Statistical techniques for edge detection in histological images

摘要: A review of the statistical techniques available for performing edge detection on histological images is presented. The tests under include Student's T Test, Fisher test, the Chi Square Kolmogorov Smirnov and Mann Whitney U test. All utilize a novel two sample edge detector to compare properties image regions surrounding central pixel. performance statistical tests compared using biomedical which traditional gradient based are not as successful, therefore giving an overall methods, results. Comparisons also made to more traditional Canny and Sobel, filters. results show that in the presence noise clutter both parametric non-parametric tests compare well robustly extracting information series images.