The role of non-CRF inhibition in contour detection

Abstract

We propose a biologically motivated computational step, called non-classical receptive field (non-CRF) inhibition, to improve the performance of contour detectors. Non-CRF inhibition is exhibited by 80% of the orientation selective neurons in the primary visual cortex of macaque monkeys and has been demonstrated to influence the visual perception of man as well. We introduce an image processing operator, the bar cell operator, which consists of a Gabor energy operator augmented with non-CRF inhibition. This operator responds strongly to isolated lines, edges and contours, but exhibits a weaker or no response to edges that make part of texture. We evaluate the contour detection performance of the proposed operator for images of natural scenes with associated ground truth edge maps. The bar cell operator consistently outperforms the Canny edge detector, mostly due to a reduced number of false positives.