Comparative survey of thinning algorithms

Mark, Vincze; Bence, Kovari [Kővári, Bence András (informatika), szerző] Automatizálási és Alkalmazott Informatikai Tanszék (BME / VIK)

Angol nyelvű Tudományos Konferenciaközlemény (Könyvrészlet)
    Thinning or skeletonization is a process for reducing foreground regions in a binary image to a skeletal remnant that largely preserves the extent and connectivity of the original region while throwing away most of the original foreground. Thinning is commonly used in digital image processing, pattern recognition, image analysis and not least, in signature verification. The goal of this paper is to introduce the most common thinning methodologies and propose a method to evaluate their performance, especially in the field of signature recognition. The proposed evaluation method is intended to be objective, therefore it takes into account various properties of a thinned skeleton and compares them to those of an ideal reference image. Fifteen different algorithms have been implemented and rated using this method, the results showed that different kinds of skeletonization techniques have different benefits and drawbacks, however none was found to give perfect results.
    Hivatkozás stílusok: IEEEACMAPAChicagoHarvardCSLMásolásNyomtatás
    2021-03-05 04:34