Grayscale Image Enhancement Using SMQT (Successive Mean Quantization Transform) Algorithm

MATLAB implementation based on a 2005 ICIP conference paper, including result visualization. This code applies the SMQT (Successive Mean Quantization Transform) algorithm for grayscale image enhancement. The key advantage over traditional methods like histogram equalization is its ability to preserve the histogram waveform shape while revealing underlying image structures and automatically enhancing details through intelligent quantization levels. The algorithm recursively divides pixel values based on mean thresholds to maintain tonal relationships.

Total Variation Minimization: A Regularized Restoration Method for Image Detail Preservation

In image restoration techniques, Total Variation (TV) minimization serves as a regularization-based recovery method designed to preserve image details. This algorithm implements an alternating minimization strategy with precise iterations to simultaneously restore both the image and point spread function. Code implementation typically involves solving partial differential equations using gradient descent or primal-dual optimization methods. Experimental results demonstrate robust image restoration even under high-noise conditions, with key functions handling edge-preserving regularization and noise suppression.

MATLAB Median Filter Implementation

This MATLAB function implements median filtering, an image processing technique that effectively preserves image details while providing excellent salt-and-pepper noise removal capabilities.