针对高分辨率遥感影像暗背景下亮居民地的提取问题,该文提出了一种基于形态学多轮廓结构元素假高帽变换的提取方法。该方法首先采用腐蚀型边缘检测算子对影像进行预处理,接着利用多轮廓结构元素分别对影像进行假高帽变换,然后对各个处理结果进行加权融合,最后对加权融合后图像区域生长完成轮廓跟踪。实验结果表明,该方法能够提取居民地,特别适合于暗背景下较亮区域居民地的提取,并且提取效率较高,对于遥感影像居民地提取的研究与应用具有一定的参考价值。 In order to solve the problem of bright residential areas under the dark background of high resolution remote sensing images, this paper proposes a method based on morphological multi-contour structural elements pseudotyped cap transformation. Firstly, the image is preprocessed by using the edge detection operator. Then, the image is transformed by using the multi-contour structure elements. Then, the fusion results are weighted. Finally, track. The experimental results show that this method can extract residential areas and is especially suitable for the extraction of residential areas in bright areas with dark background, and has high extraction efficiency, which is of certain reference value for the research and application of remote sensing image extraction.