论文部分内容阅读
依据苏州东桥试验区水稻生长期观测数据,采用多时相水稻拔节到抽穗期全极化Radarsat-2数据,分别分析了不同极化HH、VV、CROSS、比值HH/VV的雷达后向散射系数时域变化特征与生物量的相关关系,构建水云模型、二次多项式模型和指数模型反演水稻生物量。反演结果表明:HH、CROSS水云模型都有不错的反演效果,相关系数分别为0.910、0.902,而HH水云模型反演生物量尤佳,均方根误差为0.190。指数模型普遍优于二次多项式模型,HH/VV指数模型效果出众,相关系数为0.929,均方根误差为0.164。通过比较分析不同极化的水云模型、二次多项式模型和指数模型,HH水云模型与HH/VV指数模型反演水稻生物量精度相对较高。
According to the observation data of rice growth period in Dongqiao experimental area of Suzhou, the radarsat-2 data from jointing stage to heading stage of multi-phase rice were used to analyze the radar backscatter coefficients of HH, VV, CROSS and HH / VV Time-domain variation characteristics and biomass correlation, building water cloud model, quadratic polynomial model and index model to retrieve rice biomass. The inversion results show that both HH and CROSS water cloud models have good inversion results, and the correlation coefficients are 0.910 and 0.902, respectively. However, the HH water cloud model is the best for retrieving biomass with a root mean square error of 0.190. The index model is generally superior to the quadratic polynomial model, and the HH / VV index model is superior, with a correlation coefficient of 0.929 and a root mean square error of 0.164. Through comparative analysis of water polarization model, quadratic polynomial model and exponential model, HH water cloud model and HH / VV index model of rice biomass inversion accuracy is relatively high.