A MEASURE OF QUALITY FOR EVALUATION METHODS OF SEGMENTATION AND EDGE DETECTION

Abstract

A new measure of quality is proposed for evaluation the performance of available methods of image segmentation and edge detection. The technique is intended for the evaluation of low-error results and features objective assessment of discrepancy with the theoretical edge, in tandem with subjective visual evalutation using the neighbourhood and error-interaciton criteria. The proposed mathematical model is extremely simple, even from the perspective of computational execution. Encouraging results were obtained for a selection of test images, especially in relation to other recently-proposed and/or currently employed quality measures.