耕地“非粮化”遥感监测与应用研究

Research on Remote Sensing Monitoring and Application of"Non Grain"Cultivated Land

传统的国耕地“非粮化”监测主要依靠人工实地核查或群众舆论举报等,这些手段费时费力,且错误信息较多,对监测工作帮助不大。本研究采用遥感分类与反演手段对地表覆盖类型进行变化分析与全要素分类提取,提取多时期耕地范围,并对比耕地变化区域,提取出耕地“非粮化”图斑。结果表明:遥感技术在耕地“非粮化”监测中具有很好的适用性,时效性上大幅优于人工实地核查;多时相分析可较好地提取出农作物种植区;遥感分类手段中的面向对象识别方法在耕地“非粮化”监测中的识别精度最高。

The traditional"non grain"monitoring of cultivated land in China mainly depends on manual field verification or public opinion reporting.This means is time-consuming and laborious,and there are many wrong information,which is not helpful to the monitoring work.In this study,remote sensing classification and inversion are used to analyze the change of surface coverage types and extract the total factor classification,extract the range of cultivated land in multiple periods,compare the change areas of cultivated land,and extract the"non grain"map spot of cultivated land.The results show that remote sensing technology has good applicability in the monitoring of"non grain"of cultivated land,and its timeliness is much better than manual field verification;multi-temporal analysis can better extract the crop planting area;the object-oriented recognition method of remote sensing classification has the highest recognition accuracy in the"non grain"monitoring of cultivated land.