Efficacy of Gaussian smoothing in Canny edge detector

The authors examine the efficacy of the noise-filtering stage in the Canny edge detector using the receiver operating characteristic paradigm and find that omitting the filtering stage can give superior results. They conclude that the non-maximal suppression stage is mainly responsible for the success of the Canny detector.

[1]  Sudeep Sarkar,et al.  Robust Visual Method for Assessing the Relative Performance of Edge-Detection Algorithms , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Neil A. Thacker,et al.  Adaptive step edge model for self-consistent training of neural network for probabilistic edge labelling , 1996 .

[3]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Josef Kittler,et al.  Optimal Edge Detectors for Ramp Edges , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Rangachar Kasturi,et al.  Machine vision , 1995 .

[6]  Sean Dougherty,et al.  Edge detector evaluation using empirical ROC curves , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[7]  Owen Robert Mitchell,et al.  Precision Edge Contrast and Orientation Estimation , 1988, IEEE Trans. Pattern Anal. Mach. Intell..