9 June 1986 Edge Detection By Differences Of Gaussians
Author Affiliations +
Proceedings Volume 0595, Computer Vision for Robots; (1986) https://doi.org/10.1117/12.952277
Event: 1985 International Technical Symposium/Europe, 1985, Cannes, France
The Differences of Gaussians (DOGs) are of fundamental importance in edge detection. They belong to the human vision system as shown by Enroth-Cugell and Robson [ENR66]. The zero-crossings of their outputs mark the loci of the intensity changes. The set of descriptions from different operator sizes forms the input for later visual processes, such as stereopsis and motion analysis. We show that DOGs uniformly converge to the Laplacian of a Gaussian (ΔG2,σ) when both the inhibitory and excitatory variables converge to σ. Spatial and spectral properties of DOGs and ΔGs are compared: width and height of their central positive regions, bandiwidths... Finally, DOGs' responses to some features such as ideal edge, right angle corner, general corner..., are presented and magnitudes of error on edge position are given.
© (1986) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ph. Marthon, Ph. Marthon, B. Thiesse, B. Thiesse, A. Bruel, A. Bruel, "Edge Detection By Differences Of Gaussians", Proc. SPIE 0595, Computer Vision for Robots, (9 June 1986); doi: 10.1117/12.952277; https://doi.org/10.1117/12.952277


Optical Robotic Vision System
Proceedings of SPIE (December 10 1985)
Fuzzy morphological filters
Proceedings of SPIE (October 31 1992)
Connectionist learning procedure for edge detector
Proceedings of SPIE (January 31 1992)

Back to Top