湍流退化红外图像校正算法

IR image correction algorithm for turbulence-degraded

针对高速湍流场引起的红外图像模糊问题,提出了一种基于改进增量Wiener滤波的复原校正算法。首先,基于先验知识对湍流退化图像的降晰函数进行辨识并得到复原图像的起始估计;其次,提取起始复原图像中的强边缘并平滑边缘区域;最后,利用改进的增量维纳滤波算法迭代复原图像。实验结果表明:该算法与传统的迭代盲复原算法及基于Fuzzy滤波器的后期去振铃算法相比,复原图像的振铃测度有较大下降,同时提高了复原图像的质量,降低了算法的时间复杂度。

In order to deblur the fuzzy infrared image caused by high﹣speed turbulent flow field, a novel correction algorithm based on the modified incremental Wiener filter was proposed. Firstly, the degradation process was simplified as parameter﹣describing 2- D Gaussian function according to the prior knowledge, and the Point Spread Function（PSF） was estimated via an image quality assessment based blur identifier, which also can provide the initial restored image estimation. Then, strong edges in the initial restored image were detected and their corresponding area was smoothed to get the Edge Smoothed（ES） initial restored image. Finally, the ES image was restored through the modified incremental Wiener filter using the estimated PSF. Experimental results show that this algorithm can effectively suppress ringing artifacts, reduce the ringing metric of the restored image compared with those of the traditional iterative restoration algorithm and the Fuzzy filter based algorithm, and better the recovery image quality evidently, significantly reduce the time complexity.