针对大气湍流复杂背景红外退化图像的复原问题,构造了各向异性权重差分项,在此基础上将基于二阶差分的2范数极小作为空间相关性约束应用在复杂背景湍流退化图像相邻两帧点扩展函数的优化求解过程中,提出了针对复杂背景的湍流退化图像复原算法。在微机上进行了一系列复原和方法对比实验,实验结果表明本文方法有效,效果较好。
For the problem of the restoration of the infrared turbulence-degraded image with complex background, the items of some anisotropic weighted differences are constructed mainly in this paper, The minimization of 2-norm of the second-order differences to be functioned as a spatial correction constraint is incorporated in the estimation process of solving for the values of the point spread functions. A restoration algorithm for two neighboring frames of turbulence-degraded images with complex background is proposed. Series of experiment and comparing results show that the proposed algorithm is effective for the restoration of infrared turbulence-degraded image with complex background.