Autowave principles for parallel image processing

Abstract A highly parallel autowave method for pattern analysis and topological feature detection is presented. It is invariant against translations, rotations and scaling of the input pattern. The method yields an increase in computational speed of 3 to 6 orders of magnitude in comparison with a sequential (von Neumann) computer. The method can be realized in principle using only one chip with simple uniform connections of elements.

[1]  Amlan Kundu.  Robust edge detection , 1990, Pattern Recognit..

[2]  V. I. Krinsky,et al.  Image processing using light-sensitive chemical waves , 1989, Nature.

[3]  Xinhua Zhuang,et al.  Image Analysis Using Mathematical Morphology , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.