Hilditch Thinning Algorithm and Row-by-Row Column-by-Column Phase Unwrapping Algorithm

In phase map computations, several key algorithms are frequently employed: the Hilditch thinning algorithm (hilditch.m), row-by-row column-by-column phase unwrapping (RC_unwrap.m), least squares phase unwrapping (least_unwrap.m), Fourier transform phase extraction (FTP.m), and sine-cosine filtering (sincosfilter.m). These algorithms play crucial roles in electronic speckle pattern interferometry, holographic interferometry, and phase-shifting techniques. The hilditch.m algorithm implements a thinning-based approach that detects and refines edges through iterative pixel removal while preserving connectivity, typically using 8-connected neighborhood analysis to improve image quality and precision. RC_unwrap.m performs phase unwrapping by first processing rows and then columns (or vice-versa), comparing adjacent pixels to resolve 2π ambiguities through simple difference calculations. least_unwrap.m employs least squares minimization to fit phase surfaces, often solving Poisson's equation using discrete cosine transforms to reduce integration errors. FTP.m extracts phase information by applying Fourier transform to interference patterns, filtering frequency domains, and performing inverse transforms to obtain phase maps. sincosfilter.m utilizes sine and cosine transformations for noise reduction, typically working in frequency domain to remove high-frequency noise while preserving phase information integrity, thereby enhancing image quality and measurement accuracy. These algorithms collectively provide comprehensive solutions for phase analysis, from edge detection and noise filtration to phase extraction and unwrapping, forming essential tools in optical measurement and interferometry applications.