论文部分内容阅读
准确、快速提取大范围地表土壤盐渍化空间分布是一个迫切需要解决的科学难题。本文以河套灌区解放闸灌域土壤盐分雷达监测为例,研究基于RADARSAT-2数据的盐渍化信息提取,利用成熟的BP神经网络技术,建立四极化雷达影像灰度值反演土壤盐分的人工智能模型,经实测数据检验能够在一定程度上满足盐渍化监测的需要,优于传统盐渍土分类方法,可促进微波遥感在土壤盐渍化监测中的开拓应用。
Accurate and rapid extraction of soil salinization spatial distribution over a large area is an urgent scientific problem to be solved. Taking the monitoring of soil salinity radar in the watershed of Jiefang Gate in Hetao Irrigation District as an example, the salinization information extraction based on RADARSAT-2 data was studied. By using the mature BP neural network technology, the gray value of quadrupole radar image was used to retrieve the soil salinity The artificial intelligence model, verified by measured data, can meet the needs of salinization monitoring to a certain extent, which is better than the traditional saline soil classification method, which can promote the development and application of microwave remote sensing in soil salinization monitoring.