Title :
Anisotropic edge detection using mean field annealing
Author :
Acton, Scott T. ; Bovik, Alan C.
Author_Institution :
Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
Abstract :
An edge detection technique that optimizes edge localization while providing edge continuity and edge thinning is introduced. The solution is obtained by annealing a mean field neural network, providing inexpensive solutions with high parameter insensitivity. Anisotropic diffusion is used to provide localized edge data through the scale-space. Analysis of network parameters, diffusion parameters, network convergence, and scale-space equivalence is provided. Results are shown for real image data and compared with the results of other important edge detection schemes
Keywords :
edge detection; neural nets; anisotropic diffusion; anisotropic edge detection; diffusion parameters; edge continuity; edge localization; edge thinning; localized edge data; mean field annealing; mean field neural network; network convergence; network parameters; parameter insensitivity; real image data; scale-space; Anisotropic magnetoresistance; Annealing; Detectors; Equations; Gaussian processes; Image edge detection; Joining processes; Laboratories; Machine vision; Smoothing methods;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0532-9
DOI :
10.1109/ICASSP.1992.226037
