基于TV正则化和局部约束的遥感图像恢复(英文)

Remote Sensing Image Restoration Based on TV Regularization and Local Constraints

阐述了基于总变分理论和基于像元亮度局部约束的退化图像恢复算法,为利用二者的优点获得更好的恢复效果.把总变分方法和局部约束方法结合在一起,提出了一种新的混合恢复算法.对最小二乘问题进行总变分正则化约束,形成迭代公式,在迭代过程中对所获得的结果利用局部均值和局部方差进行局部约束.实验中对退化的遥感图像分别用总变分约束方法恢复和本文提出的方法进行恢复,结果表明,该方法具有良好的图像恢复能力,图像恢复效果有了明显的提高.

Image restoration algorithms based on total variation theory and pixels intensity local constrains method were introduced. To obtain a better restored effect, a novel hybrid algorithm was presented to estimate the original image from the degraded one. Total variation （TV） regularization term was added into the object function to form the iteration formula. And, local mean and local variance were used to constraint the solution. In numerical experiment, the degraded images were restored using the TV method and the proposed method. The results show that better performance of the proposed approach can be obtained.