| 雷达相干性与NDVI经验模型 |
| |
| 引用本文： | 刘智勇,张晨,刘泽楷,袁俊健,祁宏昌,潘屹峰,朱焕廉,吴希文,王华.雷达相干性与NDVI经验模型[J].测绘通报,2021,0(4):45-51,59. |
| |
| 作者姓名： | 刘智勇 张晨 刘泽楷 袁俊健 祁宏昌 潘屹峰 朱焕廉 吴希文 王华 |
| |
| 作者单位： | 1. 广东电网有限责任公司广州供电局, 广东 广州 510620;2. 广州地理研究所, 广东 广州 510075;3. 中科云图科技有限公司, 广东 广州 510075;4. 深圳建设工程质量监测中心, 广东 深圳 518052;5. 广东工业大学, 广东 广州 510006 |
| |
| 基金项目： | 中国南方电网有限责任公司广州供电局有限公司重点科技项目（080000KK52190001）；广州市科技计划（201902010033）；广东省自然科学基金（2018A030310538） |
| |
| 摘 要： | 植被的覆盖程度是造成雷达影像失相干的重要因素。通常，在森林等植被覆盖严重的地区，相干性相对较低，而在城市等植被覆盖率较低的地区，相干性较高。本文基于2017年珠江三角洲地区的MODIS归一化植被指数（NDVI）与Sentinel-1雷达卫星影像相干性，建立线性回归和幂函数回归模型，并利用两种模型预测该地区2016年的相干性，最后通过F分布检验及残差分布比较两者的精度。试验结果表明，总体上幂函数模型的精度在拟合及预测方面均高于线性函数模型，因此，本文建议以幂函模型作为相干性及NDVI的经验模型。利用幂函数经验模型不仅能够较好地预测2016年珠三角地区的相干性，同时在预测云南地区的相干性中也获得了较高的精度，因此证明该经验模型具有一定的推广价值。
|
| 关 键 词： | 归一化植被指数 相干性 InSAR 幂函数 线性函数 |
| 收稿时间： | 2020-09-30 |
Empirical relationship between radar coherence and NDVI |
| |
| LIU Zhiyong,ZHANG Chen,LIU Zekai,YUAN Junjian,QI Hongchang,PAN Yifeng,ZHU Huanlian,WU Xiwen,WANG Hua.Empirical relationship between radar coherence and NDVI[J].Bulletin of Surveying and Mapping,2021,0(4):45-51,59. |
| |
| Authors: | LIU Zhiyong ZHANG Chen LIU Zekai YUAN Junjian QI Hongchang PAN Yifeng ZHU Huanlian WU Xiwen WANG Hua |
| |
| Institution: | 1. Guangdong Power Grid Co., Ltd., Guangzhou Power Supply Bureau, Guangzhou 510620, China;2. Guangzhou Institute of Geography, Guangzhou 510075, China;3. ZhongkeYuntu Technology Co., Ltd., Guangzhou 510075, China;4. Shenzhen Construction Engineering Quality Monitoring Center, Shenzhen 518052, China;5. Guangdong University of Technology, Guangzhou 510006, China |
| |
| Abstract: | The degree of vegetation coverage is an important factor that causes the loss of coherence between radar images. Generally, in areas with severe vegetation coverage such as forests, coherence is relatively low, while in areas with low vegetation coverage such as cities, coherence is relatively high. We establish linear regression and power function regression models based on MODIS normalized vegetation index (NDVI) and InSAR in the Pearl River Delta region in 2017. We use these two models to predict the coherence in this region in 2016 and use F distribution to test the accuracy of the models. The results show that the accuracy of the power function regression model is higher than the linear function in both fitting and prediction. Therefore, we suggest the power function model as an optimal empirical model of coherence and NDVI. Our results show that the power model can well predict the coherence not only in the Pearl River Delta in 2016, but also can predict that in high precision in Yunnan. Therefore, we suggest the empirical model might be widely used in other regions. |
| |
| Keywords: | normalized vegetation index coherence InSAR power function linear function |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
|
