A bio-inspired system for boundary detection in color natural scenes
ICCSA'12 Proceedings of the 12th international conference on Computational Science and Its Applications - Volume Part II
Hi-index | 0.01 |
This fully automated process uses edge map information to eliminate false boundaries in an image's region map, and region map information to remove noise in its edge map. It then integrates the two maps into a single, final result. Experiments on a large dataset of natural color images show that this approach matches human perception better than individual methods in terms of both quantity and quality.