Adaptive Acceleration of MAP with Entropy Prior and Flux Conservation for Image Deblurring

In this paper we present an adaptive method for accelerating conventional Maximum a Posteriori (MAP) with Entropy prior (MAPE) method for restoration of an original image from its blurred and noisy version. MAPE method is nonlinear and its convergence is very slow. We present a new method to accelerate the MAPE algorithm by using an exponent on the correction ratio. In this method the exponent is computed adaptively in each iteration, using first-order derivatives of deblurred images in previous two iterations. The exponent obtained so in the proposed accelerated MAPE algorithm emphasizes speed at the beginning stages and stability at later stages. In the accelerated MAPE algorithm the nonnegativity is automatically ensured and also conservation of flux without additional computation. The proposed accelerated MAPE algorithm gives better results in terms of RMSE, SNR, moreover, it takes 46% lesser iterations than conventional MAPE.