Hyperspectral Data Mapping In Soil Organic Matter

Soil is a natural resources for human survival,is an important part of sustainable agricultural development,and soil organic matter content reflect important characteristics of soil fertility.Cycle is long past the traditional determination of the conventional methods of chemical analysis of soil organic matter content,high cost,high-precisionmeasurement points are scarce,can not meet the needs of precision agriculture on the status of temporal an spatial variation of soil components.Therefore,the soil organic matter content in a short time to obtain the required information is particularly important. Since the1960s, the emergence and development of remote sensing technology,soil science has opened up new areas.Hyperspectral remote sensing technology can be exemsively applied in soil nutrient research due to its three special advantages,high spectral resolution,strong waveband continuity as well as a great deal of spectral information. In this experiment, spectral reflectance of soil samples were measured using ASD2500hyperspectral meter. It was analyzed that there were soil OM changing characteristics and spectral reflectance curve shaps. And organic matter content for soil spectral reflectance transform correlation analysisto identify relatively high correlation band,and the use of multiple stepwise linear regression method to establish the appropriate estimation model,and mode is were tested.The present paper studies the relationship between soil spectrum and soil organic fraction based on spectrum curves(ranging from350to2500nm)of44soil samples,which were collected in DeXing county,jiangxi province.First,soil reflection spectrum was mathematically manipulated,And then analyze the spectral reflectance transform correlation between form and content of organic matter.Correlation analysi showed that The first derivative of the soil organic matter content and spectral reflectance was significantly related band490~770nm、1420~1440nm、1906~1910nm. Model to establish the regression,Finally,right HYPERION data mapping.