一种无参考运动模糊图像的质量评价方法

摘要

针对高动态环境下视觉导航系统的可靠性受到运动模糊制约，而现有的结构相似度算法（SSIM）对模糊图像评价不敏感的问题，提出了一种基于改进结构相似度的无参考运动模糊图像质量评价算法。首先，将原始运动模糊图像经过同运动方向的模糊算子生成再模糊图像；之后，将原始模糊图像与再模糊图像进行8×8分块，并在相应子块中找到边缘信息丰富的子块；然后，计算原始模糊图像和再模糊图像中对应边缘信息丰富子块的结构相似度；最后，通过边缘置信度对子块加权得到整幅图像的模糊评价指标。实验结果表明，此评价方法结果与主观评价结果具有较高的一致性，能够准确地对运动模糊图像进行评价。由于本方法能有效地优化边缘子块的权值，与传统的平均结构相似度（MSSIM）评价方法相比，具有更高的灵敏性。%In order to solve the problem that the reliability of visual navigation is restricted by motion blur under high dynamic environment,a novel non-reference motion-blurred image quality assessment algorithm based on improved structural similarity is proposed.Firstly,the re-blurred image is produced by the original blurred image with the same blurred direction.Then two images are divided into 8 ×8 blocks,and find the edge block,and then structural similarity of corresponding edge blocks in original blurred image and re-blurred image.Finally the blur estimation of the blurred image is obtained by the different weights of the edge block’s structural similarity index.The experimental results show that the proposed approach can obtain good performances,and the algorithm is more sensitive than traditional MSSIMalgorithm.