Artificial Intelligence – NBE Key 2024-25

Unit 6: Computer Vision

Subjective Type
A. Short answer type questions:
Note: Skip Q2 and 3 from
1 What do you mean by resolution?
Ans The number of pixels in an image is called its resolution.

2 What are considered good features of image processing?

Ans In image processing features like edges, corners, blobs and key point features are considered
as good features.

3 What are pixels?

Ans Pixel stands for picture element. A pixel is the smallest unit of information that constitutes
texts, images, or videos on a computer.

4 State the objective of Computer Vision.

Ans The goal of CV is to give AI models the ability to comprehend and interpret visual data from
all around the world in a manner that is similar to the human vision.

5 How are Computer Vision algorithms trained to identify image content?

Ans Models are pre-trained by being fed thousands of labelled images. They use this knowledge to
interpret, analyze and produce the required results or take appropriate action.

6 Explain the concept of feature extraction in Computer Vision.

Ans Feature extraction in computer vision refers to the process of identifying and isolating
important patterns or characteristics (such as edges, textures, and shapes) from an image that
are useful for tasks like classification, recognition, or detection. It reduces the dimensionality
of the data while preserving essential information for model learning.

Subjective Type – Application-based questions

B. Long answer type questions:
1 How do computers store RGB images?
Ans Every RGB image is stored on a computer in the form of three different channels: red, green
and blue channels. Each channel contains a number of pixels, with the value of each pixel
ranging from 0 to 255. When all channels are combined, they form a colored image. This
means that in an RGB image, each pixel has a set of 3 different values, which together give
color to that particular pixel. Each pixel within a colored image is a unique composition of the
three channels. For e.g. Purple is achieved by combination of RGB values (153,55,251)

2 Explain CV and provide any two applications of it.

The Global Edge School – Ahead Of The Learning Curve 1

 Artificial Intelligence – NBE Key 2024-25
Ans CV is a domain of AI that enables machines to interpret visual data, process it, and analyze it
using algorithms and methods to understand real-world phenomena. It helps machines derive
meaningful information from digital images, videos, and other visual inputs and take actions
based on that information.
<any two applications explained from page 294 to 297>

3 What is meant by the pixel value of a digital image?

Ans Each pixel in a digital image on a computer has a pixel value that determines its brightness or
color. The most common pixel format is the byte image, where this value is stored as an 8-bit
integer with a range of possible values from 0 to 255, where 0 is no color and 255 considered
as white.

4 Describe the CV tasks for a single object in an image.

Ans Classification and Classification + Localization are the CV tasks for a single objects
application.
 Classification – helps identify the main category of the object present in the photos.
This task involves assigning a label to the entire image based on its contents.
 Classification + Localization – this task finds the main category of the object and
where the object is in the image. It involves both processes of identifying what object
is present in the image and, at the same time, identifying at what location that object is
present in that image.

5 State the difference between CV and human vision

Ans Computer Vision Human Vision
Requires training with large datasets and Learns from experience and real-world
algorithms. interactions.
Can process images faster but depends on Relatively slower but highly efficient and
computational power. adaptable.
Requires specialized algorithms (e.g., stereo Has innate depth perception through
vision, CNN). binocular vision.
Relies on predefined rules and patterns for Interprets based on context, memory,
interpretation. intuition, and experience.

The Global Edge School – Ahead Of The Learning Curve 2

 Artificial Intelligence – NBE Key 2024-25
Cannot process emotional or subjective Can perceive emotions, feelings, and non-
information. verbal cues.
Limited by data and algorithms; struggles Excellent at understanding context and
with complex contexts. ambiguous scenes.

6 The facelock feature of a smart phone is an example of CV. Briefly discuss this feature.
(research based question)
Ans The Face Lock feature in smartphones uses computer vision to enable secure, hands-free
unlocking. It works by detecting the user's face, extracting unique facial features (like the
distance between eyes or shape of the nose), and comparing them with a pre-stored template.
If the match is successful, the phone unlocks. To enhance security, advanced smartphones
may use infrared or depth sensors to distinguish real faces from photos or videos, ensuring
accurate 3D recognition. This combination of convenience and security makes Face Lock a
popular biometric authentication method.

7 What is the difference between ‘image classification’ and ‘image classification and
localization’? Give an example to illustrate this difference.

Ans 1.Image Classification

 Task: It identifies what objects or categories are present in an image.
 Output: The model outputs a label (or multiple labels) that classify the entire image.
 Example: A model classifies an image as "dog" or "cat" but doesn't specify where in
the image the object is located.
 Illustration: If you input an image of a dog, the model simply outputs the label "dog,"
identifying the object.
2. Image Classification and Localization
 Task: It not only classifies the objects in an image but also identifies their location
within the image by providing bounding boxes around the objects.
 Output: The model outputs a label for the object and a bounding box (coordinates)
that specifies the object's position.
 Example: For an image with a dog, the model not only classifies it as "dog" but also
draws a box around the dog to show where it is in the image.
 Illustration: If you input an image of a dog, the model outputs "dog" along with a
bounding box showing the exact location of the dog within the image.
Key Difference
 Image classification tells what is in the image.
 Image classification and localization tells what is in the image and where it is located.

8. Apart from text-based search, search engines also carry out image-based searches. How
does CV help in this task? (research based question)
Ans Computer vision (CV) plays a crucial role in enabling image-based searches on search
engines through several techniques and processes:

The Global Edge School – Ahead Of The Learning Curve 3

 Artificial Intelligence – NBE Key 2024-25
 Image Recognition: Analyzes and identifies objects, scenes, or patterns within images
through feature extraction and classification.
 Content-Based Image Retrieval (CBIR): Measures similarity between the query image
and database images using extracted features for visually similar results.
 Object Detection: Identifies specific objects in images using bounding boxes and
semantic segmentation to understand content better.
 Image Tagging and Annotation: Automatically tags images with relevant keywords based
on content, improving organization and search accuracy.
 Reverse Image Search: Allows users to upload images for searching for similar images or
information by extracting features and matching them against a database.
 Identifying Image Sources: Helps find the origin of images or higher-resolution versions
by analyzing and comparing content.
 Augmented Reality (AR) Integration: Enables real-time object recognition for instant
information retrieval when users point their camera at an object.
 Contextual Information: Associates images with relevant metadata (like location or
activity) for improved search relevance and accuracy.

The Global Edge School – Ahead Of The Learning Curve 4