Computer Vision MCQs – Class 10 AI (417) | Prepare Smart, Score High
Get exam-ready with latest, important, syllabus-aligned CBSE Class 10 AI-417 Computer Vision MCQs covering key concepts and exam-oriented questions.
Ready to get started? 🚀 Let’s dive right into the MCQs! 💯
Q1. Computer Vision is the domain of AI that enables machines to
a) hear and respond to sounds
b) see, process and make sense of visual data
c) store only textual data
d) generate new hardware
Q2. Which domain of AI allows machines to “see” through images or visual data?
a) Data Sciences
b) Natural Language Processing
c) Computer Vision
d) Robotics Process Automation
Q3. In the comparison of the human vision system with a computer vision system, the sensing device (camera) corresponds to the
a) eye
b) brain
c) nose
d) ear
Q4. A school wants to automatically record student attendance by identifying their faces. Which technology is most suitable?
a) Image rescaling
b) Spreadsheet processing
c) Facial recognition using computer vision
d) Audio classification
Q5. Which of the following shows the correct order of layers in a CNN?
a) Pooling, Convolution, ReLU, Fully Connected
b) Convolution, ReLU, Pooling, Fully Connected
c) ReLU, Convolution, Fully Connected, Pooling
d) Fully Connected, Convolution, ReLU, Pooling
Q6. In the diagram relating AI, Machine Learning, Deep Learning and Computer Vision, which field encloses all the others?
a) Artificial Intelligence
b) Machine Learning
c) Deep Learning
d) Computer Vision
Q7. Which statement about Computer Vision and Image Processing is correct?
a) Both are exactly the same
b) Image Processing is a superset of Computer Vision
c) Computer Vision is a subset of Image Processing
d) Computer Vision is a superset of Image Processing
Q8. Which of the following is an example of Image Processing?
a) Object detection
b) Handwriting recognition
c) Correcting brightness
d) Identifying a face in a crowd
Q9. Which of the following is an example of Computer Vision?
a) Rescaling an image
b) Handwriting recognition
c) Changing the tones of a picture
d) Correcting brightness
Q10. Image Processing is mainly focused on
a) enhancing raw input images or preparing them for other tasks
b) predicting the meaning of visual input
c) detecting objects in videos
d) translating languages
Q11. The concept of Computer Vision was first introduced in the
a) 1950s
b) 1960s
c) 1970s
d) 1990s
Q12. Google’s “Search by Image” uses Computer Vision because it
a) compares features of the input image with a database of images
b) only reads the text typed by the user
c) converts the image into a video
d) deletes unwanted images
Q13. A shop studies customers’ movements, navigation routes and walking patterns. Which application is most appropriate?
a) Image compression only
b) Medical diagnosis
c) Facial filters
d) Computer vision in retail
Q14. Computer Vision in self-driving cars involves
a) only playing music
b) only fuel monitoring
c) only engine repair
d) identifying objects, getting navigational routes and monitoring the environment
Q15. In medical imaging, Computer Vision is used to convert
a) 3D models into 2D scans
b) 2D scan images into interactive 3D models
c) X-rays into audio files
d) text reports into images
Q16. A system must identify an object and indicate its location in an image containing a single object. Which task matches this requirement?
a) Image enhancement
b) Classification alone
c) Image compression
d) Classification and localisation
Q17. Which two tasks of Computer Vision are meant for multiple objects?
a) Classification and Classification + Localisation
b) Classification and Instance Segmentation
c) Classification + Localisation and Object Detection
d) Object Detection and Instance Segmentation
Q18. The image Classification task involves
a) drawing boxes around all objects
b) assigning an input image one label from a fixed set of categories
c) labelling every pixel of an image
d) enhancing the brightness of an image
Q19. Classification + Localisation identifies
a) what object is present and where it is located, for a single object
b) only the colour of the object
c) all objects and their pixels
d) only the size of the image
Q20. Finding instances of real-world objects such as faces, bicycles and buildings in images or videos is called
a) Classification
b) Convolution
c) Object Detection
d) Pooling
Q21. In Instance Segmentation, after detecting objects and giving them a category, each _____ is given a label.
a) image
b) video
c) kernel
d) pixel
Q22. The word “pixel” means
a) picture element
b) picture pencil
c) pixel pattern
d) printed pixel
Q23. As the number of pixels in an image increases,
a) the image becomes darker
b) the image becomes smaller in meaning
c) the colours disappear
d) the image resembles the original more closely
Q24. The number of pixels in an image is sometimes called its
a) kernel
b) resolution
c) channel
d) feature
Q25. A monitor resolution of 1280×1024 means
a) 1280 pixels from one side to the other and 1024 from top to bottom
b) 1024 pixels from one side to the other and 1280 from top to bottom
c) 1280 colours and 1024 shades
d) 1280 images and 1024 videos
Q26. One megapixel is equal to:
a) One million pixels
b) One thousand pixels
c) One hundred pixels
d) One billion pixels
Q27. The most common pixel format is the
a) word image
b) byte image
c) bit image
d) kilobyte image
Q28. In a CNN, the first convolution layer is conventionally responsible for capturing
a) low-level features such as edges, colour and gradient orientation
b) the final label of the image
c) probability values
d) the size of the dataset
Q29. In a byte image, each pixel value lies in the range
a) 0 to 100
b) 0 to 128
c) 0 to 255
d) 1 to 256
Q30. In a grayscale image, the pixel values 0 and 255 represent
a) white and black respectively
b) black and white respectively
c) red and green respectively
d) blue and yellow respectively
Q31. In a grayscale image, each pixel has a size of
a) 1 bit
b) 3 bytes
c) 1 byte
d) 1 kilobyte
Q32. A grayscale image has
a) three planes of 2D arrays
b) a single plane of a 2D array
c) two planes of 2D arrays
d) no pixel array
Q33. The main purpose of the pooling layer is to
a) reduce the spatial size of the convolved feature while retaining important features
b) increase the number of kernels
c) assign labels to the image
d) remove all edges
Q34. When R = G = B = 255, the output colour is
a) black
b) white
c) red
d) grey
Q35. When R = G = B = 0, the output colour is
a) white
b) red
c) green
d) black
Q36. How many separate colour channels are used to store an RGB image?
a) Three
b) One
c) Two
d) Four
Q37. In an RGB image, each pixel has
a) one value
b) two values
c) three values
d) four values
Q38. A pixel has RGB values (255, 0, 0). Which colour does it represent?
a) Blue
b) Green
c) Red
d) White
Q39. The online application Piskel is used to
a) train a classifier
b) mine data
c) translate text
d) create pixel art and make a GIF
Q40. Lobe.ai is described as
a) an Auto-ML, no-code AI tool for image classification
b) a programming language
c) a type of camera
d) a data mining tool for text
Q41. Teachable Machine was developed by
a) Microsoft in 2010
b) Amazon in 2015
c) Google in 2017
d) Apple in 2019
Q42. Teachable Machine runs on top of
a) Python.org
b) tensorflow.js
c) Orange
d) Piskel
Q43. The Orange Data Mining tool is used in the chapter to build a classification model for
a) coral bleaching
b) face filters
c) self-driving cars
d) pixel art
Q44. In computer vision, a feature is
a) the brand of the camera
b) the size of the monitor
c) the colour of the paper
d) a piece of information relevant for solving a computational task, such as a point, edge or object
Q45. Which layer of a CNN introduces non-linearity in the feature map?
a) ReLU layer
b) Pooling layer
c) Input layer
d) Fully connected layer
Q46. An image convolution is
a) addition of two images without a kernel
b) removal of all pixels
c) element-wise multiplication of the image array and a kernel followed by a sum
d) conversion of colour to grayscale
Q47. A kernel is
a) a type of image file
b) a matrix slid across the image to produce an enhanced output
c) a camera sensor
d) a label given to an image
Q48. The objective of the fully connected layer in a CNN is to
a) take the results of convolution/pooling and classify the image into a label
b) apply kernels on the image
c) remove negative numbers
d) reduce the image size
Q49. A Convolutional Neural Network (CNN) is
a) a type of camera
b) a Machine Learning tool for text only
c) a Deep Learning algorithm that takes in an input image and differentiates one aspect/object from another
d) a no-code tool
Q50. The first layer of a CNN is the
a) Fully Connected layer
b) Convolution layer
c) Pooling layer
d) ReLU layer
Q51. The output of the convolution layer is called
a) a label map
b) a pixel map
c) a kernel map
d) a feature map (also called an activation map)
Q52. What happens when pixel values in an image are changed?
a) The appearance of the image can change.
b) The image always becomes a video.
c) The image loses its dimensions automatically.
d) The image becomes a spreadsheet.
