Based on original Nonnegativity and Support constraint Recursive Inverse Filtering(NAS-RIF) algorithm,an improved blind image restoration algorithm was proposed.In order to achieve the trade-off between the fidelity of observed image and the smoothness of restored image and to prevent noise amplification,a new cost function of the NAS-RIF algorithm was introduced by adding space-adaptive terms and a regularization term.The space-adaptive terms could be calculated through the local properties of the observed image and the the noise variance.A algorithm using conjugate-gradient routine to calculate the optimal result was proposed to estimate the noise variance in unknown transcendental condition.The improved algorithm and original algorithm were both used to restore three different degraded images.The experimental results show that the ΔSNRs by proposed algorithm are increased by 0.207 3 dB,1.023 9 dB,2.862 8 dB as compared with those of original NAS-RIF algorithm.It demonstrates that the improved algorithm is more efficient.