MALLA REDDY ENGINEERING COLLEGE (AUTONOMOUS)

IV B. Tech II Semester (MR 21) I Mid Question Bank 2024-25 (Objective)

Subject: Introduction to Data Science (C6703) Branch: CSE-CS
Name of the Faculty: Mr S Mahipal
S.
Questions Ans
No.
Module-I

What is computer vision primarily concerned with?

1 A) Image compression B) Image enhancement C
C) Extracting information from images D) Image storage

Which of the following is a key challenge in computer vision?

2 A) Image resolution B) Image processing speed C
C) Accurate object recognition D) Image file format

In the pinhole camera model, what does the aperture represent?

3 A) The lens B) The focal point D
C) The image sensor D) The opening that allows light through

What is the purpose of histogram equalization in image processing?

4 A) To increase the image resolution B) To enhance the contrast of an image B
C) To convert the image to grayscale D) To compress the image

Which color space model is based on the additive combination of red, green, and blue?
5 D
A) HSV B) YUV C) CMYK D) RGB

What does the term 'image filtering' refer to?

6 A) Changing the color of an image B) Removing noise or emphasizing features in an image B
C) Compressing the image file D) Increasing the image size

Which of the following is a common edge detection algorithm?

7 B
A) Gaussian blur B) Canny edge detector C) Median filter D) Dilation
 What is the role of the convolution operation in image processing?
8 A) To change the image color B) To combine two images C
C) To apply a filter to an image D) To enhance image resolution

In image processing, what does the term 'color space conversion' refer to?
A) Changing the image resolution
9 B) Translating image data from one color model to another B
C) Adjusting the brightness of an image
D) Compressing the image file

What is the primary goal of image segmentation?

A) To enhance image contrast
10 B) To separate an image into different regions or objects B
C) To resize the image
D) To apply color corrections

What does the term 'morphological operation' in image processing refer to?
A) Changing the color of an image
11 B) Adjusting image contrast C
C) Operations based on the shape of objects in an image
D) Compressing the image file

Which of the following is a common technique for image noise reduction?

12 A) Histogram equalization B) Median filtering B
C) Edge detection D) Color space conversion

What does the term 'image resolution' refer to?

A) The number of colors in an image
13 B) The size of the image file C
C) The detail an image holds, typically measured in pixels
D) The image brightness
 Which algorithm is used for detecting corners in an image?
14 A) Canny edge detector B) Harris corner detector B
C) Gaussian blur D) Hough transform

What is the purpose of applying a Gaussian blur in image processing?

15 A) To detect edges B) To smooth an image and reduce noise B
C) To increase contrast D) To enhance colors

What does the term 'feature extraction' in computer vision refer to?
A) Adding features to an image
16 B) Identifying and extracting key points or objects from an image B
C) Compressing the image data
D) Changing the image resolution

Which color space is often used in image compression?

17 C
A) RGB B) CMYK C) YUV D) HSV

What is the primary function of a convolution kernel in image processing?

18 A) To detect edges B) To modify pixel values based on their neighbors B
C) To compress image data D) To change the color space

In image processing, what is a ' thresholding ' technique used for?

A) To detect edges
19 B) To segment an image by separating foreground from background B
C) To apply a color filter
D) To enhance image resolution

20 B

What is the purpose of using a Sobel filter in image processing?

 A) To detect colors B) To enhance edges C) To smooth an image D) To segment objects

What does the term 'object recognition' involve in computer vision?

A) Identifying the color of objects
21 B) Determining the location of objects in an image C
C) Identifying and classifying objects within an image
D) Changing the image resolution

Which algorithm is commonly used for edge detection in computer vision?

22 A) K-means clustering B) Hough transform C
C) Canny edge detector D) Principal Component Analysis

What is the purpose of using a dilation operation in morphological image processing?

23 A) To remove noise B) To expand the boundaries of objects B
C) To detect edges D) To compress image data

What is the main advantage of using the HSV color space over the RGB color space?
B
A) HSV provides better color compression
24 B) HSV is more intuitive for adjusting colors like hue and saturation
C) HSV has higher resolution
D) HSV is easier to convert to grayscale

In computer vision, what does 'image registration' refer to?

A) Converting an image to grayscale
25 B) Aligning multiple images into a single coordinate system B
C) Enhancing image contrast
D) Compressing image data

What is the main function of a Hough transform in image processing?

26 A) To detect circles and lines B) To enhance image contrast A
C) To perform edge detection D) To convert color spaces
 Which technique is used to correct geometric distortions in images?
27 A) Histogram equalization B) Image warping B
C) Median filtering D) Edge detection

What is the key feature of the ORB (Oriented FAST and Rotated BRIEF) algorithm?
28 A) It detects edges in an image B) It is a fast feature detector and descriptor B
C) It converts color spaces D) It enhances image resolution

Which operation in image processing is used to smooth out noise?

29 C
A) thresholding B) Erosion C) Blurring D) Dilation

What is the purpose of applying a high-pass filter in image processing?

A) To enhance high-frequency components such as edges
30 B) To remove noise A
C) To detect color changes
D) To convert images to grayscale

In the context of image processing, what is an 'artifact'?

A) A type of image filter
31 B) An error or distortion introduced during image acquisition or processing B
C) A method of edge detection
D) A color space model

What does 'local binary pattern' (LBP) feature describe in texture analysis?
A) Color histograms
32 B) Edge orientations C
C) Texture by comparing pixel intensity with neighboring pixels
D) Shape features
 What is the primary purpose of the Fourier transform in image processing?
A) To convert an image into its frequency domain
33 B) To detect edges A
C) To perform color space conversion
D) To smooth an image

In image processing, what does 'image normalization' refer to?

A) Adjusting the brightness and contrast of an image
34 B) Scaling pixel values to a specific range B
C) Converting the image to grayscale
D) Reducing image resolution

What is the main purpose of using a Kalman filter in computer vision?

A) To enhance image contrast
35 B) To track the movement of objects over time B
C) To detect edges
D) To segment objects

What is 'image binarization'?

A) Converting an image to black and white
36 B) Enhancing image colors A
C) Detecting edges
D) Increasing image resolution

Which technique is commonly used for image sharpening?

37 A) Gaussian blur B) Histogram equalization C
C) High-pass filtering D) Thresholding

38 A

What does the term 'depth of field' refer to in imaging?

A) The distance between the closest and farthest objects in focus
B) The range of colors in an image
C) The resolution of an image
 D) The size of the image file

In image processing, what does 'anisotropic filtering' do?

A) Reduces image noise
39 B) Enhances image sharpness based on the angle of viewing B
C) Converts colors from RGB to grayscale
D) Compresses image files

What is the main function of 'principal component analysis' (PCA) in computer vision?
A) To detect edges
40 B) To reduce dimensionality and extract key features B
C) To convert color spaces
D) To enhance image contrast

Which algorithm is used for image segmentation based on clustering?

41 A) K-means clustering B) Hough transform A
C) Canny edge detector D) Gaussian mixture model

What is 'image stitching' used for in computer vision?

A) To combine multiple images into a single panoramic image
42 B) To correct geometric distortions A
C) To enhance image resolution
D) To detect edges

Which term refers to the process of adjusting an image to correct its perspective distortions?
A) Image registration
43 B) Image warping B
C) Image normalization
D) Image binarization
 What is 'edge enhancement' used for in image processing?
A) To blur the edges of an image
44 B) To make edges more distinct and visible B
C) To change the color of edges
D) To increase image resolution

Which technique is commonly used to detect shapes in an image?

45 A) Histogram equalization B) Hough transform B
C) Median filtering D) Convolution

What does 'image contrast' refer to?

A) The number of colors in an image
46 B) The difference in brightness between different areas of an image B
C) The size of an image file
D) The resolution of an image

In computer vision, what does 'texture mapping' involve?

A) Applying a texture to a 3D model from a 2D image
47 B) Enhancing image colors A
C) Reducing image noise
D) Increasing image resolution

What is 'optical flow' used to measure in computer vision?

A) The speed of image processing algorithms
48 B) The motion of objects between frames in a video B
C) The depth of an image
D) The color changes in an image

49 A

What is 'image thresholding' commonly used for?

A) To convert an image to black and white
B) To apply a color filter
C) To enhance image resolution
 D) To compress image files

Which technique is used to detect and measure shapes like circles and ellipses in images?
A) Hough transform
50 B) K-means clustering A
C) Gaussian blur
D) Principal Component Analysis

Module-II
What is the primary goal of feature detection in computer vision?
A) To enhance image resolution
51 B) To identify and extract important points or regions from an image B
C) To convert color spaces
D) To compress image data
Which of the following is a commonly used feature descriptor?
52 A) SIFT B) Gaussian blur A
C) Median filter D) Histogram equalization
What does SIFT stand for in the context of image processing?
A) Scale-Invariant Feature Transform
53 B) Scale-Independent Feature Transform A
C) Spatial-Invariant Feature Transform
D) Scale-Invariant Feature Technique
Which algorithm is designed to detect corners in an image?
54 A) Harris corner detector B) Hough transform A
C) Canny edge detector D) ORB
What is the main advantage of using feature descriptors like ORB over SIFT?
A) ORB is more computationally expensive
55 B) ORB is invariant to scale and rotation C
C) ORB is faster and less computationally intensive
D) ORB is more accurate in detecting corners
Which technique is used to detect and describe local features in images robustly?
56 A) K-means clustering B) Hough transform C
C) SURF D) Principal Component Analysis
In feature matching, what is a 'descriptor'?
A) A color model
57 B) A method for enhancing image contrast C
C) A representation of the feature’s appearance
D) A method for edge detection
What is the purpose of feature matching in computer vision?
A) To detect edges
58 B) To identify corresponding points or features between images B
C) To convert color spaces
D) To enhance image resolution
Which of the following is NOT a feature extraction method?
59 C
A) SIFT B) ORB C) Gaussian Mixture Model D) SURF
60 What does the 'RANSAC' algorithm do in feature matching? B
 A) Detects edges in an image
B) Finds robust feature matches by iteratively removing outliers
C) Converts images to grayscale
D) Enhances image contrast
What is the main purpose of 'image descriptors' in feature matching?
A) To convert images to grayscale
61 B) To quantify the appearance of detected features B
C) To enhance image colors
D) To compress image data
Which algorithm is used for detecting blobs in an image?
A) Harris corner detector
62 B) Difference of Gaussian (DoG) B
C) Canny edge detector
D) Hough transform
What does 'ORB' stand for in image feature detection?
A) Oriented FAST and Rotated BRIEF
63 B) Original FAST and Robust BRIEF A
C) Optimal FAST and Rotated BRIEF
D) Oriented FAST and Refined BRIEF
What is a 'keypoint' in the context of image feature detection?
A) A point of high color intensity
64 B) A point of interest with distinctive features in an image B
C) A point where edges intersect
D) A point used for image compression
What is the 'FAST' algorithm used for in computer vision?
65 A) Feature detection B) Image segmentation A
C) Edge detection D) Color space conversion
Which of the following is a popular method for object recognition?
A) Hough transform
66 B) Convolutional Neural Networks (CNNs) B
C) Gaussian blur
D) Median filtering
What is the primary function of a 'convolutional layer' in a CNN?
A) To detect edges and features
67 B) To convert color spaces A
C) To apply a median filter
D) To perform image compression
What does 'transfer learning' in deep learning involve?
A) Training a model from scratch
68 B) Using a pre-trained model as a starting point for a new task B
C) Converting images to grayscale
D) Enhancing image resolution
CNNs, what is the purpose of pooling layers?
A) To detect edges
69 B) To reduce the spatial dimensions of the feature maps B
C) To apply color corrections
D) To perform image segmentation
What is 'object localization' in the context of computer vision?
A) Identifying objects in an image
70 B) Determining the precise location of objects within an image B
C) Converting images to grayscale
D) Enhancing image contrast
71 Which of the following is used to describe the spatial arrangement of features in an image? A
 A) Histogram of Oriented Gradients (HOG)
B) Principal Component Analysis (PCA)
C) Convolutional Neural Networks (CNNs)
D) K-means clustering
What is the main advantage of using HOG descriptors for object detection?
A) They are computationally inexpensive
72 B) They capture gradient information to describe object shapes B
C) They detect color changes
D) They convert images to grayscale
In the context of object recognition, what does 'classification' refer to?
A) Determining the type of object in an image
73 B) Segmenting the image into regions A
C) Enhancing image resolution
D) Detecting edges
What does 'bounding box' refer to in object detection?
A) A box used to segment an image
74 B) A rectangle that outlines the location of an object in an image B
C) A tool for edge detection
D) A method for image compression
Which algorithm is used for face detection and recognition?
75 A) Hough transform B) Convolutional Neural Networks (CNNs) B
C) Harris corner detector D) K-means clustering
What is the primary function of the 'softmax' layer in a neural network?
A) To apply non-linearity
76 B) To perform classification by converting scores into probabilities B
C) To reduce the dimensionality of data
D) To detect edges
What is 'data augmentation' in the context of training deep learning models?
A) Reducing the size of the dataset
77 B) Creating variations of training images to improve model robustness B
C) Converting images to grayscale
D) Compressing image data
What does 'feature fusion' refer to in computer vision?
A) Combining features from multiple sources to improve recognition accuracy
78 B) Converting images to grayscale A
C) Enhancing image resolution
D) Applying color filters
In object detection, what is 'non-maximum suppression' used for?
A) To remove redundant bounding boxes
79 B) To enhance image colors A
C) To convert color spaces
D) To detect edges
What is the primary use of 'Gaussian Mixture Models' (GMM) in image processing?
80 A) Image segmentation B) Edge detection A
C) Feature extraction D) Color correction
What does the 'Kalman filter' do in object tracking?
A) Enhances image contrast
81 B) Estimates the position and velocity of an object over time B
C) Converts images to grayscale
D) Detects edges
82 Which technique is used for detecting and tracking moving objects in a video? A
A) Optical flow
B) Histogram equalization
 C) Gaussian blur
D) Convolutional Neural Networks (CNNs)
In feature extraction, what is the primary goal of 'principal component analysis' (PCA)?
A) To convert color spaces
83 B) To reduce dimensionality and highlight important features B
C) To perform edge detection
D) To enhance image resolution
What is 'image clustering' used for?
A) Grouping similar images or features together
84 B) Detecting edges in an image A
C) Converting images to grayscale
D) Enhancing image colors
Which method is commonly used for detecting shapes like circles in an image?
A) Hough transform
85 B) Convolutional Neural Networks (CNNs) A
C) K-means clustering
D) Gaussian Mixture Models
What is 'object tracking' in computer vision?
A) Identifying objects in static images
86 B) Following the movement of objects across frames in a video B
C) Converting images to grayscale
D) Enhancing image contrast
What does 'mean shift' algorithm do in image processing?
A) Tracks objects by finding the mode of a probability distribution
87 B) Detects edges in an image A
C) Converts images to grayscale
D) Enhances image colors
What is 'feature matching' used for in computer vision?
A) To identify corresponding features in different images
88 B) To enhance image colors A
C) To perform image compression
D) To detect edges
Which method is used to compute the similarity between two feature vectors? A
89 A) Euclidean distance B) Gaussian Mixture Model
C) Convolutional Neural Networks D) Hough transform
What is 'visual localization' in the context of autonomous systems?
A) Determining the location of an object within a single image
90 B) Finding the location of the autonomous system within an environment B
C) Converting images to grayscale
D) Enhancing image resolution
What does 'object recognition' involve in computer vision?
A) Identifying and classifying objects in an image
91 B) Converting images to grayscale A
C) Enhancing image contrast
D) Performing edge detection
In object detection, what does 'anchor boxes' refer to?
A) Pre-defined bounding boxes used to predict the location of objects
92 B) Tools for enhancing image colors A
C) Methods for edge detection
D) Techniques for image compression
93 What is 'semantic segmentation' in computer vision? A
A) Segmenting an image into regions based on object categories
B) Converting images to grayscale
 C) Detecting edges in an image
D) Enhancing image resolution
What is 'instance segmentation'?
A) Identifying individual object instances and their boundaries in an image
94 B) Converting images to grayscale A
C) Performing image compression
D) Detecting edges
What is the purpose of the 'relu' activation function in neural networks?
A) To apply non-linearity and introduce activation
95 B) To enhance image colors A
C) To convert color spaces
D) To perform edge detection
What does 'feature engineering' involve in machine learning?
A) Creating new features from raw data to improve model performance
96 B) Converting images to grayscale A
C) Enhancing image contrast
D) Performing image compression
What is 'object classification' in the context of computer vision?
A) Determining the category or type of object in an image
97 B) Segmenting the image into regions A
C) Enhancing image resolution
D) Detecting edges
Which technique is commonly used for face recognition?
A) Histogram of Oriented Gradients (HOG)
98 B) Convolutional Neural Networks (CNNs) B
C) Hough transform
D) K-means clustering
What does the term 'depth learning' refer to in deep learning models?
A) Using multiple layers in neural networks to learn complex features
99 B) Converting images to grayscale A
C) Enhancing image resolution
D) Detecting edges
In feature matching, what is a 'homography'?
A) A transformation matrix used to map points between images
100 B) A technique for enhancing image colors A
C) A method for edge detection
D) A tool for image compression
Module-III

What is the primary goal of image segmentation?

A) To convert an image to grayscale
101 B) To divide an image into meaningful regions or objects
B
C) To enhance image resolution
D) To detect edges

102
A
Which algorithm is commonly used for image segmentation by clustering
 pixel values?
A) K-means clustering B) Harris corner detector
C) Hough transform D) Canny edge detector

What does 'thresholding' in image segmentation involve?

A) Converting an image to black and white based on pixel intensity
103 B) Enhancing image colors
A
C) Detecting edges
D) Reducing image noise

What is 'region growing' in the context of image segmentation?

A) Expanding regions by adding neighboring pixels with similar properties
104 B) Detecting edges in an image A
C) Clustering pixels based on color
D) Reducing image resolution

What does 'watershed segmentation' refer to?

A) A method based on topographic relief to segment images
105 B) A technique for edge detection
A
C) A method for image enhancement
D) A way to compress image data

In image segmentation, what is a 'superpixel'?

A) A pixel with enhanced contrast
106 B) A group of neighboring pixels with similar properties
B
C) A single pixel in an image
D) A pixel with a unique color value

107 A

What is the purpose of 'connected component labeling' in image segmentation?

A) To identify and label connected regions of pixels with similar attributes
B) To convert images to grayscale
 C) To detect edges
D) To enhance image colors

Which method is used for segmenting images by detecting boundaries between regions?
A) Edge detection
108 B) Region growing A
C) Watershed segmentation
D) K-means clustering

What does 'active contour' model do in image segmentation?

A) Detects object boundaries by evolving curves to fit the object shape
109 B) Converts images to grayscale A
C) Reduces image noise
D) Enhances image resolution

What is the main advantage of using 'graph cuts' in image segmentation?

A) It provides a global optimization of the segmentation problem
110 B) It enhances image colors A
C) It converts images to grayscale
D) It reduces image resolution

What is the primary goal of object detection?

A) To identify objects in an image and locate their positions
111 B) To convert images to grayscale A
C) To detect edges
D) To enhance image resolution

Which algorithm is used to detect objects by learning features from labeled images?
A) Convolutional Neural Networks (CNNs)
112 B) K-means clustering A
C) Gaussian Mixture Model
D) Principal Component Analysis (PCA)
 What does 'YOLO' stand for in the context of object detection?
113 A) You Only Look Once B) You Only Learn Once A
C) Your Object Locator D) Yield Object Locator

What is the purpose of 'Region-based CNNs' (R-CNNs) in object detection?

A) To extract regions from an image and classify them
114 B) To detect edges in an image A
C) To enhance image colors
D) To perform image compression

What does 'Non-Maximum Suppression' (NMS) do in object detection?

A) Eliminates redundant bounding boxes to keep the most accurate one
115 B) Enhances image colors A
C) Converts images to grayscale
D) Reduces image resolution

What is the main function of 'Region Proposal Networks' (RPNs) in Faster R-CNN?
A) To propose candidate object regions for further classification
116 B) To enhance image resolution A
C) To convert images to grayscale
D) To detect edges

What does 'Single Shot MultiBox Detector' (SSD) refer to in object detection?
A) A method that detects objects in a single pass through the network
117 B) A technique for enhancing image colors A
C) A method for converting color spaces
D) A tool for edge detection

118 A

Which algorithm uses anchor boxes for object detection?

A) SSD
B) K-means clustering
 C) Gaussian Mixture Model
D) Principal Component Analysis

What is the purpose of the 'RoI Pooling' layer in Faster R-CNN?

A) To pool features from proposed regions for object classification
119 B) To convert images to grayscale A
C) To perform edge detection
D) To enhance image resolution

What does 'Faster R-CNN' improve upon compared to R-CNN?

A) Speed and accuracy by integrating region proposal and object detection
120 B) Color enhancement A
C) Edge detection
D) Image resolution

What is the main benefit of using 'YOLO' for object detection?

A) It provides real-time detection with high speed and accuracy
121 B) It enhances image resolution A
C) It detects edges
D) It converts color spaces

What technique is commonly used for detecting multiple objects in an image using a grid-based
approach?
A) YOLO
122 A
B) K-means clustering
C) Gaussian Mixture Model
D) Principal Component Analysis

123 A

What is 'anchor-based detection' in object detection?

A) Using pre-defined bounding boxes to detect objects
B) Converting images to grayscale
C) Enhancing image colors
 D) Detecting edges

In object detection, what is the role of the 'bounding box regressor'?

A) To refine the location of detected objects by adjusting bounding box
coordinates
124 A
B) To enhance image contrast
C) To convert color spaces
D) To perform image compression

What is 'semantic segmentation' used for?

A) Labelling each pixel in an image with a class label
125 B) Detecting edges in an image A
C) Converting images to grayscale
D) Enhancing image resolution

Prepared By
Name: Mr S Mahipal
Signature:
HOD Signature