Title :
Edge detector evaluation using empirical ROC curves
Author :
Bowyer, Kevin ; Kranenburg, Christine ; Dougherty, Sean
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of South Florida, Tampa, FL, USA
Abstract :
A method is demonstrated to evaluate edge detector performance using receiver operating characteristic curves. It involves matching edges to manually specified ground truth to count true positive and false positive detections. Edge detector parameter settings are trained and tested on different images, and aggregate test ROC curves presented for two sets of 10 images. The performance of eight different edge detectors is compared. The Canny and Heitger detectors provide the best performance
Keywords :
computational geometry; computer vision; edge detection; edge detector evaluation; empirical ROC curves; parameter settings; receiver operating characteristic curves; Computer science; Computer vision; Detectors; Humans; Hysteresis; Image edge detection; Image sampling; Materials testing; Packaging; Web pages;
Conference_Titel :
Computer Vision and Pattern Recognition, 1999. IEEE Computer Society Conference on.
Conference_Location :
Fort Collins, CO
Print_ISBN :
0-7695-0149-4
DOI :
10.1109/CVPR.1999.786963
