ON THE LOCALIZATION PERFORMANCE-MEASURE AND OPTIMAL EDGE-DETECTION

In a recent paper, Tagare and deFigueiredo present a localization perf ormance measure for edge detectors. They correctly point out a flaw in ,Canny's formulation of the localization criterion, which was subseque ntly adopted by Sarkar and Boyer. They motivate their form of the loca lization criterion along a different line of reasoning. In this commen t, ae show that although Canny's derivation was in error, the final fo rm of his criterion is adequate and can, in fact, be derived from Taga re and deFigueiredo's formulation of the problem. We also point out so me problems with Tagare and deFigueiredo's localization criterion.