CS-803-A Image Processing and Computer Vision - May 2023 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 in detail the various types of image filtering techniques.
   How to answer

   • Definition of image filtering and purpose (noise removal, enhancement)
   • Spatial domain techniques: linear smoothing, median, sharpening
   • Frequency domain techniques: low-pass, high-pass, homomorphic
   • Comparison and applications with examples
2. 7 Marks
   bWhat is meant by Image Transformation? Explain its needs in digital image processing.
   How to answer

   • Definition of image transformation
   • Need: decorrelation, energy compaction, enhancement, compression
   • Examples: Fourier, DCT, wavelet, Hough
   • Role in digital image processing pipeline
3. 7 Marks
   aWhat are the different types of computer vision models? Explain in detail.
   How to answer

   • Definition of computer vision models
   • Types: geometric, photometric, statistical, learning-based
   • Explanation of each model with working
   • Applications and comparison
4. 7 Marks
   bDiscuss in detail boundary descriptors of image processing.
   How to answer

   • Definition of boundary descriptors
   • Types: chain codes, Fourier descriptors, moments, shape numbers
   • Properties: invariance, compactness
   • Applications in shape representation
5. 7 Marks
   aWhat are the various representation schemes of digital image processing? Explain.
   How to answer

   • Concept of representation schemes
   • Chain codes, polygonal approximation, signatures, skeletons
   • Boundary vs regional representations
   • Selection criteria and examples
6. 7 Marks
   bWhat is Thresholding? Discuss its use and also write its limitations.
   How to answer

   • Definition of thresholding and basic principle
   • Uses in segmentation and binarization
   • Global vs local/adaptive thresholding
   • Limitations: illumination, noise, multi-modal histograms
7. 7 Marks
   aDiscuss how to measure the properties of the region in a labeled image?
   How to answer

   • Concept of labeled image and regions
   • Area, perimeter, centroid, compactness measures
   • Moment-based and topological properties
   • Steps to compute properties with example
8. 7 Marks
   bWhat is distance in Image processing? Discuss the distance relational approach in image matching frame work.
   How to answer

   • Definition of distance measures: Euclidean, city-block, chessboard
   • Distance transforms and relational approach
   • Use in template and image matching framework
   • Example of matching using distance
9. 7 Marks
   aWhat is Boundary analysis? Discuss the properties in analyzing the boundary.
   How to answer

   • Definition of boundary analysis
   • Properties: length, curvature, bending energy, shape numbers
   • Fourier and moment-based analysis
   • Steps in analyzing boundary
10. 7 Marks
    bExplain in brief how photogrammetric processing is done from 2D to 3D?
    How to answer

    • Concept of photogrammetry and 2D to 3D principle
    • Steps: camera calibration, stereo correspondence, triangulation
    • Depth estimation and 3D reconstruction
    • Applications and limitations
11. 7 Marks
    aWhat is image intensity in image processing? Discuss in brief the Intensity matching of ID signals in an image.
    How to answer

    • Definition of image intensity / gray levels
    • Concept of 1D signal intensity matching
    • Correlation / SSD matching approach
    • Challenges: illumination, noise
12. 7 Marks
    bExplain the importance of shape recognition in image processing.
    How to answer

    • Definition of shape recognition
    • Importance in object detection, OCR, medical imaging
    • Role in classification and retrieval
    • Examples of applications
13. 7 Marks
    aGive a brief summary on object recognition methods.
    How to answer

    • Definition of object recognition
    • Methods: template matching, feature-based, deep learning
    • Steps: detection, description, classification
    • Comparison of methods
14. 7 Marks
    bDiscuss how neural networks are used for image shape recognition.
    How to answer

    • Basics of neural networks for vision
    • CNN architecture: convolution, pooling, fully connected
    • Training for shape features and classification
    • Advantages with example
15. 14 Marks
    Explain in brief any two of the following:

    1. a)Thinning
    2. b)Motion-based segmentation
    3. c)View class matching
    4. d)Knowledge representation
    How to answer

    • Select any two topics from list
    • Definition and key principle of each
    • Steps / algorithm and diagram if relevant
    • Applications and limitations