为了对单帧气动光学效应退化图像进行有效恢复,对非负和支持域限制递归逆滤波(NAS-RIF)复原算法进行有效地改进,提出一种保边缘的递归逆滤波盲目复原算法。为了保存图像的细节信息,在代价函数中引入一个保边缘规整化惩罚泛函,采用线性快速搜索方法来寻找最优解。算法不需要图像支持域的先验信息,能够复原具有较复杂背景的目标图像。
To restore aero - optic effect degraded images, an improved recursive inverse filtering algorithm for blind restoration is presented. A penalty term is added into the cost function to preserve the edges of image effectively and then the problem is transformed into minimization on the cost function by linear search method. There is no need of a priori information on image support and the image with complex background can be restored well. Experimental results illustrate the behavior of this algorithm.