基于SHALSTAB模型与面向对象遥感影像分析的地震滑坡信息快速检测*
王晓波1,2,鲁 恒2,3,4,刘雪梅5,杨正丽3,4,项 霞3,4,蔡诗响3,4

(1.青海省基础地理信息中心,青海 西宁 810001; 2.青海省地理空间信息技术与应用重点实验室,青海 西宁 810001; 3.四川大学 水力学与山区河流开发保护国家重点实验室,四川 成都 610065; 4.四川大学 水利水电学院,四川 成都 610065; 5.四川省地震局,四川 成都 610041)

坡度稳定性模型; 面向对象; 高分遥感影像; 滑坡; 快速检测

Rapid Detection of Seismic Landslide Information Based on SHALSTAB Model and Object-oriented Remote Sensing Image
WANG Xiaobo1,2,LU Heng2,3,4,LIU Xuemei5,YANG Zhengli3,4,XIANG Xia3,4,CAI Shixiang3,4

(1. Provincial Geomatics Center of Qinghai,Xining 810001,Qinghai,China)(2. Geomatics Technology and Application Key Laboratory of Qinghai Province,Xining 810001,Qinghai,China)(3. State Key Laboratory of Hydraulics and Mountain River Engineering,Sichuan University,Chengdu 610065,Sichuan,China)(4. College of Hydraulic and Hydroelectric Engineering,Sichuan University,Chengdu 610065,Sichuan,China)(5. Sichuan Earthquake Agency,Chengdu 610041,Sichuan,China)

slope stability model; object oriented; high-resolution remote sensing image; landslide; rapid detection

备注

滑坡是最为常见的地震次生灾害之一,对其进行有效监测一直都是业界研究的热点。基于此,提出了一种高分遥感影像地震滑坡信息快速检测方法,该方法将SHALSTAB模型与面向对象影像分析相结合,首先对遥感影像进行多尺度分割,并根据稳定性模型赋权,然后根据深度学习机制对滑坡对象进行检测,最后对检测结果进行过滤,并将该方法应用于2013年芦山地震滑坡检测,与目视解译结果进行对比。结果 表明:该方法能快速检测高分遥感影像上滑坡,滑坡检测正确率达85%以上。

Landslide is one of the most common geological disasters caused by earthquakes. How to quickly and effectively monitor landslide has always been a research hotspot. Based on this,we proposed a seismic landslide information fast detection method based on high resolution remote sensing image. This method is combining SHALSTAB(Shallow Land sliding Stability)model with object-oriented image analysis. Firstly,the multi-scale segmentation of remote sensing images is carried out,and the weights are assigned according to the SHALSTAB model. Then the landslide objects are detected according to the deep learning mechanism. Finally,the detection results are filtered,and the method is applied to the landslide detection of Lushan M7.0 earthquake in 2013 and compare with the visual interpretation results. The results show that the proposed method can detect landslide in high-resolution remote sensing image rapidly,and the detection accuracy of landslide is over 85%.