基于HJ1A-HSI超光谱影像的耕地有机质遥感定量反演  

Remote Sensing Quantitative Retrieval of Farmland Organic Matter Based on HJ1A-HSI Image

作　　者：顾晓鹤[1] 王堃[1] 潘瑜春[1] 蔡玉梅[2] 黄文江[1] 王慧芳[1,3] 

机构地区：[1]北京农业信息技术研究中心,北京100097 [2]中国土地勘测规划院,北京100035 [3]浙江大学农业遥感与信息技术应用研究所,浙江杭州310029

出　　处：《地理与地理信息科学》2011年第6期69-73,共5页Geography and Geo-Information Science

基　　金：国家科技支撑计划项目(2008BAJ08B03)

摘　　要：以环境小卫星超光谱影像为主要数据源,在耕地有机质实测样本的支持下,开展光谱反射率及其变化形式与有机质的相关性分析,筛选耕地有机质响应的敏感波段与特征组合算法,利用多元回归分析方法,建立基于HSI影像的耕地有机质定量反演模型。研究结果表明,耕地有机质的HSI影像响应波段均位于可见光与近红外波段间,其中以540～860nm范围最佳,相关系数均在0.5左右;HSI反射率对数一阶微分预测模型精度最高,模型及其检验的决定系数都在0.7以上,均方根误差在0.2%左右,可用于顺义区耕地有机质全覆盖空间填图。因此,环境小卫星的超光谱影像对耕地有机质含量具有较好的光谱响应能力,其空间覆盖能力有助于开展县域尺度的耕地有机质遥感反演和空间填图。The research proposed a model to generate farmland organic matter distribution map using HJ1A-HSI images.The model was developed in the forms of quantitative retrieval model,considering correlation between organic matter and HSI reflectance variation,sensitive wavebands on organic matter and feature combing algorithm.The results suggested that：1） The most sensitive waveband on organic matter of HSI images are visible and near infrared spectrum,in which 540~860 nm bands is the best with a correlation coefficients around 0.5;2） The model using HSI reflectance in first order differential logarithm can get the highest accuracy,with the coefficients of determination R higher than 0.7 and the RMSE around 0.2%,which proves this model can be applied in Shunyi County.The sensitive response on farmland organic matter and the ability of larger coverage strongly support the HJ1A-HSI suitable for organic matter distribution mapping.

关 键 词：有机质 HSI 多元回归分析 空间填图 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]