单时相双极化ENVISAT ASAR数据水稻识别

许多研究已表明合成孔径雷达(SAR)对水稻识别及作物长势监测很有潜力。但是，以往的研究多是采用单极化多时相SAR数据进行水稻监测的。该文本着探讨多极化方式的优势以及降低数据购买成本和减少数据处理量的目的，对单时相双极化的ENVISAT ASAR APP数据的水稻识别能力进行了评价。在水稻生长季节，获取了覆盖江苏洪泽县的ASAR APP时间序列数据。首先，分析比较不同地物的后向散射系数，选择出最能区分水稻与非水稻的单时相数据；然后，采用决策阈值法将水稻信息从图像中提取出来；最后，利用DGPS实测的样地数据对水稻识别进行精度验证。结果表明，利用水稻齐穗期至近成熟期的HH和VV极化的ENVISAT ASAR APP图像能较好区分水稻与非水稻， 水稻识别精度可达86％以上。

Rice field mapping and monitoring using singe-temporal and dual polarized ENVISAT ASAR data

In this paper, the main purpose is to select the best single-temporal ENVISAT alternating polarization ASAR APP data for rice mapping. The authors investigated the backscatter coefficient of rice fields and other land covers using multi-temporal ASAR APP data with co-polarized(HH,VV) channels. Then the best single-temporal data were selected out for further analysis. Experiments were carried out to verify the effectiveness of single temperal and dual polarization ASAR APP data for extracting rice information in Hongze county, Jiangsu Province of China. Results indicate that the image acquired around September is the best for mapping rice area. The classification accuracy for rice and no-rice was assessed using DGPS samples and it was more than 86%. The study indicates that the whole procedure of rice mapping using single temporal ASAR APP data could become a low-cost and convenient operational system for rice area mapping.