Algorithm study of edge detection in profiled fibers cross-section measurement

Edge detection is a basic content of digital image processing,and is the base of image segmentation,feature extraction and image understanding.The fiber characteristic parameters are the principal indexes of showing cross-sectional shape of profiled fiber.The characteristic parameters of profiled fibers can be acquired by computer digital image processing system and the type of profiled fibers will be distinguished.So the performances of several typical edge detection algorithms in pentagonal shape of profiled fiber are discussed.Meanwhile,the advantages and disadvantages of these algorithms are demonstrated in the experiments and some conclusions are drawn.This study will provide some basis for the image processing of automatic detection system for profiled fiber cross-section,which is being developed.