面向土壤分类的高光谱反射特征参数模型

发布时间：2018-04-11 09:13

本文选题：土壤分类 + 光谱特征参数　；参考：《遥感学报》2017年01期

【摘要】：提出了一种无损、快速、成本低的土壤分类方法,选取松嫩平原4种典型土壤(黑土、黑钙土、风砂土和草甸土)耕层(0—20 cm)土样的实验室反射光谱数据作为研究对象,采用重采样、包络线消除法处理光谱数据,提取反映反射光谱特征的光谱特征参数,利用K均值聚类(K-means clustering)和决策树(decision tree)分别进行聚类分析和分类模型构建,实现土壤的快速分类。结果表明,利用表层土壤反射光谱特征参数构建的决策树分类模型可以对研究区土壤进行分类。研究成果有望加快土壤制图,为土壤理化性质的时空变化研究提供技术支持。
[Abstract]:A fast and low cost soil classification method was proposed. The laboratory reflectance spectrum data of four typical soils (black soil, calcareous soil, aeolian sandy soil and meadow soil) were studied.Resampling and envelope elimination are used to process spectral data, and spectral characteristic parameters reflecting the characteristics of reflection spectrum are extracted. Cluster analysis and classification model are constructed using K-means clustering of K-means and decision tree tree, respectively.The rapid classification of soil is realized.The results show that the decision tree classification model based on the characteristic parameters of surface soil reflectance spectrum can be used to classify the soil in the study area.The research results are expected to accelerate soil mapping and provide technical support for the spatiotemporal variation of soil physical and chemical properties.
【作者单位】：东北农业大学资源与环境学院;黑龙江省农垦科学院科技情报研究所;
【基金】：国家自然科学基金项目(编号:41501357) 黑龙江省普通高等学校新世纪优秀人才培养计划黑龙江省博士后启动基金(编号:LBH-Q13026)~~
【分类号】：S155

【相似文献】