CAT1 Computer Vision Syllabus
CAT1 Computer Vision Syllabus
CAT1 Computer Vision Syllabus
1. Introduction
2. Understanding the fundamentals of Computer Vision and Images Apply and analyze the
methods for pre-processing
3. What is computer vision? -A brief history
4. Image formation -Geometric primitives and transformations
5. Photometric image formation
6. Understanding the Image age Acquisition and the role of camera
7. The digital camera
8. Image processing
9. Point operators
10. Understanding and applying the basics of image transformation
11. Linear filtering
12. More neighbourhood operators
CAT 1 Question
1 Discuss the components of image processing and analyse each component in detail. (7)
Ans :- Image processing is a field of study that involves the
manipulation of digital images using computer algorithms. It
involves a series of operations performed on an image to enhance
its quality, extract useful information, or transform it into a more
meaningful representation. The process of image processing is
composed of several stages, each with its own set of techniques
and algorithms. In this answer, we will discuss the various
components of image processing and analyze each of them in
detail.
1. Image Acquisition
2. Image Enhancement
3. Image Analysis and Recognition
1. Image Acquisition:
2. Image Enhancement:
2 Discuss the Colour models like RGB, CMY, and HSV etc?(5)
Ans:- Color models are mathematical representations that define
how colors can be represented numerically.
The RGB color model stands for Red, Green, Blue. It is an additive
color model used to represent colors in electronic systems, such as
computer monitors, televisions, and digital cameras. In this model,
colors are created by combining different intensities of red, green,
and blue light.
Ans:- A monochrome image and a grayscale image are two different types of images
commonly used in digital image processing. Although they are often used
interchangeably, they have some fundamental differences.
A monochrome image is an image that consists of only one color, typically black or
white. It is a binary image where each pixel is represented by a single bit of
information, 0 or 1, indicating whether the pixel is black or white. Monochrome images
are commonly used in text-based documents, such as faxes, where only black or white
information is needed.
On the other hand, a grayscale image is an image that consists of different shades of
gray. Each pixel in a grayscale image is represented by a single value, typically ranging
from 0 to 255, indicating the intensity of gray. Grayscale images are commonly used in
image processing applications where color information is not necessary, but the
intensity or brightness of the image is important.
The main difference between monochrome and grayscale images is the number of
possible values that each pixel can take. In a monochrome image, each pixel can only
take one of two possible values, whereas in a grayscale image, each pixel can take any
value between 0 and 255. This difference in the number of values has implications for
image processing applications. For example, grayscale images can be used for more
complex image processing operations, such as edge detection, while monochrome
images are more suitable for simple operations like thresholding.
In summary, while both monochrome and grayscale images are commonly used in
image processing applications, they differ in the number of values that each pixel can
take. Monochrome images have only two possible values, black or white, while
grayscale images can take any value between 0 and 255, indicating different shades of
gray.
4 Explain the simple image model. How the basic nature of image is characterized by the
two components, called illumination and reflectance components.(7)
Ans:-
The simple image model is a mathematical model used in digital image
processing to describe the basic nature of an image. According to this
model, any image can be decomposed into two components: the
illumination component and the reflectance component.
6 Analyse the Role of Digital Cameras and How can we control the depth of field?(7)
Ans:- Digital cameras have become ubiquitous in modern society and have
revolutionized the way we capture and share images. They use digital technology to
convert light into electrical signals that can be stored and processed by a computer,
resulting in high-quality images that can be easily edited and shared.
One important aspect of digital photography is the control of depth of field, which
refers to the range of distances in an image that are in focus. Controlling depth of field
can help to create a more aesthetically pleasing image by drawing attention to specific
objects or areas of the scene.
There are several factors that can affect the depth of field in a digital image, including
aperture size, focal length, and distance to the subject. To control depth of field, we can
adjust the aperture size, which refers to the size of the opening in the lens that lets
light into the camera. A larger aperture (smaller f-number) will result in a shallower
depth of field, while a smaller aperture (larger f-number) will result in a deeper depth
of field.
In addition to aperture size, we can also control depth of field by adjusting the focal
length of the lens and the distance to the subject. A longer focal length and a closer
distance to the subject will result in a shallower depth of field, while a shorter focal
length and a farther distance to the subject will result in a deeper depth of field.
Overall, digital cameras play a critical role in capturing high-quality images and allow
us to control various aspects of the image, including depth of field. By adjusting the
aperture size, focal length, and distance to the subject, we can create images that are
aesthetically pleasing and convey our artistic vision.
O=L-1-I
O = c * log(1 + I)
where c is a scaling constant and I and O are the input and output
intensity values, respectively. Log transformation is a spatial domain
method that is used to enhance the contrast of an image,
particularly in the darker regions. The log function compresses the
dynamic range of the image, making the darker regions brighter
and easier to see.
8 State different limitations of a pinhole camera and how to overcome these limitations.
Write a short note on thin lenses. (7)
Ans:- Pinhole cameras are simple optical devices that use a tiny
aperture to project an inverted image of the outside world onto a
screen or film. However, pinhole cameras have several limitations
that can affect the quality of the image:
Thin lenses are one of the most common types of lenses used in
cameras and other optical devices. A thin lens is a simple lens that
is characterized by its focal length and refractive index. Thin lenses
can be used to form images of objects by bending the light that
passes through them.
There are two types of thin lenses: converging lenses and diverging
lenses. A converging lens is a lens that converges light rays towards
a single point, called the focal point. A diverging lens, on the other
hand, diverges light rays away from a single point, called the virtual
focal point.
Thin lenses can be used in different ways to form images,
depending on the position of the object and the lens. For example,
a converging lens can be used to form a real image of a distant
object by placing the object beyond the focal point of the lens. A
diverging lens can be used to form a virtual image of a nearby
object by placing the object inside the focal point of the lens.
9 List out the different types of images in the view of the Computer Vision.(4)
Ans:- In the context of Computer Vision, images can be classified into several types
based on their characteristics and properties. The different types of images in
Computer Vision include:
1. Binary image: A binary image is a black and white image in which each pixel is either
black (0) or white (1). Binary images are commonly used for image processing tasks
such as edge detection, segmentation, and morphology.
2. Grayscale image: A grayscale image is an image in which each pixel has a single
intensity value, typically ranging from 0 (black) to 255 (white). Grayscale images are
commonly used for image analysis tasks such as feature extraction, object recognition,
and classification.
3. Color image: A color image is an image in which each pixel is represented by a
combination of red, green, and blue (RGB) values. Color images are commonly used in
computer vision applications such as object detection, segmentation, and tracking.
4. Depth image: A depth image is a type of image that represents the distance of each
pixel from the camera. Depth images are commonly used in applications such as 3D
reconstruction, object recognition, and human pose estimation.
These are some of the common types of images used in Computer Vision, and each
type has its own unique characteristics and applications.
10 Analyse the Image Noise Filters techniques you know and explain with an example. (7)
Ans:- Image noise is an unwanted variation in pixel intensity that can
be caused by several factors such as electronic interference, sensor
noise, or environmental factors. Image noise can reduce the quality
of an image and make it difficult to perform image analysis tasks.
Therefore, noise removal is an important pre-processing step in
image processing.
There are several techniques available for removing image noise,
and these techniques can be broadly classified into two categories:
spatial domain filters and frequency domain filters.
11 List the difference between Linear Filters and Non-Linear Filters? With suitable example.
(7)
Ans;- Linear filters and non-linear filters are two broad categories of
image filters used for image processing tasks. The main difference
between these two types of filters is how they operate on the pixel
values of the image.
Linear Filters: Linear filters are a class of image filters that operate
on the pixel values of an image using a linear function. These filters
are based on the principle of convolution, which involves sliding a
small matrix, called a kernel or mask, over the image and
computing the weighted sum of the neighboring pixels at each
location.
12 Discuss the Opening and closing morphology operations and its uses in Image Processing.
(7)
Opening and closing are two fundamental morphological operations in image
processing that are used for various image enhancement tasks, such as noise
reduction, object extraction, and shape analysis. Both operations are based on the
idea of structuring elements, which are small binary images that define the shape
and size of the features to be extracted or modified.
1. Image Denoising - The opening operation is used to remove small noisy regions
from an image, while the closing operation is used to fill in gaps and smooth the
edges of larger objects.
2. Object Extraction - The opening operation is used to remove small objects from an
image, while the closing operation is used to fill in gaps and connect the broken
edges of larger objects.
3. Shape Analysis - The opening and closing operations are used to extract the
boundaries and contours of objects in an image, which can be used to measure the
shape, size, and orientation of objects.
4. Edge Detection - The opening and closing operations can be used to detect the
edges and boundaries of objects in an image by comparing the original image with
the opened or closed image.
13 Discuss the problem of median filter with even number of points in the window. (7)
Ans;- The median filter is a common technique used in image
processing for removing noise from an image. It is based on the
principle of replacing each pixel value with the median value of its
neighboring pixels. The median filter is effective in removing
impulsive noise, such as salt and pepper noise, from an image.
However, when the size of the median filter window is even, it can
lead to some issues. Specifically, the issue arises when computing
the median value for a set of even number of pixels in the window.
In this case, there is no unique median value, which leads to some
ambiguity in the result.
Copy code
14 Define image sampling and image quantization and discuss their role in the quality of an
image. (7)
Ans;-
Image sampling and image quantization are two important
concepts in digital image processing that have a significant impact
on the quality of an image.
15 Explain the term 'computer vision' and it's need with the help of a suitable example. (4)
Ans:- Computer vision refers to the field of artificial intelligence and
computer science that deals with enabling computers to interpret
and understand digital images and video. The goal of computer
vision is to replicate the abilities of human vision, such as
recognizing objects, identifying patterns, and making decisions
based on visual input.
The need for computer vision arises from the vast amounts of visual
data that are generated and consumed every day, from security
cameras to social media photos. With the help of computer vision,
machines can analyze, interpret, and extract insights from this visual
data, enabling them to automate tasks, make predictions, and assist
humans in decision-making.
1. Structure:
2. Power consumption:
CCD sensors consume more power than CMOS sensors due to their
higher operating voltage and the need for external clock signals to
read out the stored charges.
3. Image quality:
CCD sensors are known for their high image quality, low noise, and
excellent dynamic range, making them ideal for applications where
image quality is of utmost importance, such as professional
photography and scientific imaging.
CMOS sensors have improved significantly in recent years and are
now capable of producing high-quality images with low noise and
good dynamic range. However, they may not match the image
quality of CCD sensors in certain applications.
4. Speed:
CMOS sensors are generally faster than CCD sensors due to their
on-chip circuitry and the ability to read out pixels in parallel.
CCD sensors are slower than CMOS sensors due to the sequential
readout of pixels along the electrodes.
17 What do you mean by the aperture of the camera, discuss aperture's role in the quality of
an image. (5)
Ans:- In photography, the aperture of a camera refers to the
opening in the lens through which light enters the camera. It is
represented by an f-number, which determines the size of the
aperture. A smaller f-number represents a larger aperture, while a
larger f-number represents a smaller aperture.
1. Depth of field:
2. Exposure:
The aperture also plays a key role in determining the exposure of
the image. The larger the aperture, the more light enters the
camera, which can result in a brighter image. The smaller the
aperture, the less light enters the camera, resulting in a darker
image. Controlling the aperture in conjunction with shutter speed
and ISO allows for precise control of the exposure of the image.
cssCopy code
In the diagram, the object is illuminated with a light source, and the
reflected light enters the camera lens. The lens refracts the light to
focus it onto the camera's sensor, where the image is captured as
pixels and stored as a digital image.
19 Electromagnetic Spectrum defines the ranges of Electro-Magnetic Radiation from the Sun.
Explain how it helps in vision process. (7)
Ans:- The electromagnetic spectrum is a range of frequencies of
electromagnetic radiation that includes visible light, as well as other
forms of radiation such as radio waves, microwaves, infrared
radiation, ultraviolet radiation, X-rays, and gamma rays.
20 Illustrate how image processing, computer graphics and computer vision are related to
each other. (7)
Ans:-
Image processing, computer graphics, and computer vision are all closely
related fields in computer science, each with its own specific focus and
applications. While there is some overlap between these fields, they all have
unique characteristics and applications.
Computer graphics, on the other hand, deals with the generation of visual
content using computer programs. The goal of computer graphics is to
create realistic or stylized images, animations, or visual effects for various
applications, including video games, movies, and virtual reality. Computer
graphics involves the use of computer programs to create, manipulate, and
render images, as well as to simulate complex physical phenomena.
21 Image processing is the process of transforming an image into a suitable form that is more
appropriate for a particular application. Explain the steps required to execute in the entire
transformation process. (7)
Ans:- The image processing transformation process involves several
steps that are required to execute in a specific sequence to achieve
the desired outcome. These steps are as follows:
Once the features have been extracted, they can be used for various
tasks such as image classification, object detection, and recognition.
The choice of feature extraction technique depends on the specific
application and the type of features required.