AL-701 Computer Vision - Jun 2025 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
   aWhat are color spaces and how are they used in image processing? Discuss the role of histogram equalization in image enhancement.
   How to answer

   • Define color spaces with examples RGB, HSV, YCbCr, LAB
   • Use in image processing: segmentation, enhancement, analysis
   • Define histogram equalization and CDF-based contrast stretching
   • Role in enhancement with diagram of histogram before/after
2. 7 Marks
   bExplain convolution in image filtering. Compare and contrast different smoothing techniques such as Box Blur, Gaussian Blur and Median Blur.
   How to answer

   • Define convolution with kernel sliding operation and formula
   • Purpose of smoothing for noise removal
   • Compare Box vs Gaussian vs Median working principle
   • Contrast edge preservation and salt-pepper vs Gaussian noise handling
3. 7 Marks
   aDefine computer vision and explain its significance in modern applications. How does it differ from traditional image processing?
   How to answer

   • Define computer vision as high-level scene understanding
   • Significance in modern applications: autonomous, medical, surveillance
   • Traditional image processing vs vision: pixels vs semantics
   • Examples differentiating low-level and high-level tasks
4. 7 Marks
   bDiscuss various problems in computer vision and explain how basic image operations help in solving them.
   How to answer

   • List core problems: segmentation, recognition, motion, 3D reconstruction
   • Explain challenges of illumination, occlusion, viewpoint
   • Role of basic operations: filtering, thresholding, morphology
   • How preprocessing aids solving higher-level problems
5. 7 Marks
   aExplain the concept of binary image processing. How do thresholding, erosion, and dilation contribute to object recognition?
   How to answer

   • Define binary image and thresholding types
   • Explain erosion and dilation with structuring element
   • Contribution to noise removal and shape extraction
   • Use in object recognition pipeline
6. 7 Marks
   bDiscuss contour analysis and connected component analysis. How are these methods used in object segmentation?
   How to answer

   • Define contours and connected components labeling
   • Steps of contour extraction and component analysis
   • Use for counting, shape features, segmentation
   • Example of object segmentation workflow
7. 14 Marks
   Discuss the various techniques used for image segmentation. How does image classification differ from object detection?
   How to answer

   • List segmentation techniques: threshold, edge, region, clustering, deep learning
   • Explain U-Net, Mask R-CNN briefly
   • Define image classification vs object detection outputs
   • Difference with bounding box, localization and examples
8. 14 Marks
   Explain the significance of gesture recognition in computer vision. How can motion estimation be applied in real-world scenarios?
   How to answer

   • Define gesture recognition and its HCI significance
   • Pipeline: detection, tracking, classification
   • Define motion estimation: optical flow, background subtraction
   • Real-world uses: surveillance, gaming, autonomous driving
9. 14 Marks
   How does face detection work in computer vision? Explain the role of deep learning techniques in improving face detection accuracy.
   How to answer

   • Outline face detection pipeline: Viola-Jones to CNN
   • Explain Haar, HOG vs MTCNN, RetinaFace, YOLO-face
   • Role of deep learning: feature learning, robustness
   • Accuracy improvements with large datasets and variations
10. 14 Marks
    Discuss the ethical considerations in computer vision applications. How can bias in AI models be minimized?
    How to answer

    • List ethical issues: privacy, surveillance, consent, fairness
    • Explain dataset bias and its consequences
    • Methods to minimize bias: diverse data, auditing, fairness metrics
    • Regulation and responsible AI practices
11. 7 Marks
    aHow can computer vision techniques be adapted for remote sensing and satellite image analysis? Discuss the key challenges.
    How to answer

    • Adaptations for remote sensing: multispectral, high resolution handling
    • Techniques: classification, change detection, object extraction
    • Key challenges: scale, cloud, annotation, variability
    • Examples of satellite image analysis
12. 7 Marks
    bExplain the role of AI-driven computer vision in medical imaging. How has it transformed the healthcare industry?
    How to answer

    • Role in medical imaging: X-ray, MRI, CT, pathology
    • AI tasks: segmentation, detection, diagnosis assistance
    • Transformation: early detection, automation, telemedicine
    • Benefits and limitations in healthcare