CS-803-A Image Processing and Computer Vision - May 2022 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aDescribe the concept of computer vision does it require in image processing.
   How to answer

   • Define computer vision and its goal of scene understanding from images
   • Relation to image processing: image processing as low-level preprocessing for vision
   • Whether/why image processing is required: enhancement, restoration, segmentation
2. 7 Marks
   bDescribe various elements of digital image processing system.
   How to answer

   • Block diagram of digital image processing system
   • Elements: acquisition, storage, processing, communication, display
   • Role of each element briefly
3. 7 Marks
   aDefine digital image. What do you mean by image sampling and quantization?
   How to answer

   • Define digital image as 2D function f(x,y) sampled and quantized
   • Define sampling: discretization of spatial coordinates
   • Define quantization: discretization of intensity levels
   • Effect on resolution and gray levels
4. 7 Marks
   bDiscuss morphological algorithm operations performed on gray scale images.
   How to answer

   • Grayscale morphology using min/max with structuring function
   • Dilation and erosion for gray-scale images
   • Opening, closing and reconstruction algorithms
   • Applications: smoothing, gradient, top-hat
5. 7 Marks
   aDescribe various computer vision models and their applications.
   How to answer

   • Overview of geometric/photometric vision models
   • Examples: camera projection, illumination, shape models
   • Applications in recognition, reconstruction, navigation
6. 7 Marks
   bExplain Hough transform with the help of suitable derivations.
   How to answer

   • Principle: point to curve mapping in parameter space
   • Derivation for line: $\rho = x\cos\theta + y\sin\theta$ and voting
   • Extension to circles/shapes and peak detection steps
7. 14 Marks
   Explain the following operations.

   1. i)dilation
   2. ii)erosion
   3. iii)cloting
   4. iv)thinning
   How to answer

   • Define dilation: expansion using structuring element
   • Define erosion: shrinking using structuring element
   • Define closing (cloting) and thinning with expressions
   • Effect/example of each on binary image
8. 7 Marks
   aExplain global and adaptive thresholding techniques.
   How to answer

   • Define thresholding and global threshold selection
   • Adaptive/local thresholding for non-uniform illumination
   • Algorithms: Otsu, mean/gaussian adaptive windowing
9. 7 Marks
   bCompare weak perspective projection and orthographic projection in affine projection models.
   How to answer

   • Define affine projection models
   • Orthographic projection equations and assumptions
   • Weak perspective: scaled orthographic with depth scaling and comparison
10. 7 Marks
    aExplain the use of neural network structures for pattern recognition with an example.
    How to answer

    • Neural network structure for classification: input, hidden, output layers
    • Training with feature vectors and backpropagation
    • Example: character/face recognition workflow
11. 7 Marks
    bDiscuss back tracking algorithm by an example.
    How to answer

    • Principle of backtracking search with constraints
    • Step-by-step example e.g., N-queens or segmentation search
    • Backtrack on failure and complexity remarks
12. 7 Marks
    aDefine and explain knowledge representation and Information Integration.
    How to answer

    • Define knowledge representation for vision systems
    • Methods: semantic nets, frames, rules, ontologies
    • Information integration/fusion across sources
13. 7 Marks
    bHow to evaluate extracted shape descriptors in 3D vision? Discuss.
    How to answer

    • Types of 3D shape descriptors: global, local
    • Evaluation metrics: precision, recall, robustness to pose/noise
    • Benchmarking procedure and discussion
14. 7 Marks
    aExplain the identifying road signs in vehicle vision system.
    How to answer

    • Pipeline: detection, segmentation, classification of signs
    • Features: color, shape e.g., Hough/circles, template matching
    • Challenges: illumination, occlusion in vehicle vision
15. 7 Marks
    bDiscuss about accurate centre location by using Hough transform.
    How to answer

    • Circle Hough transform for centre $(a,b,r)$ estimation
    • Voting in accumulator and peak for accurate centre
    • Refinement for sub-pixel accuracy and example