论文部分内容阅读
提出了一种利用SPOT5影像自动提取崩塌信息的改进方法。对MSAVI指数法进行了改进,使改进后的方法能更好地去除植被信息和突出裸地信息。首先利用改进的CMSAVI指数法去除SPOT5影像中的植被信息,进而对提取的结果进行去阴影、坡度筛选、形态学滤波、栅矢转换、面积和顺坡性筛选,并基于改进的多峰直方图阈值自动选取算法,实现了崩塌信息的自动提取。实验表明,该方法既能有效去除非崩塌等干扰信息,又真正实现了崩塌信息的自动提取,从而极大地提高了崩塌的识别、提取效率和精度。
An improved method of automatically extracting collapse information using SPOT5 images is proposed. The MSAVI index method has been improved so that the improved method can better remove the vegetation information and highlight the bare land information. Firstly, the vegetation information in SPOT5 image was removed by the improved CMSAVI index method, and then the results of extraction were screened out, shaded, morphological filtered, converted to grids, area and sequence. Based on the improved multi-peak histogram threshold Automatic selection of algorithms to achieve the automatic extraction of collapse information. Experiments show that this method not only can effectively remove the non-collapse interference information, but also realizes the automatic extraction of collapse information, which greatly improves the identification, extraction efficiency and accuracy of collapse.