论文部分内容阅读
针对传统遥感影像质量评价中云层覆盖量无法定量评价的问题,文章提出采用利用边缘信息的阈值分割结合数学形态学方法来提取遥感影像云层覆盖范围。利用边缘信息的阈值分割方法能够有效利用影像自身信息来改善分割结果,再结合形态学方法,进而能消除道路、房屋等大部分噪声信息,最终实现遥感影像上不同特征云层覆盖范围的自动提取。基于浙江全省高分辨率遥感影像的实验结果表明:该方法能够快速有效地识别出遥感影像上云层覆盖范围,研究结果对于遥感影像云层覆盖的自动评价具有参考价值。
In view of the problem that cloud cover can not be quantitatively evaluated in the traditional remote sensing image quality assessment, this paper proposes threshold extraction using edge information combined with mathematical morphology to extract cloud coverage of remote sensing images. The threshold segmentation method based on edge information can effectively use the image information to improve the segmentation results. Combined with the morphological method, most of the noise information such as roads and houses can be eliminated. Finally, the coverage of different characteristic clouds in remote sensing images can be automatically extracted. The experimental results based on the high resolution remote sensing images of Zhejiang province show that this method can quickly and effectively identify the coverage of cloud cover on remote sensing images and the research results have reference value for the automatic evaluation of cloud cover of remote sensing images.