AL-701 Computer Vision - Dec 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
   aExplain the key differences between Image Processing and Computer Vision. Illustrate your answer with suitable examples.
   How to answer

   • Define Image Processing (image to image) vs Computer Vision (image to understanding/decision).
   • Compare on goal, input/output, techniques, intelligence level.
   • Give examples e.g. enhancement/filtering vs object detection/face recognition.
2. 7 Marks
   bPerform the following numerical operations on a $3\times3$ image matrix: $$I = \begin{bmatrix}100 & 150 & 200 \\ 50 & 100 & 150 \\ 0 & 50 & 100\end{bmatrix}$$

   1. i)Apply contrast stretching operation with new intensity range (0,255).
   2. ii)Apply brightness enhancement by +50 intensity shift.
   How to answer

   • Find min=0, max=200 and apply contrast stretching formula s = (r-min)*255/(max-min).
   • Compute stretched 3x3 matrix values.
   • Add +50 to each pixel with clipping at 255 for brightness enhancement.
3. 7 Marks
   aDiscuss the mathematical operations performed on digital images. Explain datatype conversion with example in OpenCV.
   How to answer

   • List arithmetic/logical operations: addition, subtraction, multiplication, AND/OR/NOT.
   • Explain datatype conversion need e.g. uint8 to float32 for processing.
   • Give OpenCV example using astype / convertTo / cv2.convertScaleAbs.
4. 7 Marks
   bExplain the process of binary image thresholding, implement adaptive thresholding numerically on a $4\times4$ grayscale image.
   How to answer

   • Define thresholding: g(x,y)=1 if f>T else 0, global vs adaptive.
   • Explain adaptive mean/Gaussian threshold in local window.
   • Show numerical steps on 4x4 example to get binary output.
5. 7 Marks
   aDescribe morphological operations- Erosion and Dilation. Show the result of dilation on the binary image. $$I = \begin{bmatrix}0 & 1 & 0 \\ 1 & 1 & 0 \\ 0 & 1 & 1\end{bmatrix}$$ Using structuring element $$B = \begin{bmatrix}1 & 1 \\ 1 & 1\end{bmatrix}$$
   How to answer

   • Define erosion (shrink) and dilation (expand) with set theory.
   • Describe structuring element B operation.
   • Compute dilation of given I with 2x2 all-ones B step by step.
6. 7 Marks
   bWhat is Contour Analysis? How can contours be used for object shape recognition.
   How to answer

   • Define contour as boundary of same-intensity connected pixels.
   • Outline steps: preprocessing, findContours, approximation.
   • Explain shape recognition via moments, Hu moments, matching.
7. 7 Marks
   aExplain different types of color spaces and their transformation equations (RGB$\rightarrow$HSV, RGB$\rightarrow$LAB).
   How to answer

   • Describe RGB, HSV, LAB, YCbCr characteristics.
   • Give RGB to HSV and RGB to LAB transformation equations.
   • Mention use cases for each color space.
8. 7 Marks
   bNumerically compute Histogram Equalization for an image having pixel intensities. [52, 55, 61, 66, 70, 61, 64, 73]
   How to answer

   • Compute histogram, PDF and CDF for given intensities.
   • Apply equalization formula s = (L-1)*CDF.
   • Map old values to new equalized values.
9. 7 Marks
   aCompare Image Filtering methods Box blur, Gaussian blur and Median blur with help of example.
   How to answer

   • Define Box (averaging), Gaussian (weighted) and Median (non-linear) blur.
   • Compare kernel, edge preservation, noise handling.
   • Give example output on sample patch.
10. 7 Marks
    bDerive the Sobel operator used for gradient calculation and show its effect on a sample $3\times3$ image.
    How to answer

    • Derive Sobel Gx and Gy 3x3 kernels from smoothing + differentiation.
    • State gradient magnitude and direction formulas.
    • Show convolution on 3x3 sample and edge response.
11. 7 Marks
    aDifferentiate between first order and second order derivative filters with mathematical examples.
    How to answer

    • Define first-order (gradient: Sobel, Prewitt) vs second-order (Laplacian, LoG).
    • Give mathematical masks and derivative formulas.
    • Compare noise sensitivity and edge type detected.
12. 7 Marks
    bDiscuss various object detection techniques in Computer Vision with suitable examples.
    How to answer

    • List traditional (Haar, HOG+SVM) vs deep (R-CNN, YOLO, SSD).
    • Explain working of each category briefly.
    • Give application examples.
13. 7 Marks
    aExplain the process of Gesture Recognition using OpenCV and Deep Learning.
    How to answer

    • Outline pipeline: capture, preprocessing, hand segmentation, feature extraction.
    • Explain OpenCV part and CNN classifier part.
    • Mention real-time prediction and applications.
14. 7 Marks
    bExplain Motion Estimation and Object Tracking algorithms with help of examples.
    How to answer

    • Define motion estimation (optical flow, block matching).
    • Explain tracking algorithms: MeanShift, KCF, SORT/DeepSORT.
    • Give example of video surveillance / vehicle tracking.
15. 14 Marks
    Write a short note on any two:

    1. i)CLAHE (Contrast Limited Adaptive Histogram Equalization)
    2. ii)Image Segmentation
    3. iii)Connected Component Analysis
    4. iv)Applications of Deep Learning in Computer Vision
    How to answer

    • Select any two topics with definition and key steps.
    • Explain working principle with diagram/formula.
    • Mention applications and advantages.