鄱阳湖湿地植物实测高光谱数据分类

发布时间：2018-03-26 05:16

本文选题：鄱阳湖　切入点：湿地　出处：《遥感信息》2017年05期

【摘要】：利用高光谱数据对湿地植物进行分类历来是植被遥感研究的难点重点之一。选取鄱阳湖湿地为研究区域,利用ASD FieldSpec4便携式地物光谱仪测取了6种优势种植物的反射光谱曲线,在对原光谱数据预处理的基础上,运用导数法、Log(1/R)法、包络线法等方法对光谱曲线进行对比分析,最后根据各项光谱特征参数对植物进行了分类提取。研究结果显示:导数法、Log(1/R)法、包络线等方法能有效地提高植物的光谱区分度,统计各项光谱特征参数能有效对6种植物进行分类,总体分类精度为85.8%,最高的狗牙根为100%,最低的灰化薹草和水蓼为75%。研究成果为植物分类和种类识别奠定了理论基础,同时也为大尺度、快捷监测湿地植物的时空变化提供技术支撑,对湿地植物以及生态环境的保护决策提供科学依据。
[Abstract]:Classification of wetland plants using hyperspectral data has always been one of the difficult points in vegetation remote sensing research. The reflectance spectrum curves of 6 dominant species of plants were measured by ASD FieldSpec4 portable ground object spectrometer in Poyang Lake wetland. On the basis of preprocessing the original spectral data, the spectral curves are compared and analyzed by the derivative method Logan 1 / R method and the envelope method. Finally, the classification and extraction of plants were carried out according to the spectral characteristic parameters. The results show that the derivative method and the envelope method can effectively improve the spectral differentiation of plants. Six species of plants can be classified effectively by statistical spectral characteristic parameters. The total classification accuracy is 85.8, the highest is 100, the lowest is Carex ash and Polygonum hydropiper. The research results lay a theoretical foundation for the classification and classification of plants. At the same time, it also provides technical support for large-scale and rapid monitoring of the temporal and spatial changes of wetland plants, and provides scientific basis for the decision-making of wetland plants and ecological environment protection.
【作者单位】：江西理工大学建筑与测绘工程学院;
【基金】：国家自然科学基金(41101322) 江西省自然科学基金(20114BAB213022) 江西省教育厅科学技术研究项目(GJJ160617)
【分类号】：Q948;TP79

【相似文献】