JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD B.Tech III IMAGE PROCESSING Previous Question Paper (March - 2021 - R18)

Digital image processing involves the manipulation of digital images using computer algorithms. Key problems include image degradation (noise, blur, or low contrast), data redundancy requiring compression, and the challenge of feature extraction from complex, high-dimensional datasets. Applications are vast and include: 1) Medical Imaging: Enhancing X-rays, MRIs, and CT scans for better diagnosis. 2) Remote Sensing: Processing satellite imagery for weather forecasting, urban planning, and agricultural monitoring. 3) Biometrics: Fingerprint, iris, and facial recognition for security. 4) Computer Vision: Enabling autonomous vehicles and robotics to perceive their environment. 5) Image Compression: Reducing storage requirements for multimedia transmission.

The process of converting a gray-level image to a binary image is known as thresholding. It maps each pixel intensity value to either 0 (black) or 1 (white) based on a predefined threshold value (T).

1. Choose a threshold value T, typically in the range [0, 255].
2. Iterate through every pixel (x, y) in the input image with intensity f(x, y).
3. Apply the transformation function: g(x, y) = 1 if f(x, y) ≥ T, and g(x, y) = 0 if f(x, y) < T.

For example, consider a 3x3 image patch with intensities:

If we set a threshold T = 128, pixels with values 128 or higher become 1, and others become 0:

This results in a binary image where the object of interest is isolated from the background.