MATLAB Implementation of Canny Edge Detection with Otsu-based Adaptive Thresholding

This MATLAB implementation demonstrates Canny edge detection using adaptive threshold selection based on Otsu's algorithm. The method automatically determines optimal threshold values, significantly improving edge detection accuracy and performance. Adaptive thresholding operates by analyzing the statistical distribution of pixel intensity values in the image, calculating the optimal separation point between foreground and background classes. This approach excels in handling images with varying lighting conditions and effectively extracts edge information through a multi-stage process: Gaussian filtering for noise reduction, gradient magnitude calculation using Sobel operators, non-maximum suppression for edge thinning, and dual-threshold hysteresis for edge linking. The implementation leverages MATLAB's built-in functions such as graythresh for Otsu threshold computation and edge with 'Canny' method, while demonstrating custom code for gradient calculation and hysteresis thresholding. This combination provides a robust solution for reliable edge detection across diverse image types and lighting scenarios.