基于高光谱成像的土壤速效氮含量预测研究OA

Taking soil before sowing in the experimental field of Shanxi Agricultural University as the research object,a total of 230 soil sam-ples to be tested were collected as the research objects.The hyperspectral data of the region of interest(ROI)of the samples were collected and averaged by the self-made sampling procedure,and then the soil available nitrogen content was chemically determined.Genetic algorithm(GA),variable combination population analysis(VCPA)combined with GA,and Cuckoo search(CS)combined with GA were used to extract key bands respectively for the raw average spectral variables.The extracted spectral data was subjected to multiple scattering correction(MSC)and the data results were established by partial least square regression(PLSR)and multiple linear regression(MLR).The correlation coeffi-cient(R)and root mean square error(RMSE)of the prediction results obtained by VCPA-GA-MSC-PLSR were 0.899 12 and 86.262 mg/kg respectively,the prediction effect was the best,and it can more truly reflect the level of soil available nitrogen content.The research results can provide theoretical support for the application of hyperspectral imaging technology to the detection of soil available nitrogen content and provide a reference for the rapid detection of other soil components.