Blur identification using an adaptive ADALINE network

There are different techniques available for solving of the restoration problem including Fourier domain techniques, regularization methods, and so on. But without knowing at least approximate parameters of the blur, these filters show poor results. Fourier domain techniques seriously suffer from the additive noise and non-uniform motion. In this paper a new approach is proposed for blur parameters identification using an adaptive ADALINE network. The weights of the ADALINE network are taken as the estimation of the blur PSF. Simulation results for the non-uniform straight motion-blurred images demonstrate the identification and restoration is effective.