ASSIGNMENT-SESSION 1 CLASS: X SUBJECT: AI

PART B- UNIT 5 COMPUTER VISION

1. What is Computer Vision?

Computer Vision (CV) is a subset of artificial intelligence that makes computers, machines
and devices visually enabled by giving them capability to analyze and understand the image
captured by camera.
2. What are some applications of Computer Vision?
(a) Facial Recognition System (b) Face Filters
(c) Image Search or Search Engines (d) Computer Vision in Retail
(e) Inventory Management (f) Self-Driving Cars
(g) Google Lens/Translate App (h) Medical Imaging
(i) Livestock Farming (j) Agriculture
3. What is facial recognition system?
A Facial Recognition System is a computer vision-based technology that identifies or
verifies or matches a digital image (live or stored) of a human face against a database of
stored face images. It is performed in three steps:
(a) Face detection. (b) Face capture. (c) Face match.
4. What are face filters? Where are these used?
A Face Filter is a computer-generated effect that applies present (predesigned) edits or
changes to a loaded facial image.
The face filters are computer-generated effects that first identify the image of face and
then get superimposed on facial images. It can work with your camera, adding a layer or
imagery in the foreground or background of your image.
5. What is reverse image search?
Reverse Image search refers to a search where the search query is formed using a pasted
image and the search engine looks for matching or similar images from around the Web.
6. How is CV helpful in retail and inventory management?
Computer vision (CV) is transforming retail and inventory management by automating
processes, improving efficiency, and enhancing decision-making.
Here’s how CV is helpful in these areas:
(a) Automated Inventory Tracking. (b) Loss Prevention (Theft Detection).
(c) Customer Behavior Analysis. (d) Automated Checkout and Cashier less Stores.
(e) Pricing and Labeling. (f) Improving Employee Productivity.
7. What is the role of CV in self-driving cars?
In self-driving cars, computer vision is used along with a good quality fast camera and with
automatic route navigation.
Computer Vision solves the following issues in self-driving cars:
(a) Lana Line Detection (b) Obstacle and Road Signs/Lights Detection
(c) Steering Angle Computation (d) Vehicle Tracking
8. What is the role of CV in Lens or translate apps?
Google Lens is an AI-powered technology that uses your smartphone camera and deep
machine learning to not detect the object in front of the camera lens but understand this
and offer options such as scanning, translation, shopping and more.
In Lens and translate apps, computer vision (CV) plays a crucial role in enabling real-time
interaction with the physical world by recognizing and interpreting visual information.
Here’s how CV contributes to these apps:
(a) Image Recognition for Object and Scene Understanding.
(b) Augmented Reality (AR) Integration.
(c) Handwriting Recognition.
(d) Improved User Interaction.
9. How is CV useful in medical imaging?
Computer vision (CV) is transforming medical imaging by improving accuracy, efficiency,
and early detection of diseases.
Here are keyways in which CV is useful in medical imaging:
(a) Automated Image Analysis. (b) Early Disease Detection.
(c) Image Segmentation and Classification. (d) Enhancing Image Quality.
CV has many applications such as:
(a) Skin Cancer Detection. (b) Surgery Simulations.
(c) Diagnosis Aid. (d) Pneumonia Detection.
10. How is CV useful in agriculture in Livestock farming and in crop monitoring?
Computer vision (CV) is becoming increasingly useful in agriculture, particularly in livestock
farming and crop monitoring, by automating processes, enhancing productivity, and
enabling data-driven decision-making.
Here's how CV plays a key role in these areas:
1. Computer Vision in Livestock Farming:
(a) Animal Health Monitoring. (b) Automated Milking and Feeding.
(c) Disease Spread Prevention. (d) Livestock Weight Estimation.
2. Computer Vision in Crop Monitoring:
(a) Disease and Pest Detection. (b) Weed Detection and Management.
(c) Automated Crop Sorting and Grading. (d) Yield Estimation and Harvest Planning.
 ASSIGNMENT-SESSION 2 CLASS: X SUBJECT: AI
PART B- UNIT 5 COMPUTER VISION

1. Define these terms:

(a) pixel: represents the smallest piece of the screen that can be controlled individually in
terms of colour and intensity.
(b) resolution: refers to the quality and size of an image in terms of the number of pixels
it contains, typically as ‘width x height’.
(c) bitmap graphic: also known as a raster image, is a type of digital image composed of a
grid of individual pixels (picture elements). Each pixel holds specific color information, and
when combined, these pixels create the complete image.
(d) colour depth: the amount of information that is stored about a pixel determines its
colour depth (pixel value), which controls how precisely the pixel’s colour can be specified.
2. What are two types of digital images?
(a) Bitmap (Raster) Images are composed of a grid of individual pixels, where each pixel
represents a color and brightness value.
(b) Vector Images are composed of paths, which are defined by mathematical equations.
These paths consist of points, lines, curves, and shapes.
3. How are RGB images internally processed?
RGB images are internally processed by computers as a grid of pixels, with each pixel having
three color components: Red (R), Green (G), and Blue (B). These three colors are combined
in varying intensities to produce a wide range of colors. By combining these colours, all
other colours are generated in digital images.
4. How are grayscale images internally process?
Grayscale images are processed as 2D matrices of pixel values, where each pixel
represents an intensity level rather than color. In a grayscale image, each pixel carries an
intensity value that represents the brightness at that point.
The range of intensity values typically spans from 0 to 255 in an 8-bit image, where:
(a) 0 represents black.
(b) 255 represents white.
(c) Intermediate values (1 to 254) represent shades of gray.
5. What are different computer vision tasks?
(a) Image Classification (b) Localisation
(c) Object Detection (d) Image Segmentation
6. What do you understand by image classification?
Image Classification refers to the act of identifying and classifying a given image as
belonging to one of a set of predefined categories or classes. Image classification is a
fundamental task in computer vision where an algorithm assigns a label or category to an
entire image based on its visual content. The goal is to categorize the image into one or
more predefined classes, such as identifying whether an image contains a cat, dog, car, or
other objects.
7. What is localization task in computer vision?
Localization refers to identifying the position or location of an object. Location performs
classification along with localizing the object. It looks for class-specific details
(classification) in an image or a video and identifies their position in that image
(localization).
8. What do you understand by the object detection task in computer vision?
Object Detection refers to a combined action of localization and classification, carried out
on different items/things/subjects of interest in an image/video to identify and label them
as objects of specific types and know their count.
9. What is image segmentation? How does it take place?
Image segmentation in computer vision involves dividing an image into multiple segments
or regions to simplify its analysis. Its goal is to assign a label to each pixel in the image, such
that pixels with the same label share certain visual characteristics like color, intensity, or
texture. This helps in understanding the structure of objects within an image, allowing for
more precise object detection, recognition, and analysis.
Steps in the Image Segmentation Process:
(a) Preprocessing: Normalize the image and remove noise, Resize the image if necessary.
(b) Feature Extraction: Extract features like color, intensity, texture, or edge information
depending on the segmentation method.
(c) Segmentation: Apply a segmentation algorithm based on the type of task.
(d) Post-processing: Refine the segmentation map to eliminate noise, small unwanted
regions, or fill gaps.
(e) Evaluation: Assess segmentation accuracy using different metrics.
10. What is the role of algorithms in computer vision?
Algorithms are fundamental to the field of computer vision, serving as the backbone that
enables machines to interpret, analyze, and understand visual data from the world. Their
roles span a wide range of functions, from basic image processing to complex decision-
making tasks.
Algorithms are indispensable in computer vision, enabling machines to perform a vast
array of tasks that mimic human visual perception. From the foundational steps of image
preprocessing to the sophisticated processes of deep learning-based recognition and
segmentation, algorithms provide the necessary tools for computers to understand and
interact with the visual world. As technology advances, the role of algorithms continues to
expand, driving innovations in areas such as artificial intelligence, robotics, healthcare, and
beyond.
11. What is OpenCV? How is it useful?
OpenCV (Open Source Computer Vision Library) is an open-source software library that
provides a comprehensive set of tools and functions for real-time computer vision and
image processing. It was initially developed by Intel and later supported by Willow Garage
and Itseez. OpenCV is widely used in academic research and industrial applications for
tasks ranging from simple image manipulations to complex machine learning and deep
learning tasks.
Uses:
(a) OpenCV can be used to manipulate images with operations such as resizing, cropping,
rotating, filtering, and color conversion.
(b) OpenCV supports reading video streams from cameras or video files and processing
them frame by frame.
(c) OpenCV offers tools for reconstructing 3D structures from 2D images using stereo
vision.
(d) OpenCV is used for camera calibration and pose estimation, which are essential for
overlaying virtual objects onto real-world scenes.