基于卫星遥感的输电沿线建筑物变化检测

准确、快速的实现建筑物变化检测对识别输电沿线隐患区具有重要的意义。随着卫星遥感技术的不断发展和进步，基于高分辨率遥感影像进行建筑物变化检测已成为研究热点。本文提出了一种融合阴影、建筑物指数、纹理、形状等特征的高斯概率模型实现像元级的建筑物初提取，在此基础上，结合多尺度分割结果实现对象级的建筑物变化检测，最后，将前后两时相对象级的建筑物提取结果进行空间叠加分析，并将叠加结果划分为建筑物新增、建筑物拆迁和建筑物改建三类。本文在研究区和验证区分别采用不同分辨率的遥感影像进行实验和验证，结果表明，该方法在不同尺度和影像分辨率下的建筑物变化检测均取得了实际效果和检测精度，可为输电沿线建筑物变化检测提供良好的技术支持。

The Research on Building Change Detection Along Transmission Line Based on Satellite Remote Sensing

Accurate and rapid detection of building changes is of great significance to identify latent areas along transmission lines. With the development and progress of satellite remote sensing technology, building change detection based on high resolution remote sensing image has become a research hotspot. In this paper, a Gaussian probability model integrating features such as shadow, building index, texture and shape is proposed to realize initial extraction of buildings at the pixel level. Based on this, object-level change detection of buildings is realized by combining multi-scale segmentation results. Finally, The spatial superposition analysis is carried out on the extraction results of object-level buildings in the front and back phases, and the superposition results are divided into three categories: building addition, building demolition and building reconstruction. In this paper, remote sensing images with different resolutions are used in the research area and the verification area respectively for experiments and verification. The results show that the proposed method achieves practical results and detection accuracy under different scales and image resolutions, and can provide good technical support for the detection of building changes along the transmission line.