CS-803-A Image Processing and Computer Vision - Dec 2024 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
   aDiscuss the history and evolution of Computer Vision and Image Processing (CVIP).
   How to answer

   • Define Computer Vision and Image Processing (CVIP)
   • Timeline: early vision, digital image processing, AI and learning era
   • Key milestones and applications evolution
   • Current trends and relationship between CV and IP
2. 7 Marks
   bExplain the significance of image representations and image filtering.
   How to answer

   • Define image representation models
   • Need for filtering: noise, enhancement
   • Types: spatial and frequency domain filtering
   • Significance with examples
3. 7 Marks
   aDefine morphological image processing and describe its primary operations.
   How to answer

   • Define morphological image processing and structuring element
   • Principle of set theory on binary/grayscale images
   • Primary operations: erosion, dilation, opening, closing
   • Applications
4. 7 Marks
   bDifferentiate between dilation and erosion in binary and grayscale images.
   How to answer

   • Define dilation: expansion, max operation
   • Define erosion: shrinking, min operation
   • Difference in binary images with set logic
   • Difference in grayscale images with min/max filtering
5. 7 Marks
   aWhat is image segmentation? Discuss thresholding and connected component labeling techniques.
   How to answer

   • Define image segmentation and its goal
   • Thresholding: global, adaptive, Otsu principle
   • Connected component labeling: algorithm steps
   • Examples and comparison
6. 7 Marks
   bDescribe the split-and-merge technique used in hierarchical segmentation.
   How to answer

   • Concept of hierarchical quadtree split-and-merge
   • Splitting criterion for non-homogeneous regions
   • Merging of similar adjacent regions
   • Steps, advantages and example
7. 7 Marks
   aExplain edge detection and the Hough transform in area extraction.
   How to answer

   • Define edge detection and edge operators
   • Hough transform principle for line/curve detection
   • Parameter space voting process
   • Role in area extraction
8. 7 Marks
   bDescribe the least-square fitting technique in curve fitting and its applications.
   How to answer

   • Principle of least-square error minimization
   • Mathematical formulation for line/curve fitting
   • Steps to compute best-fit parameters
   • Applications in vision and measurement
9. 7 Marks
   aDefine region analysis and describe the concept of spatial moments.
   How to answer

   • Define region analysis
   • Define spatial moments and central moments
   • Formulas for area, centroid, orientation
   • Use in shape description
10. 7 Marks
    bExplain the role of signature properties and shape numbers in boundary analysis.
    How to answer

    • Define boundary, signature concept
    • Signature properties: translation, rotation invariance
    • Shape numbers and chain code basis
    • Role in boundary matching and recognition
11. 7 Marks
    aWhat is the backtracking algorithm? Explain its application in solving consistency labelling problems.
    How to answer

    • Define backtracking search
    • Consistency labelling problem formulation
    • How backtracking prunes inconsistent labels
    • Example application
12. 7 Marks
    bDescribe perspective projective geometry and inverse perspective projection.
    How to answer

    • Define projective geometry and perspective projection
    • Geometry of vanishing points and homogeneous coordinates
    • Inverse perspective projection concept
    • Use in 3D reconstruction
13. 7 Marks
    aWhat is knowledge-based vision and how does it support object recognition?
    How to answer

    • Define knowledge-based vision system
    • Components: knowledge base, inference, models
    • How prior knowledge constrains recognition
    • Example of object recognition support
14. 7 Marks
    bExplain the Hough transform in the context of simple object recognition methods.
    How to answer

    • Hough transform voting principle
    • Application to lines, circles for objects
    • Steps in simple object recognition
    • Advantages under noise/occlusion
15. 7 Marks
    aDefine Principal Component Analysis (PCA) and its application in feature extraction.
    How to answer

    • Define PCA and dimensionality reduction goal
    • Steps: covariance, eigenvectors, projection
    • Feature extraction by principal components
    • Application in face/image recognition
16. 7 Marks
    bDescribe how neural networks are used in image shape recognition.
    How to answer

    • Neural network architecture for shape recognition
    • Training with shape features/datasets
    • Classification and generalization process
    • Advantages and examples