Abstract: By introducing the monotone smooth logarithm and compounding it with the original cost function, we propose an improved nonnegativity and support constraints recursive inverse filtering algorithm (NAS-RIF) for image restoration. It has been deduced that the search direction remains the same but that the step length decreases in optimizing the improved cost function with the conjugate gradient algorithm from the gradient of the cost function that the FIR filter apporaches the contrary point spread function (PSF) more, and that the estimated image is closer to the original one. The experiment on a blurred text image restoration shows that the mean square error(MSE) of the improved NAS-RIF algorithm has a better convergence, and that the signal to noise ratio is improved.

Key words: nonnegativity and support constraints recursive inverse filtering algorigthm, blind image deconvolution, logarithm, gradient