基于遥感技术的滑坡识别及滑坡检测系统开发  

Development of Landslide Identification and Detection System Based on Remote Sensing Technology

作　　者：魏永杰 WEI Yongjie(Xinjiang Railway Survey and Design Institute Co.,Ltd.,Urumqi 830011,China)

机构地区：[1]新疆铁道勘察设计院有限公司,乌鲁木齐830011

出　　处：《自动化与仪器仪表》2023年第8期37-43,共7页Automation & Instrumentation

摘　　要：遥感技术及图像处理技术的快速发展提高了滑坡自动化解译的能力。由于能够综合考虑遥感影像中的光谱、纹理、形态和地形特性,近年来面向对象的图像分析技术(Object-Based Image Analysis,OBIA)在滑坡自动化解译中得到广泛利用。当前许多相关研究都是基于已有商业软件开展的,为降低滑坡检测技术的使用门槛,在本研究中利用OBIA和机器学习创建了一个基于开源Python包的半自动滑坡检测系统。基于该系统对精伊霍铁路沿线约330 km2的面积开展滑坡检测,将检测结果与人工野外调查结果相对比并进行定量精度评价,结果表明kappa系数为0.803,该系统在滑坡检测中具有较高的精度,可以适用于大范围快速滑坡提取。The rapid development of remote sensing technology and image processing technology has improved the ability of automatic interpretation of landslide.In recent years,Object-Based Image Analysis(OBIA)has been widely used in landslide automatic interpretation because it can comprehensively consider the spectral,texture,morphological and topographic characteristics of remote sensing images.At present,many related researches are carried out based on existing commercial software.In order to reduce the use threshold of landslide detection technology,in this study,we use OBIA and machine learning to create a semi-automatic landslide detection system based on open-source Python package.Based on this system,about 330km along JINGYIHUO Railway.The results show that the kappa coefficient is 0.803.The system has high accuracy in landslide detection and can be applied to large-scale rapid landslide extraction.

关 键 词：滑坡 遥感技术 OBIA 机器学习 精伊霍铁路 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]