论文部分内容阅读
遥感影像中薄云的存在为影像的判读带来了极大的影响,通过研究薄云对Landsat影像造成的影响,提出一种加权梯度融合变分模型。通过在无云区域采用较小权重以保持影像自身信息,薄云区域则采用较大权重将参考影像的梯度信息融入待修复影像,改进了梯度模型在无云区域过度增强细节而造成的失真。采用暗通道法和梯度融合法与该方法进行比较,实验结果表明:该方法在有效去除薄云的同时对无云区域有较好的保真效果。
The existence of thin clouds in remote sensing images has a great impact on the interpretation of images. By studying the influence of thin clouds on Landsat images, a weighted gradient fusion variational model is proposed. By using less weight in the cloudless area to maintain the information of the image itself, the thin cloud area takes more weight to merge the gradient information of the reference image into the image to be repaired, and improves the distortion caused by over-enhancing the detail of the gradient model in the cloudless area. Compared with this method by the dark channel method and the gradient fusion method, the experimental results show that this method has a good fidelity to the cloudless area while effectively removing the thin cloud.