢ ⎥
(iv) R(-45) ⎢⎣0 0 b⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1⎥⎦
(v) R(0,2)
⎡ cos θ sin θ 0⎤ ⎡1 0 0 ⎤
⎢− sin θ cos θ 0⎥⎥ ⎢0 1 − b ⎥
# Determine transformation responsible for reflection of object ⎢ ⎢ ⎥
the line y+x = 2. ⎢⎣ 0 0 1⎥⎦ ⎢⎣0 0 1 ⎥⎦
(1) T(0,-2) ⎡1 0 0⎤ ⎡cos θ − sin θ 0⎤ ⎡cos θ sin θ − b sin θ ⎤
(2) R(45) = ⎢⎢0 1 b⎥⎥ ⎢⎢ sin θ cos θ 0⎥⎥ ⎢ sin θ
⎢ − cos θ b cos θ ⎥⎥
(3) Rfx ⎢⎣0 0 1⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ 0 0 1 ⎥⎦
(4) R(-45)
41
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡1 0 0 ⎤ Reflection:
= ⎢⎢0 1 b⎥⎥
⎢⎣0 0 1⎥⎦ # Determine the X’-formation matrix responsible for reflection
about y = x+3 .
⎡ 1 m ⎤⎡ 1 m bm ⎤
⎢ − 0⎥ ⎢ − ⎥ (1) T(0,-3)
⎢ 1+ m 1 + m2 ⎥ ⎢ 1+ m 1 + m2 1 + m2 ⎥
2 2
(2) R(-45)
⎢ m 1 m 1 1
0⎥ ⎢ − ⎥ (3) Rfx
⎢ 1 + m2 1 + m2 ⎥ ⎢ 1 + m2 1 + m2 1+ m ⎥2 ⎥
⎢ ⎥⎢ (4) R(45˚ )
⎢ 0 0 1⎥ ⎢ 0 0 1 ⎥
⎢⎣ ⎥⎦ ⎢⎣ ⎥⎦ (5) T(0,3)
⎡1 − m 2 2m − 2bm ⎤
⎢ − ⎥
⎡1 0 0⎤ ⎢1 + m 1+ m 1+ m2 ⎥
2 2 y
2m m 2 − 1 − bm 2 + b ⎥
= ⎢⎢0 1 b⎥⎥ ⎢
⎢1 + m 2 1+ m2 1+ m2 ⎥
⎢⎣0 0 1⎥⎦ ⎢ 0 0 1 ⎥
(0,3)

⎢ ⎥
⎢⎣ ⎥⎦ 45
x
⎡1 − m 2 2m − 2bm ⎤ (-b,0)
⎢ ⎥
⎢1 + m 1+ m 1 + m2 ⎥
2 2

2m m −12
2b ⎥
=⎢
⎢1 + m 2
1+ m 2
1 + m2 ⎥
⎢ 0 0 1 ⎥
Shearing: Transformation that causes deformation in the
⎢ ⎥ original object by supposing as if the object is composed of
⎢⎣ ⎥⎦
internal layers and these layers are caused to slide over is
y
shearing.

(0,b)

θ
x
(-b,0)

42
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np

c(x,y) c’(x’,y’) clip

y y’

A B x1
The procedure that identifies these portions of the picture that are
x
x’ either inside or outside of the specified region of span is
About x-axis clipping .
y’ = y
x’= x+x1 where, Line clipping:
x1 α y p2
A
p3

x1 = Shx.y p2’ A’
where, Shx is shearing constant.
y’ = y
x’= x+shx.y p1’
In matrix form, clip
p1
⎡ x'⎤ ⎡1 sh x 0⎤ ⎡ x ⎤
⎢ y '⎥ = ⎢0 1 0⎥⎥ ⎢⎢ y ⎥⎥
B’
⎢ ⎥ ⎢ B
⎢⎣ 1 ⎥⎦ ⎢⎣0 0 1⎥⎦ ⎢⎣ 1 ⎥⎦
Similarly about y-axis Cohen Sutherland line clipping:
x’ = x In this method, every line endpoint is assigned a four digit
y’=y+shy.x binary code(region code) that identifies the location of the point
⎡ x'⎤ ⎡ 1 0 0⎤ ⎡ x ⎤ relative to the boundary.
⎢ y '⎥ = ⎢ sh 1 0⎥⎥ ⎢⎢ y ⎥⎥
⎢ ⎥ ⎢ y For,
⎢⎣ 1 ⎥⎦ ⎢⎣ 0 0 1⎥⎦ ⎢⎣ 1 ⎥⎦ b1 : left
P’ = SHy.p b2 : right
b3 : below
2D window clipping: b4 : above
43
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
1000 The lines which con not be identified as completely inside or
outside a window by these tests are checked for intersection with
1001 0000 1010 the window boundary. Such lines may or may not cross into the
window interior.
0101 0100 In the fig. aside , region code of
0110
P1 = 0001
P2 = 1000
p3
p2 P1 AND P2 = 0
(xwmin, ywmax) (xwmax, ywmax)
So we need future calculation.
p2’ p3’
p4 Starting form p1,
Intersection of P1 with left boundary is calculated.
p1 Region code of P1’ = 0000
(xwmin, ywmin) (xwmax, ywmin) P1’ AND P2 = 0 .
Intersecting of P2 with above boundary is calculated region code
of p2’ = 0000
Since both end points have region codes (0000) .So P1’, P2’
portion of the line is saved.
Similarly,
The value 1 indicates its relative position. If a point is within For P3 , P4.
clipping rectangle then region code is 0000. So , P3 = 1000
If x – xwmin < 0 , b1 = 1 P4 = 0010
If xwmax – x < 0 , b2 = 1 P3 AND P4 = 0
If y – ywmin < 0 , b3 = 1 So we need further calculations; starting form P3 region code of
If ywmin – y < 0 , b4 = 1 P3 is 1000, i.e b4 is high, so intersection of P3 with upper
If the region codes of both end points are 0000 then we accept boundary which yields P3’ having region code 0010.
the line. Again p3’ AND P4 ≠ 0
Any line that have one in the same bit position is rejected i.e So P3P4 is totally clipped.
if RA AND RB ≠ 0 The intersection point with vertical boundary can be obtained by
Line is completely outside . y = y1 + m(x-x1)

44
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Where (x1,y1) and (x2,y2) are end points of line and y is the co- view areas are selected, these areas can be placed in separate
ordinate value of intersection point where x value is either xwmin display locations or some areas could be inserted into other,
or xwmax and m = y2 –y1 / x2 – x1 . larger display area. Transformations from world to device co-
Similarly , intersection point with horizontal boundary ordinates involve translation rotation and scaling operations, as
x = x1 + (y-y1)/m well as procedures for deleting those parts of the picture that are
Where , y = ywmin or ywmax outside the limits of a selected display area.
A world co-ordinate area selected for display is called a
# Use cohen Sutherland algorithm to clip the line for the window. An area on the display device to which a window is
following dimensions. Line endpoints (15, 45) and (25,15) mapped is called a view port. The window defines what is to be
Window: viewed. The viewport defines where it is to be displayed.
Lower left: ( 10, 20) Often, windows and viewports are rectangles in standard
Upper right : ( 30, 40) positions, if the rectangle edges are parallel to the co-ordinate
axis.
“ The mapping of a port of a world co-ordinate scene to a
Date: 2065/6/7 device co-ordinates is referred to as viewing transformation.”
Sometimes , the 2D viewing transformation is simply referred to
Two dimensional object to screen viewing Transformation: as the window – to – viewport transformation or the windowing
transformation.
Mechanism for displaying view of a picture on an output Fig. given below illustrates the mapping of a picture
device: selections that falls within a rectangular window onto a
designated rectangular viewpoint.
A graphical package allows a user to specify which part of the
defined picture is to be displayed and where that part is to be
placed on the display device. Any convenient Cartesian co- y wmax y vmax
ordinate system; referred to as the world co ordinate reference
frame can be used to define the picture.
For a 2-D picture, a view is selected by specifying a subarea of y wmin y vmin
the total picture area. A user can select single area for displaying
or several areas could be selected for simultaneous display. The
picture parts within the selected areas are then mapped onto x wmin x wmax x vmin x vmax
specified areas of the device co-ordinates. When the multiple World Co-ordinate Device co-ordinate
45
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Fig. A viewing transformation using standard rectangles for the yworld

window and viewpoint.

1
Steps to be followed for window to viewpoint transformations Viewport
area:
yo
- Generate world co-ordinate
- Convert world co-ordinate to view co-ordinate.
- Map the view co-ordinate to normalized viewing co-
ordinate. xo xworld

- Map the normalized viewing co-ordinate to device co- World co-ordinate Normalized device co-ordinate
ordinate system.
These steps can be illustrated through a pipeline.
By changing the position of the viewport we can view objects at
different positions on the display area of an output device. Also
Construct world
co-ordinate Convert world
Map viewing
co-ordinates to Map normalized
DC
by varying size of the viewport we can change the size and
VC Normalized viewing NVC viewpoint to
scene using
modeling co-ordinate
WC Co-ordinates
to viewing co-ordinates device
co-ordinate
proportions of displayed objects.
co-ordinate using window
transformation viewpoint
Specification
Window to viewport transformation:
Fig. The two dimensional viewing transformation pipeline

Fig. given below show a rotating view co-ordinate reference y wmax

window y vmax
frame and the mapping to normalized co-ordinates.
y vmin

y wmin y vmin

x wmin x w max x vmin x vmax

World co-ordinate Device co-ordinate

46
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
It is the mechanism for displaying view of a picture on an output The value of sx and sy is found by calculating the position of
device. The world co-ordinate selected for display is called (x,y) in the window to the corresponding position of point (u,v)
window. The area on the display device to which window is in the viewpoint.
mapped is called viewport. So, window defines what is to be To maintain the same relative position, so
viewport defines where it is to be displayed. The mapping of part x − x w min u − x v min
=
of world co-ordinate scence to device co-ordinate is called x w max − x w min x v max − x v min
viewing transformation or window-to-viewport transformation. y − y w min u − y v min
Or, =
y w max − y w min y v max − y v min
x − x w min
S x = v max
x w max − x w min
(xwmax,ywmax) (xvmax, yvmax)
y − y w min
S y = v max
(x,y) (u,v) y w max − y w min
∴ Twv =
(xwmin,ywmin) (xvmin,yvmin) − x w min
⎡ x v max ⎤
⎢x 0 0⎥
⎡1 0 x v min ⎤ − x w min ⎡1 0 − x w min ⎤
⎢ w max ⎥
⎢0 1 y v max − y w min
⎢ y v min ⎥⎥ ⎢ 0 0⎥ ⎢0 1 − y
⎢
⎥
w min ⎥
⎢ y w max − y w min ⎥
Window – to – viewport transformation can be explained as: ⎢⎣0 0 1 ⎥⎦ ⎢ ⎢⎣0 0 1 ⎥⎦
- Choose the world co-ordinate in a rectangle. 0 0 1⎥
⎢ ⎥
- Transform it to origin. ⎣⎢ ⎥⎦
- Scale it with appropriate value. Is the required window- to – viewpoint transformation.
- Transform it to the relative position in viewport.
# Determine window to view port transformation for the
Step 1: T (-xwmin, -ywmin) following dimensions of windows and viewpoint.
Step:2: S (sx, sy) Window viewpoint
Step:3: T(xvmin, yvmin) Lower left corner (5,10) (8,12)
Upper right corner (15,20) (12,18)
∴ Net transformation,
Different types of Graphical package are:
Twv = T (xvmin, yvmin). S(sx, sy).T(-xwmin, -ywmin)
1. GKS (Graphical kernel system)
47
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
2. PHIGS(programmers Hierarchical interactive graphics - Update – Update information associated at some location
standard) in the stack.
3. GL(Graphics Library) The most recently pushed element can be checked prior to
4. Open GL. performing a pop operation.
A stack is based on the last in first out algorithm (LIFO) in
which insertion and deletion operations are performed at one end
Data Structural Concept: of the stack. The most accessible information in a stack is at the
top of the stack and least accessible information is at the bottom
Introduction: of the stack.
Data may be organized in different ways; the logical or If someone wants to delete on item from an empty stack, it
mathematical model of a particular organization of data is called causes underflow and if someone wants to add an element to the
data structure. In other words, data structure is a collection of stack (which is already full) then it is the case of overflow.
data elements whose organization is characterized by accessing
operations that are used to store and retrieve individual data
A Top of the stack
elements.
B
Data structure is of two types:
1. Linear data structure. E.g Array, stack, Queue etc. C

2. Non linear data structure. E.g Trees. D Bottom of the stack

(i) Array: An array is collection of similar elements all Implementation of Stack:

elements of any given array must be of the same type. Stack can be implemented in two ways:
1 2 3 4 5 ..................... 100 1. pointer (Linked list)
Fig. Array of 100 students 2. Array
(ii) Stack: A stack is defined formally as a list (a linear data
structure) in which all insertion and deletions are made at Push operation:
one end called the top of the stack(TOS). The fundamental Step 1: Check for the overflow
operations which are possible on a stack are: If TOS ≥ = size
- push - insertion Output: “stack overflow and exit”.
- pop - deletion Step 2: Increment the pointer value by one
- peep – extract information TOS = TOS +1
Step 3: Perform insertion
48
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
S[TOS] = value
Step 4: Exit Step 1: [Check for underflow]
If TOS-i+1<0
POP operation: Output “stack underflow” and exit.
Step 1: Check for the underflow Step 2: [Change the element]
If TOS = 0 S(TOS-i+1) = value
Output: “Stack underflow and exit” Step 3: Return
Step 2: Decrement the pointer value by one
Value = S[TOS] Polish Notation: The process of writing the operator of an
TOS = TOS-1 expression either before their operands or after them is called
Step 3: Return the former information of the stack the polish notation in honor of its discover, the “Polish
Return [value] mathematician Jan Luksiewicz.”.
Step 4: Exit. The polish notations are classified into three categories:
these are:
PEEP operation: If one interested only about an information - Infix
stored at some location in a stack, then peep operation is - Postfix
required. In this operation we simply move the pointer to the - Prefix
desired location and then fetch the information associated with For example, take an expression a+b, for infix +sign is placed
the location. between two operands ‘a+b’. For postfix, we can write the same
expression as ‘ab+’ i.e operator is placed after the operands.
Step 1: [ Check for the stack underflow] For prefix the same expression can be written as +ab.
If TOS-i+1 < 0
Output : “ Stack underflow” and exit.
Step 2: [Return the ith element from the top of the stack]
Return[S(TOS –i+1)]
Date: 2065/6/8
Update Operation: It is required when the content of some
location in a stack is to be changed. Suppose one wants to update Queue:
information at the ith location in the stack ‘s’. We move TOS
pointer to the ith location from the top of the stack and input the Deletion
9 5 4 2
insertion
new value of that location. Front Rear
49
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
This is a linear data structure used to represent a linear list and Q[rear] = value
permits deletion to be performed at one end of the list and Step 4: [ Set the front pointer ]
insertion on the other end of the list. The information is such a If front = 0
list is processed in the same order as it was received, i.e on first Front = 1
come first serve basis.(FCFS) Step 5: Return
A queue has two pointes; front and rear , pointing to the
front and rear elements of the queue respectively. Consider a Algorithm to Delete an element form queue:
queue consisting of ‘n’ elements and an element value which we
have to insert into the queue. The value NULL (0) of front Step 1: [check the underflow condition ]
pointer employees an empty queue. Figure given above If front = 0
illustrates a queue. Output: “ Underflow” and return.
1 2 3 4 1 2 3 4

11 22 33 44 22 33 44 Step 2: [Remove an element ]

Value = Q[front]
Step 3: [check for empty queue]
Front Rear Front Rear
If front = rear
1 2 3 4 5
Front = 0
22 33 44 55 Rear = 0
Else
Front Rear
Front = front + 1

Step 4: Return (Value )

1 2 3 4 5 6
Algorithm to insert an elements into queue:
11 22 33 44 55 22 33 44 55
Step 1: [check overflow condition]
(b)
If rear ≥ size (a)
o/p : “overflow” and return 33 44 55 33 44 55 66

(c) (d)
Step 2: [ Increment rear poi
nter] Circular queue:
Rear = rear+1 Suppose we have an array Q that contains n element in
Step 3: [Insert an item ]
50 which Q1 comes after Qn in the array. When this technique is
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
used to construct a queue then the queue is called the circular (viii) 77 66 55 (x) 44 33 77 66 55
queue. In other words we can say that a queue is called circular
when the last room comes just before the first room. Fig given Delete 88
Insert 33
F R
below show the circular queue . R F

Q[2]
Linked List :
Q[1]

A linked list is defined as the collection of nodes.

Q[n]
Node 1 Node 2 Node 3 Node 4
Start
1 . 2 . 3 . 4 \o

node
Each node has two port.
(i) (iv) 99 88 77
• information
• Pointer to the next node.
Front Rear F R

(ii) 99 (v) 99 88 77 66 Information part may consist of one or more than one fields. In
other words a linked list consist of a series of a structure contains
Insert 99 Insert 66 one or more then one contiguous information fields and a pointer
FR F R to a structure containing its successor. The last node of the list
contains NULL (‘\0’).
(iii) 99 88 (vi) 88 77 66
The pointer contains the address of the location where
Insert 88 the next information is stored. A linked list also contains a list
F
Delete 99
pointer variable called start, which contains the address of the
F R
first node is the list. Hence , there is an arrow drawn from start
(ix) 44
(vii) 88 77 66 55 77 66 55
to the node in the linked list. If there is no node in the list, the is
Insert 44
called NULL list or EMPTY list.
Insert 55
F R F The different operations performed on the linked list are:
(1) Insertion:
a. Inset a node at the beginning of the list.
51 b. Inset a node at the end of the list.
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
c. Inset a node between the nodes. in the figure above. Now it is possible to define a doubly linked
(2) Deletion as a collection of nodes, each nodes having three fields.
a. Delete a node at the beginning of the list. - pointer to previous node ( pointer to predecessor)
b. Delete a node at the end of the list. - Information field
c. Delete a node between the nodes. - Pointer to next node (pointer to successor)
(3) Traversing . : Traveling from one node to another node. Different operations performed on doubly linked list are:
1. Insertion 2. Deletion 3. Traversing.
Doubly linked list:
A singly linked list is so named because each list element Non linear data structure:
contain a pointer to the next element . In that list traveling is (1) Trees
possible only in one direction. Some times it is required to
transverse the list in either direction forward or backward. This N1
T1
involves the performance and efficiency of algorithms. So T3
T2
traversing of linked list in both the directions requires nodes N2 N13
N4
having the links as given in figure below: N3
N12
information
N5 N6 N8 N11
right link N9
N10
right link N18
N14 N7
left link
left link
N15
N16
N17
node
N
Fig. Tree T having 18 nodes
N =NULL = ‘\0’
A tree T is defined as a set of finite set of one or more nodes
such that.
Doubly linked list
- There is a special node called the root R.
- The remaining nodes are divided into n ≥ 0
The links are used to denote the predecessor and successor of a Disjoint sets T1,T2,T3,………Tn , where each of these set is a
node. The link denoting the predecessor of a node is called the tree . T1, T2, T3,……Tn is called the subtree of the root.
left link and that denoting its successor is right link. A list with The root node N1 has three subtrees T1,T2,T3 and the roots of
this type of arrangement is called Doubly Linked List as shown 52 these subtrees are N2,N3,N4 respectively . The roots of these
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
subtrees of a node is called its degree, root N1, has degree 3. left successor of R. If T2 is not empty then the root of T2 is
Similarly node N5 has degree 5 and node N14 has degree 0. A known as right successor of R.
node with degree 0 is called leaf.
N6,N7,N8,N9,N10,N11,N12,……N18 are called leaf nodes. The R R
leaf nodes are also called Terminal nodes and rest of nodes in a ROOT 1
Binary tree T1 T2
tree is called non-terminal.
A B 2 3
Forest:
Left child Right child 4 5 7
6
N1
N2

Expression :
N5 N3
E = v + u/x* y-z
N6 N7 _
Z
N18 N8
N9
N14
N15 +
N16 V
N17

*
A forest is a set of n ≥ 0 disjoint trees, where n represents no of
y
nodes in the tree. The meaning of forest is very close to a tree .
because if we remove the root of a tree, we get a forest. For e.g
in the fig. above we remove N1, we get the forest of three trees. u
x
It is shown in the figure above.

Binary tree: Complete Binary Tree: A binary tree T is called complete if

A binary tree T is a finite set of nodes, which is either empty each node of T can have at the most two children. A binary tree
or consists of special nodes called R and two disjoint binary tree at level L can have at the most 2L nodes.
T1 and T2 (Which are called the left subtree and right subtree If L=0 20 = 1
respectively). If T1 is not empty then the root of T1 is called the If L=1 21 = 2
If L=2 22 = 4
53 If L=3 23 = 8
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Implies that there is only one node in the tree that is root. v1 e1
v2
e1 e4 e3
v1 v2
e5 e2
Level 0 1 v4
e4 e6 e7
e2
e3
Level 1 v4 v3 v5 v3
2 3 e5
e6 e8 e9
e7 v6
Level 2 4 5 7
6 v5 Fig.(e) Fig.(d)

8 9 10 11 12 13 14 15 v2
Level 3 e1
e4
Level 4 e2
e3
Level 5 v3
v1
Fig.(f)

Traversing Binary Tree:

1. Pre order Traversal RAB
From the figure it is clear that fig. (a) consists of two vertices
2. In order Traversal ARB
each one is stand alone. Fig (b) contains an undirected graph
3. Post order Traversal ABR
having one edge and two vertices. Fig. (c) consists of two
vertices and one edge and it is called directed graph. Fig. (d)
2. Graphs: graphs G is defined as set of two tuples G = (V,E).
illustrates an undirected graph with six vertices and nine edges.
Where ,
Fig (e) is a directed graph with five vertices and seven edges.
V represents set of vertices of G and E represents the set of
Fig (f) consists of three vertices and four edged in which two
edges of G. There exists a mapping from the set of edges to a set
edges are directed and two are undirected known as mixed graph.
of pairs of element of V. Fog eg. given below shows the different
types of graphs.
Path: A sequence of edges of a digraph (directed graph) such
v1 v2
that the terminal vertex of an edge sequence in the initial vertex
Fig.(a) of the next if exits is called the path. For example:
v1 v1 v2
v2
Fig.(b) Fig.(c)

54
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
v1 e1 y
v2
e5 e8
e6 e2
v4
e4 e7
xy-plane
v5 v3
e3 x
E = {(v1,v2), (v2,v3), (v3,v4),(v4,v5),(v5,v1)} xz-plane
If we consider the element of E, we find that E is a set of five
tuples, i.e for edge. z
e1 = (v1,v2)
e2 = (v2,v3) The 3D viewing process is inherently more complex then is the
e3 = (v3,v4) 2D viewing process. In 2D we simply specify specify the
e4 = (v4,v5) window on 2D world and a viewpoint on the 2D-view surface.
e5 = (v5,v1) Conceptually objects in the world are clipped against the window
From e1 and e2 we find that v2 is common in both. In the e1 , v2 and are then transformed into the viewpoint for display. The
is the terminal vertex and it is initial vertex of e2. Similarly in e2 extra complexicity 3D-viewing is caused in the part by the added
and e3 , v3 is corner and so on. Thus we can say that there is dimensions and in part by the fact that display device are only
cycle and there exist a circular path that is shown by the dark line 2D
in the above fig. The solution is mismatch between 3D object and 2D displays
is accomplished by introducing projections which transforms 3D
Date: 2065/7/24 objects onto a 2D-projection plane.
Three dimensional Graphics: In 3D-viewing , we specify a new volume in world a
6.1 Three Dimensional object to screen perspective viewing projection onto a projection plane and a viewport on the view
Transfomation: surface. Conceptually objects in the 3D-new volume and are then
projected. The content of the projection of the new volume onto
the projection plane, called the window, are then displayed
(transformed into the view port for display). Figure below shows
the conceptual model of the 3D-viewing process.
o/p Clipped world Projector Transformation into 2D-device
Co-ordination view port in 2D Co-ordinate
primitive co-ordinate Projection to
Clip against Projection plane device co-ordinate
new volume For device

55
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
projection we extend parallel lines from each vertex plane of the
Fig. Conceptual model of 3D-viewing process. screen.
Equation of line in 2D is
Projections: Projections transform points in a co-ordinate (y-y1)/(x-x1) = (y2-y1)/(x2-x1)
system of dimension ‘n’ into points in a co-ordinate system of In parametric form,
dimension less than ‘n’ projection of 3D-object is defied by x = x1+(x2-x1)u
straight projection rays (Called projectors) emanding from a y = y1+(y2-y1)u u ε [0,1]
centre of projection, passing through each point of the object and
intersecting a projection plane to form the projected object. If the Similarly in 3D,
projection is onto a plane (rather than some curved surface) and x = x1+(x2-x1)u
uses straight objectors (rather then curved) then we call it planar- y = y1+(y2-y1)u
geometric projection. y = z1+(z2-z1)u
Let direction of projection is given by vector [xp,yp,zp] and the
Types of Geometric Projection: image is to be projected on the xy-plane. If we have a point on
1. Parallel projection the object at (x1,y1,z1) and the projected point be (x2,y2) then the
2. Perpesctive projection. equation of the line is ,
x = x1+xpu
Parallel projection: y = y1+ypu
z = z1+zpu
projectors
A’ A But on the xy plane
projectors AA’ and
0 = z1+zpu
∴ u = -z1/zp
Centre of BB’ are parallel
projection at
infinity
B’ B
x2 = x1-z1(xp/zp)
projection plane
(plane of projection POP) y2 = y1-z1(yp/zp)
Fig. line AB and its parallel projection A’B’

If the distance between COP and POP is infinite then projection

is called parallel projection, i.e the rays from COP are paralleled.
So it maintains relative proportion of the object. In this
56
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡ x 2 ⎤ ⎡1 0 − xp / zp 0⎤ ⎡ x1 ⎤ [x2,y2,z2] = [ (xcz1-x1zc)/(z1-zc), (ycz1 – y1zc)/(z1 – zc), 0]
⎢ y ⎥ ⎢0 1 − yp / zp 0⎥⎥ ⎢⎢ y1 ⎥⎥ In homogenous form (Homogenous co-ordinate system)
i.e ⎢ 2 ⎥ = ⎢ [x2,y2,z2,1] = [ xcz1-x1zc, ycz1-y1zc, 0, z1-zc]
⎢ z 2 ⎥ ⎢0 0 1 0⎥ ⎢ z1 ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣ 1 ⎦ ⎣0 0 0 1⎦ ⎣ 1 ⎦
projectors
A’
However, while drawing the projected image we just ignore the z
plane. A’B’ = prospective
Centre of
projection projection
(COP)
(2) Perspective projection: If the distance between POP and B’
COP is finite and converges to a single point (Vanishing
point) then the projection is called perspective projection i.e projection plane
(plane of projection POP)
it transforms points of 3D object along projection lines that A’B’ = perspective projection of AB
meet at vanishing point. The size of the object doesnot
remains same and varies with the distance of the object from
COP. Objects further from viewing position is displayed
smaller than objects of same size that are nearer to viewing
positions. Perpesctive projection is more realistic than
parallel.
If the COP at [ xc,yc,zc] and the point on the object is [x,y,z] then
projection ray will be the line joining these points. Equation of
the line is
X = xc+(x1-xc)u
Y = yc+(y1-yc)u
Z = zc+(z1-yc)u Fig. Same vehicle
The projected point on xy plane i.e (x2,y2) will be the point
where the line intersects the xy plane. In Matrix form ,
i.e zc = 0, zc+(z1-zc)u = 0 ⎡ x 2 ⎤ ⎡− z c 0 xc 0 ⎤ ⎡ x1 ⎤
i.e u = -zc/(z1-zc) ⎢y ⎥ ⎢ 0 − zc yc 0 ⎥⎥ ⎢ y1 ⎥
⎢ 2⎥ = ⎢ ⎢ ⎥
⎢ z2 ⎥ ⎢ 0 0 0 0 ⎥ ⎢ z1 ⎥
x2 = (xcz1-x1zc)/(z1-zc) ⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣1⎦ ⎣ 0 0 1 − zc ⎦⎣ 1 ⎦
y2 = (ycz1 – y1zc)/(z1-zc) 57
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡− z c 0 xc 0 ⎤
⎢ 0 − zc yc 0 ⎥⎥ - In matrix notation using normalized homogeneous co-
∴ Tpersp = ⎢ ordinate this is performed by the matrix multiplication.
⎢ 0 0 0 0 ⎥
⎢ ⎥
⎣ 0 0 1 − zc ⎦
⎡ x ' ⎤ ⎡1 0 0 t x ⎤⎡ x⎤
However z value is needed for hidden surface detection, so ⎢ y '⎥ ⎢0
⎢ ⎥=⎢ 1 0 t y ⎥⎥ ⎢⎢ y ⎥⎥
transformation matrix changes to
⎢ z ' ⎥ ⎢0 0 1 t z ⎥⎢ z ⎥
⎡− z c 0 xc 0 ⎤ ⎢ ⎥ ⎢ ⎥⎢ ⎥
⎢ 0 − zc yc 0 ⎥⎥ ⎣ 1 ⎦ ⎣0 0 0 1 ⎦ ⎣1 ⎦
⎢
⎢ 0 0 1 0 ⎥
⎢ ⎥ Inverse translation,
⎣ 0 0 1 − zc ⎦
T-1 (tx,ty,tz) = T(-tx,-ty –tz)
However while drawing we ignore the z value. y

Date: 2065/7/25 (x’,y’,z’)

Extension of two-dimensional to three dimensional (x,y,z)

transformation:
Just as 2D-transfromtion can be represented by 3x3 matrices Fig. Translating a point
using homogeneous co-ordinate can be represented by 4x4 with translation vectro
T = (tx, ty, tz)
matrices, provided we use homogenous co-ordinate z
representation of points in 3D space as well
Scaling:
Translation: (i) About origin:
- Translation in 3D is similar to translation in the 2D except
that there is one more direction parallel to the z-axis. Explicit expression:
- If , tx,ty,and tz are used to represent the separate shifts , X’ = x*sx
then the translation of the position (x,y,z) into the point Y’ = y*sy
(x’,y’,z’) is done by Z’= z*sz
x’ = x +tx - The matrix representation for the scaling x’-formation of
y’ = y +ty a position p(x,y,z) relative to the co-ordinate origin.
z’ = z+tz Matrix form:
58
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
⎡ x'⎤ ⎡ s x 0 0 0⎤ ⎡ x ⎤ y
y
⎢ y '⎥ ⎢ 0 sy 0 0⎥⎥ ⎢⎢ y ⎥⎥
⎢ ⎥=⎢
⎢ z'⎥ ⎢ 0 0 sz 0⎥ ⎢ z ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣1⎦ ⎣ 0 0 0 1⎦ ⎣ 1 ⎦
Where scaling parameter sx,sy,sz are assigned positive values.
(xf,yf,zf)) (xf,yf,zf))

(iii) About any fixed point: (xf, yf, zf) x

- Scaling w.r.t to a selected fixed position (xf,yf,zf) can be x
z (b)
represented with the following transformation. z (a)
Step:1 Translate the fixed point with the origin T(-xf,-yf,-zf)
Step:2 Scale the object relative to the co-ordinate origin.
S(sx,sy,sz)
Step:3 Translate the fixed point back to its origin position.
T(xf,yf,zf)
y
y
Net transformation
T = T(xf,yf,zf). S(sx,sy,sz). T(-xf,-yf,-zf)
⎡1 0 0 xf ⎤ ⎡s x 0 0 0 ⎤ ⎡1 0 0 − xf ⎤
⎢0 1 0 yf ⎥⎢ 0 s 0 0⎥⎥ ⎢⎢0 1 0 − y f ⎥⎥
=⎢ ⎥⎢ y
. (xf,yf,zf))
⎢0 0 1 zf ⎥⎢ 0 0 sz 0⎥ ⎢0 0 1 − zf ⎥
⎢ ⎥⎢ ⎥⎢ ⎥ (xf,yf,zf))
⎣0 0 0 1 ⎦⎣ 0 0 0 1 ⎦ ⎣0 0 0 1 ⎦ x
⎡s x 0 0 (1 − s x ) x f ⎤
x
(d)
⎢0 sy 0 (1 − s y ) y f ⎥ (c)
=⎢ ⎥
Rotation:
⎢0 0 sz (1 − s z ) z f ⎥
⎢ ⎥
⎣0 0 0 1 ⎦

59
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
y y

x Fig. Rotation of an object

about z-axis
z z

Axis of rotation Direction of +ve rotation Rotation about x-axis:

X y to z x’ = x
Y z to x y’ = ycosθ+zsinθ
Z x to y z’ = ysinθ+zsinθ
⎡1 0 0 0⎤
Rotation About z-axis: ⎢1 cos θ − sin θ 0⎥⎥
Z-component does not change. The rotation of a point about z- ∴ Rx = ⎢
⎢0 sin θ cos θ 0⎥
axis is equivalent to rotation of point projected on xy plane ⎢ ⎥
⎣0 0 0 1⎦
(keeping z-value constant ) which is equivalent to 2D rotation i.e y
rotation angle is equal to the angle between initial and final
projected points on xy-plane
X’ = xcosθ – ysinθ
Y’ = xsinθ + ysinθ
Z’ = z

⎡cos θ − sin θ 0 0⎤ Fig. Rotation of an object

⎢ sin θ cos θ 0 0⎥⎥
about x-axis

Rz = ⎢ z
⎢ 0 0 1 0⎥ About y-axis:
⎢ ⎥
⎣ 0 0 0 1⎦ Y’ = y
z = zcosθ – xsinθ
x = zsinθ+ x cosθ
60
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
y
y

Date: 2065/7/27
Rotation about any co-axis:

(a) Parallel to any of the co-axis: When an object is to be x

rotated about an axis that is parallel to one of the co-ordinate x
z (d) Translate rotation axis to original axis
axis, we need to perform some series of transformation. z
(c) Rotate object through angle θ

1. Translate the object so that the rotation axis coincides with

the parallel co-ordinate axis. T(-a,-b,-c) where, (a,b,c) is any (b) Not parallel to any of the co-axis:
point on the rotation axis. When an object is to be rotated about an axis that is not parallel
2. Performed the specified rotation about the axis. to one of the co-ordinate axes, we need to perform some series of
Rx(θ ) transformation.
3. Translate the object so that the rotation axis is moved to its (i) Translate the object such that rotation axis passes through
original position. T(a,b,c) co-ordinate origin.
Net transformation, (ii) Rotate the axis such that axis of rotation coincides with
= T(a,b,c) Rx(θ). T(-a,-b,-c) one of the co-ordinate axis.
y (iii) Perform the specific rotation about the co-ordinate axis.
y (iv) Apply inverse rotation to bring the rotation axis back to its
original orientation.
(v) Apply inverse translation to bring the rotation axis back to
its original position.

Rotation axis
x
x
z (b) Translate rotation axis into x-axis
z (a) original position of object

61
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
r r
△ = p 2 − p1 = (x2,y2,z2) – (x1,y1,z1)
y
y

P2 = (x2-x1, y2-y1, z2-z1)

P’2

Unit vector along rotation axis is

r
P1 v ( x 2 − x1 , y 2 − y1 , z 2 − z1 )
P’1 u= r =
v ( x 2 − x1 ) 2 + ( y 2 − y1 ) 2 + ( z 2 − z1 ) 2
x
x
= (a,b,c) [let]
z
z
(b) Translate p1 to the origin Where (a,b,c) are direction cosines of rotation axis i.e a2+b2+c2 =
(a) Initial position
y
1.
y
⎡1 0 0 − x1 ⎤
⎢0 1 0 − y1 ⎥⎥
(i) T(-x1,-y1,-z1) = ⎢
⎢0 0 1 − z1 ⎥
⎢ ⎥
⎣0 0 0 1 ⎦

P’1
θ P’1 (ii) For step II rotation can not be performed directly since the
rotation is not about any of the co-ordinate axis hence. It is
x
x P’’2 done in two steps.
z
z
(c) Rotate p2’ onto the z-axis
(d) Rotate the object around z-axis (a) Rotation by angle α about x axis.
y y (b) Rotate by angle β about y axis.
y y
P2 u’
(a,b,c)
P’2
(a,b,c)

P1 Rx(α)

P’1 α α x
Ryβ)
x x β
uz(0,0,1)
z u u’’
(e) Rotate the axis to original z (f) Rotate the rotation axis to its z z
orientation original position.

Step (a)
Let p1, p2 be the two end points of axis of rotation then the axis Angle for rotation of u around x-axis into xz plane is equal to
vector is given by:
62 angle between u’ (projection of u on yz plane and z-axis ).
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
r r
u '.u z
Cosα = r r =
(0, b, c )(. 0,0,1) = c Step: 3
u'.uz b2 + c2 d ⎡cos θ − sin θ 0 0⎤
⎢ sin θ cos θ 0 0⎥⎥
r r Rz(θ) = ⎢
u '×u z (0, b, c )(
. 0,0,1) b ⎢ 0 0 1 0⎥
Sinα = r r = = Where, d = b 2 + c 2 ⎢ ⎥
u'.uz b + c2
2 d ⎣ 0 0 0 1⎦
Step: 4
⎡1 0 0 0⎤ ⎡1 0 0 0⎤ (a) R-1y(β)
⎢0 cos α − sin α 0⎥⎥ ⎢0 c / d −b/d 0⎥⎥ (b) R-1x(α )
∴ Rx (α) = ⎢ = ⎢
⎢0 sin α cos α 0⎥ ⎢0 b / d c/d 0⎥
⎢ ⎥ ⎢ ⎥ Step: 5
⎣0 0 0 1⎦ ⎣0 0 0 1⎦
Steps (b) T(x1, y1,z1)
U moves to the position u’’ after being rotated by α about x-axis ∴ Net transformation,
r
so the position of u ' ' is given by = T(x1,y1,z1). R-1x(α ) R-1y(β) R2(θ). Ry(β).Rx(α )T(-x,-y,-z)
u’’ = Rx(α ).u
⎡1 0 0 0⎤ ⎡ a ⎤ ⎡ a ⎤ Date:2065/7/28
⎢0 c / d − b / d 0⎥⎥ ⎢⎢b ⎥⎥ ⎢⎢ 0 ⎥⎥
=⎢ = Reflection:
⎢0 b / d c / d 0⎥ ⎢ c ⎥ ⎢ d ⎥
⎢ ⎥⎢ ⎥ ⎢ ⎥
⎣0 0 0 1⎦ ⎣1 ⎦ ⎣ 1 ⎦ (i) About Axis: Reflection about axis is equivalent to 180˚
r r rotation about the axis.
β = angle between u ' ' and u z
r r
u ' '×u z (a,0, d )(. 0,0,1) About z-axis:
cosβ = r r = =d ⎡cos 180 − sin 180 0 0⎤
u' '.uz a2 + d 2 ⎢ sin 180
r r cos 180 0 0⎥⎥
u ' '×u z (a,0, d ) × (0,0,1) Rfz – Rz(180˚) = ⎢
Sinα = r r = = −a ⎢ 0 0 1 0⎥
u' '.uz a2 + d 2 ⎢ ⎥
⎣ 0 0 0 1⎦

⎡ cos β 0 sin β 0⎤ ⎡ d 0 −a 0⎤ ⎡− 1 0 0 0⎤
⎢ 0 ⎢ 0 −1 0 0⎥⎥
1 0 0⎥⎥ ⎢⎢ 0 1 0 0⎥⎥ =⎢
Ry(β ) = ⎢ = ⎢0 0 1 0⎥
⎢− sin β 0 cos β 0⎥ ⎢ a 0 d 0⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣ 0 0 0 1⎦ ⎣ 0 0 0 1⎦ ⎣0 0 0 1⎦
About x-axis:
63
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
Rfx = Rx(180) ⎡1 0 0 0⎤
⎡1 0 0 0⎤ ⎢0 − 1 0 0⎥⎥
⎢0 cos 180 − sin 180 Rfxz = ⎢
0⎥⎥ ⎢0 0 1 0⎥
=⎢ ⎢ ⎥
⎢0 sin 180 cos 180 0⎥ ⎣0 0 0 1⎦
⎢ ⎥
⎣0 0 0 1⎦ (c ) about yz-plane (x-axis):
About y-axis: ⎡− 1 0 0 0⎤
Rfy = Ry(180) ⎢0 1 0 0⎥⎥
⎡ cos 180 0 sin 180 0⎤ Rfyz = ⎢
⎢0 0 1 0⎥
⎢ 0 1 0 0⎥⎥ ⎢ ⎥
=⎢ ⎣0 0 0 1⎦
⎢− sin 180 0 cos 180 0⎥ About any plane:
⎢ ⎥
⎣ 0 0 0 1⎦ Step:1: Translate the plane such that it passes through origin i.e
normal vector passes through origin.
(ii) About any axis:
Step:2: Rotate the plane such that normal vector lies on one of
Net transformation: the co-ordinate axis.

T (x1,y1,z1) Rx (-α) Ry(-β) . Rfz Ry(β). Rx (α) T(-x1,-y1,-z1) Step:3: Perform reflection about the plane whose normal vector
(iii) About plane: is one of the co-ordinate axis.
X’ = x ,
Y’ = y Step:4: Rotate back the plane such that normal vector takes its
Z’ = z original orientation.
(a) about xy plane (z-axis)
⎡1 0 0 0⎤ Step5: Translate back the plane to its original position.
⎢0 1 0 0⎥⎥
Rfxy = ⎢
⎢0 0 −1 0⎥ Shearing:
⎢ ⎥
⎣0 0 0 1⎦ In 2D, shearing about x-axis means x-values changes by
(c) about xz plane (y-axis): amount proportional to y-values and y-values remains same.
However in 3D, z-axis means x and y value change by
amount proportional to z-value and z-value remains same.
i.e x’= x+ Shx.z
64
Downloaded from www.jayaram.com.np
 Downloaded from www.jayaram.com.np
y’ = y + Shy.z
z’ = z
⎡ x ' ⎤ ⎡1 0 S hx 0⎤ ⎡ x ⎤
⎢ y '⎥ ⎢0 1 S hy 0⎥⎥ ⎢⎢ y ⎥⎥
⎢ ⎥=⎢
⎢ z ' ⎥ ⎢0 0 1 0⎥ ⎢ z ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣ 1 ⎦ ⎣0 0 0 1⎦ ⎣ 1 ⎦
Similarly,
⎡ 1 0 0 0⎤
⎢S 1 0 0⎥⎥
SHx = ⎢ hx
⎢ S hx 0 1 0⎥
⎢ ⎥
⎣ 0 0 0 1⎦
⎡1 S hx 0 0⎤
⎢0 1 0 0⎥⎥
SHy = ⎢
⎢0 S hx 1 0⎥
⎢ ⎥
⎣0 0 0 0⎦

#1. Find the transformation matrix for the reflection of 3D object

about plane 2x+y-4z+5 = 0 .
#2. Find the transformation matrix for the reflection of 3D
abject about plane passing through origin whose normal vector is
(i+j +k) (all are vector).
#3. Reflection of 3D object about axis joining the points (1,2,3)
and (4,5,6).
# 4. Rotation of 3D object by 60˚ about axis joining the points
(1,2,3) and (6,5,7).

65
Downloaded from www.jayaram.com.np