WorldView-2影像在南海岛礁浅海水深反演中的应用

发布时间：2018-03-07 19:00

本文选题：WorldView-　切入点：多光谱　出处：《遥感信息》2016年05期 　论文类型：期刊论文

【摘要】：针对南海岛礁浅海水域遥感探测研究较少且常用的中、低空间分辨率遥感影像难以满足该区域浅海水深高精度探测的问题,利用高空间分辨率WorldView-2多光谱影像和现场实测数据,对南海北岛岛礁浅海水域进行了水深反演研究。通过对WorldView-2和实测数据进行相关性分析,选择出相关性较高的数据构建了单因子、双因子、多因子线性回归模型和BP神经网络模型,利用实测数据对各个模型进行了测试与比较。结果显示,各个模型反演值与实测值的相关系数均在0.9以上,均方根误差约等于1.0,其中BP网络模型的相关系数最高,具有较高精度。研究表明,利用WorldView-2数据反演中国南海岛礁周边浅海水深的方法有效可行。
[Abstract]:In view of the low spatial resolution remote sensing image is difficult to meet the problem of high precision detection of shallow water depth in the South China Sea, the WorldView-2 multispectral image with high spatial resolution and the field measured data are used to solve the problem of the low spatial resolution remote sensing exploration in the shallow waters of the South China Sea. The inversion of water depth in the shallow waters of the North Island and Reef in the South China Sea is carried out. Through the correlation analysis between the WorldView-2 and the measured data, a single factor and a double factor are constructed by selecting the data with high correlation. The multifactor linear regression model and BP neural network model are tested and compared with the measured data. The results show that the correlation coefficient between the inversion value and the measured value of each model is more than 0.9. The root mean square error is about 1.0, among which the BP network model has the highest correlation coefficient and high accuracy. The research shows that the method of using WorldView-2 data to retrieve the shallow water depth around the islands and reefs in the South China Sea is effective and feasible.
【作者单位】：山东科技大学测绘科学与工程学院;海岛(礁)测绘技术国家测绘地理信息局重点实验室;
【基金】：测绘技术国家测绘地理信息局重点实验室开放基金(2012B05) 2015测绘地理信息公益性行业科研专项(201512034)
【分类号】：P715.7

【相似文献】