An improved nonnegativity and support constraints recursive inverse filtering (NAS-RIF) algorithm based on wavelet transform is presented to restore blind images.The degraded image is decomposed to obtain its wavelet coefficients in wavelet domain.The image′s different frequency sub-bands are also obtained.Then,NAS-RIF algorithm is used to restore degraded image in each sub-bands,different regularization terms are used in different sub-bands.By estimating the noise variance in each sub-bands,the adaptive regularization parameters can be calculated through the local properties of the observed image and the noise variance.The two simulating experiments are made and high signal to noise ratios (SNR) of 19.66 dB and 23.86 dB are obtained.The experimental results show that the method given by authors is more efficient than traditional space-adaptive regularization method.