基于遥感的南方丘陵山区耕地撂荒研究  

Study on arable land in the southern hilly mountains based on remote sensing

作　　者：吴峰 李沛鸿[1] 熊凡 袁逸敏 WU Feng;LI Peihong;XIONG Fan;YUAN Yimin(School of Civil and Surveying and Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou Jiangxi 341099,China)

机构地区：[1]江西理工大学土木与测绘工程学院,江西赣州341099

出　　处：《北京测绘》2022年第11期1519-1523,共5页Beijing Surveying and Mapping

基　　金：国家级大学生创新创业训练计划(202010407010)。

摘　　要：针对我国南方丘陵山区耕地撂荒愈加严峻的现象,为准确把握耕地的撂荒特征和影响因子,快速有效了解撂荒空间规模和数量特征情况,本文基于Landsat-8数据,用支持向量机(SVM)算法对江西省安福县进行土地利用分类,依据耕地撂荒和复垦判别规则提取撂荒和复垦信息。研究结果表明:遥感影像的总体分类精度在84.73%~89.93%之间,符合研究所需精度要求;研究时段内撂荒情况呈先增后减趋势,持续撂荒4年以内的撂荒地较多,持续撂荒面积和撂荒率在逐年递减;劳动力析出是耕地撂荒的主导因素;遥感技术可有效运用于耕地撂荒状况的监测。In view of the increasingly severe phenomenon of arable land abandonment in the hilly mountainous areas of southern China,in order to accurately grasp the characteristics and impact factors of cultivated land,and quickly and effectively understand the spatial scale and quantitative characteristics of farmland reclamation,based on Landsat-8 data,the land use of support vector machine(SVM)algorithm was applied to classify land use in Anfu county,Jiangxi province,and the information on desertification and reclamation was extracted according to the rules of arable land reclamation and reclamation.The results showed that the overall classification accuracy of remote sensing images was between 84.73%and 89.93%,which met the accuracy requirements required for the study.During the study period,the situation of famine control showed a trend of increasing first and then decreasing,and there were more wastelands that continued to be abandoned within 4 years,and the area and rate of continuous famine were decreasing year by year;labor force precipitation was the leading factor in arable land abandonment;remote sensing technology could be effectively applied to the monitoring of arable land desolation.

关 键 词：土地分类 支持向量机(SVM) 耕地撂荒 耕地复垦 

分 类 号：P237[天文地球—摄影测量与遥感]