Computer Vision (AL-701) - Important Questions

Explain what computer vision is and how it differs from traditional image processing, with suitable examples.

Perform the following numerical operations on a $3\times3$ image matrix: $$I = \begin{bmatrix}100 & 150 & 200 \\ 50 & 100 & 150 \\ 0 & 50 & 100\end{bmatrix}$$ Add 30 to each pixel, multiply by 1.2 for contrast adjustment and discuss clipping and datatype effects.

Explain data type conversion in digital image processing, including mathematical operations on images, OpenCV example, and impact on image quality and storage.

Explain binary image thresholding and demonstrate adaptive thresholding on a $4\times4$ grayscale image with a suitable numerical example.

Describe morphological erosion and dilation and compute dilation of the binary image $$I = \begin{bmatrix}0 & 1 & 0 \\ 1 & 1 & 0 \\ 0 & 1 & 1\end{bmatrix}$$ with a given $3\times3$ structuring element.

What is contour analysis and how can contours be used for object shape recognition?

Explain different types of color spaces and their transformation equations for RGB$\rightarrow$HSV and RGB$\rightarrow$LAB.

Numerically compute Histogram Equalization for an image having pixel intensities [52, 55, 61, 66, 70, 61, 64, 73] and show the equalized histogram.

Explain convolution in image filtering and compare smoothing techniques such as Box Blur, Gaussian Blur and Median Blur with examples.

Explain the concept of introduction to image filtering and its importance in image processing. Discuss linear and non-linear filters with suitable examples.

Derive the Sobel operator for gradient calculation and show its effect on a sample $3\times3$ image.

Discuss the various techniques for image segmentation and explain how image classification differs from object detection.

Explain the process of gesture recognition using OpenCV and deep learning. Discuss its applications in human-computer interaction.

Explain Motion Estimation and Object Tracking algorithms with help of examples.

Write a short note on any two: (i) Face detection using deep learning (ii) CLAHE for advanced contrast enhancement (iii) Ethical considerations and bias in computer vision applications.