A hybrid approach for image half-toning combining simulated annealing and neural networks based techniques: implementation on a zero instruction set computer based neural machine

A classes of stochastic algorithms, which are very powerful in the case of the degraded image reconstruction, are simulated annealing based algorithms. However, the reconstruction of a degraded image using iterative stochastic process require a large number of operations and is still out of real time. On the other hand, learning and generalization capability of ANN models allows a large panel of techniques improving classical techniques limitations. We are investigating in parallel implementation of image processing techniques. In this paper, we present a hybrid approach for image half-toning combining simulated annealing and neural network based techniques. Simulation and experimental results will be reported. Key-Words: Hybrid Technique, Simulated Annealing, Neural Networks, Degraded Image Reconstruction, ZISC-036, Hardware Implementation, Parallel, Image Processing, Half-Toning.