基于时空数据融合的江汉平原水稻种植信息提取

发布时间：2018-03-18 14:04

本文选题：江汉平原　切入点：遥感　出处：《长江流域资源与环境》2017年06期 　论文类型：期刊论文

【摘要】：及时、准确监测水稻种植面积,对区域粮食政策制定、粮食安全以及农业发展具有重要意义。然而我国南方地区水稻生长期内降水充沛的气候特点使得遥感影像"云污染"现象严重,为解决水稻种植信息遥感提取存在可用数据不足的问题,以江汉平原为例,利用时空数据融合模型(Spatial and Temporal Data Fusion Approach,STDFA)将Landsat 8 OLI与时序MODIS数据融合,重构出具有高时-空分辨率特征数据,然后采用面向对象的SVM分类方法对研究区内水稻种植信息进行提取,结果如下:融合后的红与近红外波段反射率与真实反射率的相关系数分别为0.84和0.81,研究区水稻提取精度为94.46%,Kappa系数为0.91。说明时空融合模型能够较好地重构出高时空分辨率数据,从而实现多云雨地区农作物种植信息遥感提取。
[Abstract]:Timely and accurate monitoring of rice planting areas, formulation of regional food policies, Food security and agricultural development are of great significance. However, the phenomenon of "cloud pollution" in remote sensing images is serious due to the abundant precipitation in the rice growing period in southern China. In order to solve the problem of the shortage of available data in remote sensing extraction of rice planting information, taking Jianghan Plain as an example, Landsat 8 OLI was fused with sequential MODIS data by using spatiotemporal data fusion model, Spatial and Temporal Data Fusion approach (STDFAA). The feature data with high temporal and spatial resolution were reconstructed, and then the information of rice cultivation in the study area was extracted by using the object-oriented SVM classification method. The results are as follows: the correlation coefficients between red and near infrared reflectance and true reflectivity are 0.84 and 0.81, respectively, and the extraction accuracy of rice is 94.46 and Kappa coefficient is 0.91.The results show that the spatio-temporal fusion model can reconstruct the high spatial and temporal resolution data. In order to realize the remote sensing extraction of crop planting information in cloudy and rainy areas.
【作者单位】：中国科学院测量与地球物理研究所;环境与灾害监测评估湖北省重点实验室;中国科学院大学;
【基金】：中国科学院科技服务网络计划(KFJ-STS-EDTP-009) 湖北省自然科学基金项目(2014CFB376)~~
【分类号】：S511;TP751

【相似文献】