无人机可见光光谱的植被覆盖度估算新方法  

A new estimation method for fractional vegetation cover based on UAV visual light spectrum

作　　者：谢兵[1] 杨武年[2] 王芳[2,3] XIE Bing;YANG Wunian;WANG Fang(Sichuan College of Architectural Technology,Deyang,Sichuan 618000,China;Chengdu University of Technology,Chengdu 610059,China;Neijiang Normal University,Neijiang,Sichuan 641000,China)

机构地区：[1]四川建筑职业技术学院,四川德阳618000 [2]成都理工大学,成都610059 [3]内江师范学院,四川内江641000

出　　处：《测绘科学》2020年第9期72-77,共6页Science of Surveying and Mapping

基　　金：国家自然科学基金项目(41671432);四川省教育厅科研项目(15ZB0451);德阳市2017年第二批应用技术研究与开发项目(2017ZZ064);德阳市科学技术和知识产权局科研项目(2015ZZ047);四川建筑职业技术教育学院项目(2015KJ10)。

摘　　要：针对常见可见光植被指数提取植被覆盖度存在的问题,该文提出一种新的可见光植被指数——红绿蓝比值植被指数,该方法直接采用红绿蓝像元值计算,方法简单快速,探索建立了新植被指数与植被覆盖度之间的相关关系,并利用相邻乡镇航片以两种方法验证了新植被指数的可靠性。结果表明:采用红绿蓝比值植被指数,通过低空无人机获取多张航片拼接成大范围航片提取植被覆盖度信息总体精度达到93.5%,为小型无人机小面积研究区域高精度植被覆盖度快速获取提供了方法,对林业调查、农业监测有着重要作用。Aiming at the problem of extracting vegetation coverage from common visible vegetation index,this paper proposed a new visible vegetation index-red,green and blue ratio vegetation index(RGBRI),which was directly calculated by using red,green and blue pixel values.The method was simple and fast.The relationship between the new vegetation index and vegetation coverage was explored,and the reliability of the new vegetation index was verified by two methods using adjacent township aerial photographs.The results showed that the overall accuracy of extracting vegetation coverage information from large-scale aerial photographs by large-scale UAV was 93.5%,which was used for low-altitude drones to obtain high-precision vegetation.The rapid acquisition of coverage provided a method that played an important role in forestry surveys and agricultural monitoring.

关 键 词：无人机 光谱特性 植被指数 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]