论文部分内容阅读
土壤盐分的定量遥感反演,为快速、准确、全面地监测盐渍化状况提供了可能。本文以黄河三角洲地区垦利县为例,实地调查采集土壤样本,并获取同时相Landsat 8影像,建立土壤盐分遥感反演的BP神经网络、偏最小二乘回归、主成分分析、多元线性回归多种模型,进而进行精度对比分析,评价、优选最佳建模方法,最后,基于最佳模型进行研究区土壤盐分的空间分布反演分析。结果显示:遥感影像的反射率与土壤盐分含量并不是单纯的线性关系,构建的盐分估测模型BP神经网络预测决定系数为0.8467,均方根误差为0.071,明显高于传统线性统计模型,能较好地模拟土壤盐分与光谱数据的关系。该研究既能为盐渍土的治理、利用提供数据支持,又能推动盐渍化区域遥感研究的定量发展。
Quantitative remote sensing of soil salinity provides the possibility of monitoring salinization status quickly, accurately and comprehensively. In this paper, Kenli County in the Yellow River Delta as an example, the field survey to collect soil samples and access to simultaneous Landsat 8 images, the establishment of soil salinity remote sensing BP neural network regression, partial least-squares regression, principal component analysis, multiple linear regression Finally, based on the best model, the spatial distribution inversion of soil salinity in the study area was analyzed. The results show that the reflectance of remote sensing image is not purely linear with the soil salinity content. The predicted BP neural network prediction coefficient is 0.8467, the root mean square error is 0.071, which is significantly higher than that of the traditional linear statistical model Better simulate the relationship between soil salinity and spectral data. This research can not only provide data support for the treatment and utilization of saline soil, but also promote the quantitative development of remote sensing research in salinization area.