退化图像复原方法研究进展

飞行器和空间成像制导装备在大气中高速飞行时会受到湍流干扰,导致光学系统接收到的图像发生模糊降质、像素偏移、信噪比降低等问题,开展退化图像复原技术及方法研究就成为空间光学成像系统获得较高性能图像的重要途径。通过对退化图像复原技术研究进展的系统梳理和分析研究,本文首先介绍了图像退化模型,接着给出了退化图像复原方法的分类,然后比较系统地介绍了确定正则化图像复原方法、随机正则化图像复原方法、基于局部相似性的图像复原方法、基于示例学习的图像复原方法等几种新型的单幅退化图像复原方法,其后分析了视频复原的特征、介绍了新近的几种典型的视频图像复原方法,最后分析总结出了图像复原的难点所在。对于促进我国退化图像复原技术的研究和发展具有一定的参考价值。

Progress of degraded image restoration methods

Owing to the existence of atmosphere, the transmission of the light waves will be interfered by the atmospheric turbulence. It will make the image blurred, the pixels got deviated and the image signal to noise ratio decreased. Therefore, the research of degraded image restoration technology has become an important way to obtain high performance image in space optical imaging system. In this paper, the research progress of degraded image restoration technology is systematically combed and analyzed. Firstly, the image degradation model is given and the classification of traditional restoration method is summarized. Secondly, degraded image restoration methods are systematically surveyed, especially new restoration technology such as regularization methods, random regularization methods, methods based on local similarity and exemplary learning. Furthermore, the feature of video restoration is analyzed, and several new methods of video image restoration are presented. Finally, the difficulty of image restoration is summarized. This review has a certain reference value for promoting the research and development of degraded image restoration technology.