Abstract: A self-adaptive image restoration algorithm based on image quality assessment is put forward to suppress the ringing effects which occur in blind image restoration.Firstly,we propose an image quality assessment based on gradient,which can assess both image definition and ringing effects.Then,we utilize the assessment to estimate PSF (Pointer Spread Function) size and treat the estimated PSF as the initial PSF of the iterative blind deconvolution (IBD) algorithm.Finally,we restore the blurred image iteratively and assess the quality of the restored image at the end of each iteration,the iteration stops as soon as the assessment becomes decreasing,the final restored image is the one with the highest assessment metric.The results of the simulated experiments show that the restored image quality assessment method in this paper is consistent with human perception,the estimated PSF size is accurate,the restored image quality is high and the ringing effects is restrained.