Abstract

An analogy can be established between image processing and statistical mechanics. Many early- and intermediate-vision tasks such as restoration, image segmentation, and motion detection can be formulated as optimization problems that consist in finding the ground states of an energy function. This approach yields excellent results, but, once it is implemented in conventional sequential workstations, the computational loads are too extensive for practical purposes, and even fast suboptimal optimization approaches are not sufficient. We elaborate on dedicated massively-parallel integrated circuits, called stochastic artificial retinas, that minimize the energy function at a video rate. We consider several components of these artificial retinas: stochastic algorithms for restoration tasks in the presence of discontinuities, dedicated optoelectronic hardware to implement thermal motion by photodetection of speckles, and hybrid architectures that combine optoelectronic, asynchronous-analog, and clocked-digital circuits.

© 2001 Optical Society of America

Dedicated optoelectronic stochastic parallel processor for real-time image processing: motion-detection demonstration and design of a hybrid complementary-metal-oxide semiconductor– self-electro-optic-device-based prototype

Alvaro Cassinelli, Pierre Chavel, and Marc P. Y. Desmulliez
Appl. Opt. 40(35) 6479-6491 (2001)

Optoelectronic retinas that perform stochastic global optimization

Ph. Lalanne, P. Chavel, P. Garda, and F. Devos
Opt. Lett. 18(18) 1564-1566 (1993)

Optoelectronic analogs of self-programming neural nets: architecture and methodologies for implementing fast stochastic learning by simulated annealing

Nabil H. Farhat
Appl. Opt. 26(23) 5093-5103 (1987)

Implementation of optical and optoelectronic circuits

Navjot Singh and Harpreet Singh
Appl. Opt. 28(1) 178-181 (1989)

Hough transform algorithm for real-time pattern recognition using an artificial retina camera

Xin Lin and Kazunori Otobe
Opt. Express 8(9) 503-508 (2001)