Image Processing with Gaussian, Uniform, and Salt-and-Pepper Noise Addition and Removal

This project implements image processing techniques starting with the addition of three common noise types: Gaussian noise (additive white noise with normal distribution), uniform noise (equal probability distribution across intensity range), and salt-and-pepper noise (random black and white pixel corruption). The implementation typically involves pixel-wise noise injection using random number generators with specific distribution parameters. For denoising, we employ two advanced filtering approaches: alpha-trimmed mean filtering (which removes extreme values from the neighborhood before averaging) and adaptive median filtering (dynamically adjusting window size to preserve details while removing impulses). The algorithms work by scanning through image pixels with sliding windows, applying statistical operations to replace noisy pixels. Finally, we conduct quantitative performance assessment using metrics like PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index) to objectively evaluate denoising effectiveness. The implementation includes comparative analysis of filter performance across different noise types and intensity levels.