基于电磁特性和反射光谱信息的土壤盐渍度快速检测方法研究

摘要

To implement rapid situ detection for soil salinization, this paper presents the quantitative relationship between soil electrical conductivity & soil spectral reflectance and soil salinization and then sets up a detection system for determination of soil salinization. First, the electromagnetic induction method is used to analyze soil salinization degree quantitatively, where the relative relationship between primary and secondary magnetic field in-duced is established by measuring the conductivity and then the degree of soil salinization is determined. Second, according to the reflectance properties of soil in visible light, infrared and microwave bands the salinization image can be obtained by combining spectra with TM data, Using TM data as the basis of image of soil salinization, the geometric correction can be apply to adapt to the actual object. In order to improve the recognition rate of the image, support vector machine classification is used for its high accuracy and stability, and a specific evaluation is pre-sented to measure the soil salinizaition. From the soil test results, the evaluation of soil salinization based on the electromagnetic characteristics and spectral characteristics is feasible.%为完成温室环境土壤盐渍化程度快速现场检测，采用基于土壤电导率和土壤光谱反射率与土壤盐渍化的定量关系描述技术，建立了一套用于现场表征土壤盐渍化程度的快速检测系统。首先，对土壤盐度进行定性分析，应用电磁感应方法测度土壤盐渍化程度，通过测量电导率推出原生磁场和诱导出的次生磁场的相对关系，进而推算出土壤盐渍化程度。接着，基于土壤在可见光、红外和微波各波段的反射特性，将光谱图和TM数据结合得到盐渍化影像。研究采用了TM数据作为土壤盐渍化的基础图像，通过几何精校正映射至实际地物点。为了提高影像识别率，采用了精度和稳定性比较高的支持向量机分类法，并给出了测度公式。从对土壤的试验结果看，以电磁特性和光谱特性来综合评价土壤盐渍化程度是可行的。