Competency based multiple choice questions:

Question 1

What is the primary function of computer vision in the field of Artificial Intelligence?

A) To allow machines to learn and solve problems . B) To give computers the ability to see and understand
the world like humans. C) To automate tasks that would be repetitive for humans. D) To mimic human
decision-making capabilities.

Answer: B) To give computers the ability to see and understand the world like humans.
Explanation:

Computer vision is a field of AI that focuses on enabling computers to interpret and understand visual data
from the real world, similar to how humans see. This allows computers to perform tasks like object
detection, image classification, and facial recognition.

Question 2

What are the two main categories of image processing tasks in computer vision applications?

A) Single objects vs. multiple objects . B) Classification vs. detection . C) Localization vs. segmentation . D)
Resolution vs. dimension .

Answer: B) Classification vs. detection.

Computer vision applications can be categorized into two main tasks based on the type of output they
produce: classification and detection. Classification involves identifying the category of an object in an
image, while detection involves identifying the location and category of objects in an image.

Question 3

What is the term for the smallest unit of information in a digital image?

A) Byte . B) Megabyte . C) Pixel. D) Resolution .

Answer: C) Pixel.
Explanation:

A pixel is the fundamental unit of information in a digital image. It represents a single point in the image
grid and stores information about the color and intensity of the light at that point.
Question 4

What are the three primary color channels that make up most colored digital images?

A) Red, Green, Yellow . B) Red, Green, Blue. C) Cyan, Magenta, Yellow . D) Black, White, Grey .

Answer: B) Red, Green, Blue.

Most colored digital images are made up of three primary color channels: Red (R), Green (G), and Blue (B).
By combining these channels in varying intensities, a wide range of colors can be produced.

Question 5

In the context of computer vision, what is the purpose of image features?

A) To determine the overall quality of the image . B) To identify and extract specific information from the
image content. C) To compress the image for storage purposes. D) To display the image on a screen .

Answer: B) To identify and extract specific information from the image content.
Explanation:

Image features are characteristics that can be extracted from an image to help computers understand its
content. These features can include edges, corners, shapes, and textures. By analyzing these features,
computer vision algorithms can identify objects, classify images, and perform other tasks.

Competency based multiple choice questions in the pattern of assertion reason:

Question 1
Assertion: Computer vision systems can be used to identify and locate specific objects within an image.
Reason: Computer vision algorithms are trained to recognize patterns and features within images.
 (a) Both Assertion and Reason are true, and the Reason is a correct explanation of the Assertion.
 (b) Both Assertion and Reason are true, but the Reason is not a correct explanation of the Assertion.
 (c) The Assertion is true, but the Reason is false.
 (d) The Assertion is false, but the Reason is true.
Answer: (a)
Explanation: Computer vision relies heavily on algorithms that identify features (like shapes, edges, and
textures). These can be localized through techniques like bounding boxes, which indicate the presence and
location of the object in the image.
Question 2
Assertion: A high-resolution image will always provide better results in a computer vision task than a low-
resolution image.
Reason: High-resolution images contain more pixels, which means more detailed information for the
computer vision system to process.
 (a) Both Assertion and Reason are true, and the Reason is a correct explanation of the Assertion.
 (b) Both Assertion and Reason are true, but the Reason is not a correct explanation of the Assertion.
 (c) The Assertion is true, but the Reason is false.
 (d) The Assertion is false, but the Reason is true.
Answer: (c)
Explanation: While high-resolution images do provide greater detail, they are not always the best choice.
High-resolution might increase computational cost; plus, depending on the task, low-resolution images
might contain sufficient information for accuracy while being significantly faster to process.
Question 3
Assertion: Computer vision can be used to automate tasks in warehouses, such as sorting and counting
inventory.
Reason: Computer vision algorithms can recognize shapes and colors, allowing them to distinguish between
different product types.
 (a) Both Assertion and Reason are true, and the Reason is a correct explanation of the Assertion.
 (b) Both Assertion and Reason are true, but the Reason is not a correct explanation of the Assertion.
 (c) The Assertion is true, but the Reason is false.
 (d) The Assertion is false, but the Reason is true.
Answer: (a)
Explanation: Warehouse automation is a key application of computer vision. Inventory objects can be
identified by shape and color, and additionally, computer vision algorithms can read barcodes or other
identifying marks, streamlining inventory management.
Question 4
Assertion: Human vision and computer vision process images in fundamentally different ways.
Reason: Human vision relies heavily on context and learned associations that computer vision systems need
to be explicitly trained for.
 (a) Both Assertion and Reason are true, and the Reason is a correct explanation of the Assertion.
 (b) Both Assertion and Reason are true, but the Reason is not a correct explanation of the Assertion.
 (c) The Assertion is true, but the Reason is false.
 (d) The Assertion is false, but the Reason is true.
Answer: (a)
Explanation: The human brain makes use of a vast knowledge base, allowing context and experience to
influence image interpretation. Computer vision systems must learn these higher-level abstractions from
massive amounts of data, which differs from human learning.
Question 5
Assertion: Pixels are the smallest units of information in a digital image.
Reason: Each pixel can only represent a single color at a given time.
 (a) Both Assertion and Reason are true, and the Reason is a correct explanation of the Assertion.
 (b) Both Assertion and Reason are true, but the Reason is not a correct explanation of the Assertion.
 (c) The Assertion is true, but the Reason is false.
 (d) The Assertion is false, but the Reason is true.
Answer: (a)
Explanation: Pixels are indeed the fundamental building blocks of digital images, with each pixel storing
the data that determines its color.

Competency based multiple choice questions in the pattern of statement1

Question 1
 Statement 1: Computer vision aims to replicate the abilities of human vision.
 Statement 2: Computer vision systems excel at understanding the overall context and meaning of an
image.
 Options:
o A. Both statements are true.
o B. Both statements are false.
o C. Statement 1 is true, Statement 2 is false.
o D. Statement 1 is false, Statement 2 is true.
 Answer: C. Statement 1 is true, Statement 2 is false.
 Explanation: Computer vision focuses on enabling machines to process and identify visual
information. However, current computer vision systems primarily focus on object identification and
pattern recognition, not the broader understanding of an image's context in the way that humans
understand images.
Question 2
 Statement 1: A pixel is the smallest controllable element in a digital image.
 Statement 2: Pixels are always represented using a single byte of data.
 Options:
o A. Both statements are true.
o B. Both statements are false.
 o C. Statement 1 is true, Statement 2 is false.
o D. Statement 1 is false, Statement 2 is true.
 Answer: C. Statement 1 is true, Statement 2 is false.
 Explanation: Pixels are indeed the basic building blocks of digital images. However, the amount of
data used to represent a pixel can vary. While the byte format is common, pixels in RGB images, for
example, would use three bytes (one for each color channel).
Question 3
 Statement 1: Object detection is a computer vision task that involves identifying the presence of
specific objects within an image.
 Statement 2: Instance segmentation goes beyond object detection by pinpointing the exact pixels
belonging to each instance of an object.
 Options:
o A. Both statements are true.
o B. Both statements are false.
o C. Statement 1 is true, Statement 2 is false.
o D. Statement 1 is false, Statement 2 is true.
 Answer: A. Both statements are true.
 Explanation: Object detection focuses on drawing bounding boxes around objects of interest, while
instance segmentation aims to create an outline for each object in the image, allowing for a finer
level of detail.
Question 4
 Statement 1: Convolutional Neural Networks (CNNs) are specifically designed for image
processing tasks.
 Statement 2 CNNs excel at recognizing patterns within images, making them ideal for computer
vision applications
 Options:
o A. Both statements are true.
o B. Both statements are false.
o C. Statement 1 is true, Statement 2 is false.
o D. Statement 1 is false, Statement 2 is true.
 Answer: A. Both statements are true.
 Explanation: CNNs are a type of neural network that have an architecture well-suited to identifying
patterns in images. They're widely used in computer vision for tasks like classification and object
detection.
Question 5
 Statement 1: Computer vision has revolutionized industries like agriculture and retail.
  Statement 2: Applications like self-driving cars and facial recognition are enabled by computer
vision technology.
 Options:
o A. Both statements are true.
o B. Both statements are false.
o C. Statement 1 is true, Statement 2 is false.
o D. Statement 1 is false, Statement 2 is true.
 Answer: A. Both statements are true.
 Explanation: Computer vision has become deeply integrated into various industries and
technologies, enabling a whole new range of applications that were not possible before.

Case study based multiple choice type questions:

Question 1
Scenario: Sarah is a farmer using an app on a drone to assess her orchard. The app uses computer vision to
estimate the number of apples ready for harvest on each tree. However, sometimes the app gets confused by
leaves or branches.
Question: Which image processing technique could help improve the app's accuracy?
 A. Blurring the image to reduce focus on smaller details.
 B. Increasing the image's brightness to make apples stand out.
 C. Applying edge detection to distinguish between apples and other objects.
 D. Converting the image to grayscale to reduce complexity.
Answer: C. Edge detection helps isolate the shapes of objects, making it easier to distinguish the roundness
of apples from irregular leaves and branches.
Question 2
Scenario: A self-driving car company is testing their vehicle's ability to navigate a busy city street. The
system needs to distinguish between pedestrians, other vehicles, and traffic signs.
Question: Which computer vision task is most critical for the car's safe operation?
 A. Classification
 B. Object detection
 C. Instance segmentation
 D. Classification + Localization
Answer: B. Object detection is crucial. The car must identify the presence of other objects and their types to
react accordingly.
Question 3
Scenario: You're building a computer vision model to recognize different dog breeds. You have a large
dataset of dog images, but there's an imbalance—far more pictures of Golden Retrievers than Chihuahuas.
Question: Which concept of machine learning is important to address in this scenario?
 A. Pixel resolution
 B. Convolutional Neural Networks (CNNs)
 C. Class imbalance
 D. Optical flow
Answer: C. Class imbalance means the model will be biased towards the overrepresented breed. You'll need
strategies to address this (e.g., collecting more Chihuahua images, oversampling them during training).
Question 4
Scenario: A security system uses facial recognition to grant building access. A new employee, Maya, is
having trouble getting the system to recognize her.
Question: Which of these factors is LEAST likely to be causing the issue?
 A. The system was trained mostly on images of people with lighter skin tones than Maya.
 B. Maya is wearing glasses, and the training images didn't include many people with glasses.
 C. The security camera is capturing images at a low resolution.
 D. Maya has recently changed her hairstyle.
Answer: C. Image resolution impacts recognition generally, but the other options are more specific to bias
and variation in appearance.
Question 5
Scenario: A museum wants to develop an interactive app where visitors can point their phones at artwork
and get information about the piece.
Question: This application would likely rely on which of the following?
 A. A database of artwork images and a classification model
 B. An object detection model trained on common museum objects
 C. Sensors that track the visitor's location in the museum
 D. A complex physics simulation of light and reflections
Answer: A. The core need is to match the user's image to an existing database and retrieve the correct
information.

Case study based multiple choice type questions in the pattern of assertion
reason:

Question 1
Assertion: A farmer using computer vision-equipped drones can accurately estimate the potential yield of
their fruit crop.
Reason: Computer vision can identify and count individual fruits, even in dense foliage.
 (a) Both the assertion and reason are true, and the reason is a correct explanation for the assertion.
 (b) Both the assertion and reason are true, but the reason is not a correct explanation for the assertion.
  (c) The assertion is true, but the reason is false.
 (d) The assertion is false, but the reason is true.
Answer: (a)
Explanation: Computer vision image analysis, especially when done from a drone to get an aerial
perspective, can see through the leaves of fruit trees. This allows the software to detect individual fruits and
provide a count. This count can be used in conjunction with other data (average fruit size/weight, etc.) to
estimate yield.
Question 2
Assertion: Computer vision algorithms can detect damage to structures after natural disasters faster than
traditional human assessment.
Reason: Satellites can capture high-resolution images of wide areas far more quickly than a team on the
ground can survey the same area.
 (a) Both the assertion and reason are true, and the reason is a correct explanation for the assertion.
 (b) Both the assertion and reason are true, but the reason is not a correct explanation for the assertion.
 (c) The assertion is true, but the reason is false.
 (d) The assertion is false, but the reason is true.
Answer: (a)
Explanation: In disaster scenarios, time is critical. Satellite imagery and computer vision analysis allow for
very rapid assessment over large areas. Algorithms designed to look for specific signs of damage (roof
collapse, broken infrastructure, etc.) can quickly identify areas in need of the most immediate attention.
Question 3
Assertion: A self-driving car cannot function under rainy conditions.
Reason: Raindrops on the camera lens will distort the computer vision system's perception of the road.
 (a) Both the assertion and reason are true, and the reason is a correct explanation for the assertion.
 (b) Both the assertion and reason are true, but the reason is not a correct explanation for the assertion.
 (c) The assertion is true, but the reason is false.
 (d) The assertion is false, but the reason is true.
Answer: (d)
Explanation: Self-driving cars are designed for various conditions, including rain. While raindrops on the
lens can be an issue, self-driving cars use a combination of sensors (cameras, LiDAR, radar, etc.). These
provide redundancy, and specific technology can help clean and mitigate the issue of a distorted camera
view.
Question 4
Assertion: An image classification system can only learn to recognize objects it has been specifically
trained on.
Reason: Computer vision models do not possess the ability to generalize knowledge beyond their training
dataset.
 (a) Both the assertion and reason are true, and the reason is a correct explanation for the assertion.
 (b) Both the assertion and reason are true, but the reason is not a correct explanation for the assertion.
 (c) The assertion is true, but the reason is false.
 (d) The assertion is false, but the reason is true.
Answer: (a)
Explanation: It's important to understand the limitations of traditional computer vision. These models are
highly specialized in tasks based on the data they've been fed during the training process. While there are
advances in transfer learning and generalization, it remains true that a major shift in the type of object the
model needs to recognize would require retraining it.
Question 5
Assertion: In an RGB image, a pixel with the value [255, 0, 0] would appear as pure red.
Reason: In the RGB color model, the red color channel is represented by the first value in the pixel's triplet.
 (a) Both the assertion and reason are true, and the reason is a correct explanation for the assertion.
 (b) Both the assertion and reason are true, but the reason is not a correct explanation for the assertion.
 (c) The assertion is true, but the reason is false.
 (d) The assertion is false, but the reason is true.
Answer: (a)
Explanation: The RGB color model is additive. A value of 255 in the red channel, with zero in the green
and blue channels, translates into the color red.

Case study based multiple choice type questions in the pattern of statement1
statement2:

Question 1
Statement 1: Computer Vision (CV) applications can identify and differentiate objects within an image.
Statement 2: A farmer could use an app equipped with CV technology to assess the ripeness of fruit on a
tree.
 A) Both statements are true and related.
 B) Both statements are true but not related.
 C) The first statement is true, the second is false.
 D) The first statement is false, the second is true.
Answer: A
Explanation: Computer vision can indeed distinguish objects, and ripeness can be determined by color
variations. Thus, a CV app could help a farmer make informed harvest decisions.
Question 2
Statement 1: CV-powered retail stores can automatically add items to a customer's virtual cart.
Statement 2: In such a store, cameras and sensors track which items customers pick up and place in their
physical shopping carts.
 A) Both statements are true and related.
 B) Both statements are true but not related.
 C) The first statement is true, the second is false.
 D) The first statement is false, the second is true.
Answer: C
Explanation: While the first statement is accurate (a la Amazon Go), the second statement is false. These
stores rely on items being picked up from shelves, not traditional carts, to trigger the item being added
virtually.
Question 3
Statement 1: The concept of pixels is fundamental to understanding how computers process images.
Statement 2: A blurry photograph suggests it was likely taken with a high-resolution camera.
 A) Both statements are true and related.
 B) Both statements are true but not related.
 C) The first statement is true, the second is false.
 D) The first statement is false, the second is true.
Answer: C
Explanation: Pixels are indeed the building blocks of a digital image. However, a blurry photo indicates
low resolution (fewer pixels), not high resolution.
Question 4
Statement 1: Computer vision has applications in autonomous vehicles.
Statement 2: A self-driving car cannot use computer vision to distinguish between a traffic light and a
streetlight.
 A) Both statements are true and related.
 B) Both statements are true but not related.
 C) The first statement is true, the second is false.
 D) The first statement is false, the second is true.
Answer: C
Explanation: CV is crucial for self-driving cars. Object detection and classification are essential for the car
to 'understand' its surroundings, which very much includes differentiating traffic lights from other light
sources.
Question 5
Statement 1: Human vision involves the eyes sensing light and sending signals to the brain for
interpretation.
Statement 2: Computer vision systems replicate this process with cameras acting as 'eyes' and algorithms as
the 'brain'.
 A) Both statements are true and related.
 B) Both statements are true but not related.
 C) The first statement is true, the second is false.
 D) The first statement is false, the second is true.
Answer: A
Explanation: The text describes how human vision works and draws an analogy to how computer vision
systems are designed to mimic this process.

Competency based very short answer type questions:

Question 1

Explain, in a sentence or two, the main goal of Computer Vision.

Answer The main goal of Computer Vision is to enable computers and machines to "see" and understand
the visual world in a similar way to humans, using cameras, algorithms, and data.
Explanation Computer vision aims to replicate aspects of human visual perception for machines.
Question 2

Name one real-world application of Computer Vision in retail businesses.

Answer Amazon Go stores, which use Computer Vision, sensors, and deep learning algorithms for
automated checkout and inventory management.
Explanation These stores track what shoppers take and automatically charge them, eliminating the need for
cashiers.
Question 3

What does the term 'pixel' stand for, and how does it relate to images?

Answer Pixel stands for "Picture Element." It is the smallest controllable unit of color in a digital image.
Images are made up of grids of pixels.
Explanation Pixels are the basic building blocks that, when arranged together, form a complete image.
Question 4

Why is object detection more complex than image classification?

Answer Object detection is more complex because it requires not only identifying the object's type
(classification) but also precisely locating it within the image (localization).
Explanation Object detection needs to determine the class of the object and create a boundary around it,
making the task more challenging.
Question 5

How does a computer represent a grayscale image differently from a colored (RGB) image?

Answer In a grayscale image, each pixel has a single intensity value (0-255), while in an RGB image, each
pixel has three values representing Red, Green, and Blue intensities.
Explanation Grayscale images have one channel, while RGB images have three separate channels for color
representation.

Case study based very short answer type questions:

Scenario 1: Agriculture with Drones

You're a farmer interested in using drones for crop assessment. Currently, you walk the fields manually.

 Question: Aside from crop yield estimation, list two other ways drones with computer vision (CV)
could provide valuable data for your farm.
Scenario 2: Self-Driving Cars

You're on the design team for a self-driving car. Bad weather (snow, fog) can confuse traditional cameras.

 Question: Beyond cameras, suggest two other sensors a self-driving car could use and explain how
the data from those sensors would improve CV decision-making in poor visibility.
Scenario 3: The Curious Cat

Your cat loves batting at hanging toys. You want to build a small robot that uses CV to track the toy.

 Question: Describe two key image features (beyond just location) that your robot's CV system
would need to track so it can successfully swat the toy.
Scenario 4: Retail Revolution

A clothing store wants to use CV to suggest outfits based on a shopper's initial selection.

 Question: To make accurate suggestions, what are two types of data the CV system would need to
be trained on, beyond just images of clothing items?
Scenario 5: Museum Security

A museum wants to use CV for security, but is concerned about privacy. They don't want to identify
individuals, just unusual behavior.
  Question: Describe one type of unusual behavior a CV system could be trained to detect without
needing to identify specific people.
Explanation:
 Creative & Innovative: These questions go beyond basic definitions and ask students to apply their
understanding of CV in new scenarios.
 Unique: The scenarios are specific and relatable, promoting deeper problem-solving.
 Case Study Based: They frame CV as a tool to solve real-world problems.
 Random Topics: Questions touch on agriculture, robotics, retail, and security, showcasing the
breadth of CV applications.

Remembering and Understanding questions:

1. What is the definition of computer vision according to the document?

Answer: Computer vision is a field of Artificial Intelligence that enables computers to see, observe and
understand digital images or data, process them by acquiring, screening, analysing, identifying and
extracting information using the machine learning and neural network algorithms.

2. What are the two main categories of computer vision tasks based on the number of input
images?

Answer: The two main categories of computer vision tasks based on the number of input images are:

 Single Objects: This involves giving one image as input to the application.
 Multiple Objects: This involves giving multiple images as input to the application.
3. What is the difference between classification and localization tasks in computer vision as
described in the document?

Answer: Classification refers to the process of finding the class or category of the input image. For example,
identifying an image of a monument as India Gate. Localization refers to finding the location of the object in
the image. Classification + Localization combines these tasks to identify the category of the object along
with its location in the image. For instance, identifying the image of a monument as India Gate and its
location in Delhi, India.

4. What are pixels and what is their role in digital images according to the document?

Answer: Pixels are the smallest unit of information in a digital image. They are arranged in a 2-dimensional
grid to form a complete image. The color of a pixel is determined by the number of bits it represents. The
more pixels an image has, the more closely it resembles the original image. Resolution refers to the number
of pixels present on a screen.
 5. What are the three primary colors that form the basis of all colored images as mentioned in the
document?

Answer: The three primary colors that form the basis of all colored images are Red (R), Green (G), and Blue
(B). By mixing these colors in varying intensities, all other colors are created. Every colored image is stored
in the form of 3 different channels, representing each primary color. Each channel has a pixel value varying
from 0-255. For instance, turquoise color is a combination of Red (64), Green (224), and Blue (208).