CAJ | 학술논문

选取清原县为研究区，基于多源遥感数据，形成一套不同尺度（以县-乡为经营单位）的森林植被提取方法。以土口子乡为例，应用ENVI EX4．8软件，采用面向对象的方法对乡内森林植被进行提取，然后基于此方法，采用协同反演TM森林植被面积的方法对清原县森林植被面积信息进行提取。结果表明：RapidEy影像分类精度达88．87％，Kappa系数为0．61；TM影像的分类精度达到92．41％，Kappa系数为0．89，具有较高的分类精度。所建立的森林植被提取方法可为林业部门对不同尺度森林植被类型面积进行监测提供参考。
선취청원현위연구구，기우다원요감수거，형성일투불동척도（이현-향위경영단위）적삼림식피제취방법。이토구자향위례，응용ENVI EX4．8연건，채용면향대상적방법대향내삼림식피진행제취，연후기우차방법，채용협동반연TM삼림식피면적적방법대청원현삼림식피면적신식진행제취。결과표명：RapidEy영상분류정도체88．87％，Kappa계수위0．61；TM영상적분류정도체도92．41％，Kappa계수위0．89，구유교고적분류정도。소건립적삼림식피제취방법가위임업부문대불동척도삼림식피류형면적진행감측제공삼고。
Based on the multi-source remote sensing data,this paper takes Qingyuan County as the study area,and the method of forest vegetation extraction is formed on different scales ( County-Township manage-ment unit) .Tukouzi Township is selected as an example and the method of Object-oriented classification is ap-plied,using ENVI EX4.8 software,and then based on this method,Collaborative inversion method is applied to TM image and the forest vegetation information of Qingyuan County is extracted.The results show that based on RapidEye image,the overall accuracy of the forest vegetation type information extraction is 88.87%and the Kappa coefficient is 0.61;the accuracy of the forest area information extracted from TM imagine is 92.41%and the Kappa coefficient is 0.89.This method has a higher classification accuracy and can provide a reference for remote sensing information extraction of forest vegetation types on different scales for forestry departments.