基于带间预测非负支撑域受限递归逆滤波盲复原算法.docVIP

基于带间预测非负支撑域受限递归逆滤波盲复原算法 　　摘要：针对非负支撑域受限递归逆滤波（NASRIF）算法对噪声敏感和耗时长等缺点，提出了一种改进的NASRIF盲复原算法。首先，为了改进原始NASRIF算法的抗噪性能和复原效果，引入了一种新的NASRIF算法代价函数；其次，为了提高算法的运算效率，结合Haar小波变换，仅对低频子频带的图像进行NASRIF算法复原，而高频子频带的信息，则通过带间预测分别从低频子频带的复原图像中预测得到；最后，为了保证高频信息的准确性，提出了一种基于最小均方误差（MMSE）的带间预测。分别对模拟退化图像和真实图像进行了仿真实验，采用该算法得到的信噪比增益分别为5.2216dB和8.1039dB。实验结果表明：该算法在保持图像边缘细节的前提下，能够较好地抑制噪声；此外，该算法的运算效率也得到了较大的提高。 　　关键词：图像盲复原；非负支撑域受限递归逆滤波算法；Haar小波变换；带间预测 　　中图分类号：TP391.413 　　文献标志码：A 　　Abstract：To overcome the shortcoming that the Nonnegativity And Support constraint Recursive Inverse Filtering （NASRIF） algorithm is noisesensitive and timeconsuming， an improved NASRIF algorithm for blind restoration was proposed. Firstly， a new cost function of the NASRIF algorithm was introduced， and then the noise resistance ability and the restoration effect were both improved. Secondly， in order to enhance computational efficiency of the algorithm， after decomposed by Haar wavelet transform， only degraded image in low frequency subbands was restored with the NASRIF algorithm， while information in high frequency subbands was predicted from the restored image of low frequency subbands by interband prediction. Finally， an interband prediction based on Minimum Mean Square Error （MMSE） was presented to guarantee the accuracy of the predicted information in high frequency subbands. The experiments on synthetic degraded images and real images were performed， and the SignaltoNoise Ratio （SNR） gain by proposed algorithm were 5.2216dB and 8.1039dB respectively. The experimental results demonstrate that the proposed algorithm not only preserves image edges， but also has good performance in noise suppression. In addition， the computational efficiency of the proposed algorithm is greatly enhanced. 　　英文关键词Key words：blind image restoration； Nonnegativity And Support constraint Recursive Inverse Filtering （NASRIF） algorithm； Haar wavelet transform； interband prediction 　　0 引言 　　在众多图像处理的应用领域中，如视频监控、医学诊断、生物鉴别、太空探索等，图像处理系统往往需要获取高质量的清晰图