Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
4 views

Week-01-lecture-3b Digital Image Fundamentals

The document covers the fundamentals of digital image processing, including the nature of light, color perception, and image acquisition methods using various sensors. It discusses image formation models, sampling, quantization, and the effects of resolution on image quality. Additionally, it addresses issues like aliasing and the Shannon Sampling Theorem in relation to image processing.

Uploaded by

aneestrader3
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
4 views

Week-01-lecture-3b Digital Image Fundamentals

The document covers the fundamentals of digital image processing, including the nature of light, color perception, and image acquisition methods using various sensors. It discusses image formation models, sampling, quantization, and the effects of resolution on image quality. Additionally, it addresses issues like aliasing and the Shannon Sampling Theorem in relation to image processing.

Uploaded by

aneestrader3
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 23

Digital Image Processing

Digital Image Fundamentals

Lecture - 1
Chapter 2: Digital Image Fundamentals
Propagating
EM waves visualization Sinusoidal waves
With wavelength λ
λ = c/v

E=hυ

Plank’s constant
The visible band of EM spectrum is ranged from 0.43 μm to
0.79 μm (violet, blue, green, yellow, orange, red)
Color perception
• Due to nature of light reflected from the object
• White object: Reflects light balanced with all
visible wavelengths
• Green object: Reflects light with wavelength in the
range from 500 nm to 579 nm
• Chromatic light: spans the EM spectrum from 0.43
μm to 0.79 μm
• Quality of chromatic light is described by
Radiance (watt), Luminance (lumens) and
Brightness
Image sensing and acquisition
Image: Combination of illumination and reflection

Three principal arrangements


used to transform illumination
energy into digital images
Image acquisition using single sensor

Most familiar sensor Photodiode:


Whose output voltage waveform is proportional to light
Image acquisition using sensor strips (4000 sensors)

Used in most flat bed scanners


4000 or more in-line sensors are possible
Image acquisition using sensor arrays

CCD (charge couple device) sensors are used in digital camera


of elements 4000×4000
A Simple Image Formation Model
• We shall denote images by two-dimensional
functions of the form
f (x, y) i(x, y) r(x, y)
0  f ( x, y )   where 0  i(x, y)   (illumination)
and 0  r(x, y)  1 (reflectance)

• illumination i(x,y) : determined by illumination


source
• Reflectance r(x,y) : determined by the
characteristics
i = 90000 of the
lm/sq. m for clear imaged
day, objects.
i = 10000 lm/sq. m for cloudy day
i = 0 .1 lm/sq. m for full moon night,
r = 0.01 for black velvet, 0.65 for stainless steel, 0.80 for flat white, 0.9
for silver plated metal, 0.93 for snow
Image sampling and quantization
Chapter 2: Digital Image Fundamentals
Representation of digital Images
Number of storage bits
Let number of gray levels L = 2k
IF M (rows) ≠ N (columns)
Bit required to store the digitized image
b=M×N×k
IF M (rows) = N (columns)
Bit required to store the digitized image
b = N2 × k = M 2 × k
Spatial and Gray Level Resolution
• Sampling determine the spatial resolution of
an image
• Resolution is the smallest discernible detail in
the image (smallest number of discernible line
pairs per unit distance
• Gray level resolution is the smallest discernible
change in gray levels
• Considerable discretion regarding number of
samples but this is not true for the number of
gray levels (usually an integer power of 2)
Effect on image quality by varying N

Allowable gray level


Resampling of images into 1024×1024 pixels
by row and column duplication
Effect on image quality by varying k (number of gray levels)
continued
Effect on image quality by varying N and k simultaneously

Little detail Large detail


Intermediate detail
Isopreference curve
Aliasing and Moire Pattern
• Functions whose area under the curve is
finite can be represented in terms of sines
and cosines of various frequencies
• The sine/cosine components with the
highest frequency determines the highest
frequency contents of the function
• Band-limited function function with finite
highest frequency and unlimited duration
• Shannon Sampling Theorem
• If the function is under-sampled, then a
phenomenon called aliasing corrupts the
sampled image
Cont.
• To reduce the aliasing effect on image
is to reduce the high frequency
component by blurring the image
• Aliasing always present in sampled
image
Chapter 2: Digital Image Fundamentals

You might also like