本平台为互联网非涉密平台,严禁处理、传输国家秘密或工作秘密

HJ-1影像中基于光谱特征的烤烟识别方法

Recognition of Flue-cured Tobacco Crop Based on Spectral Characteristics Extracted from HJ-1 Remote Sensing Images

  • 摘要: 为了监测农作物的种植面积、进而预测其产量,依据HJ-1(环境一号卫星A、B星,简称HJ-1A,1B)遥感资料分析了烤烟及其他地物的光谱反射特性,并分析比较归一化植被指数NDVI、差值植被指数NDVI和比值植被指数RVI对烤烟及其他地物的区分度,确定烤烟遥感信息识别方法。用该方法对广西靖西县烤烟进行种植信息提取,并用野外GPS实地调查采样数据进行精度验证。结果表明,总精度为92.79%,Kappa系数为0.8374。因此,利用遥感技术可快速、准确地获取烤烟种植信息。

     

    Abstract: In order to monitor crop planting area and predict its production,HJ-1 satellite(No.1 Environment satellite union of A and B) remote sensing data were used to analyze the spectral reflection characteristics of flue-cured tobacco crop and other ground objects.Three vegetation indices,including normalized difference vegetation index(NDVI),difference vegetation index(DVI) and ratio vegetation index(RVI),were calculated and their discrimination of flue-cured tobacco from other ground objects was analyzed and compared.The method for recognition of flue-cured tobacco from remote sensing data was determined,and the planting information of flue-cured tobacco in Jingxi County,Guangxi was extracted with the developed method.The method precision was verified with GPS field sampling data,and the obtained total precision was 92.79% with Kappa coefficient of 0.8374.It suggested that remote sensing technology could be used to obtain tobacco planting information quickly and accurately.

     

/

返回文章
返回