Remote Sensing Monitoring Data of Soybean Growth in Ecosystem

Abstract

: The time series images obtained by remote sensing can reflect the spectral characteristics of farmland soils and crops affected by the environment, thus providing the variation information of crop growth. In crop growing season, the dynamic changes of crop growth can be determined by different time series images. Therefore, remote sensing technology has the advantages of fast, accurate and strong current situation, it has increasingly become an important means of monitoring the dynamic changes of crop growth in a large area. Monitoring crop growth by remote sensing is of great significance for dynamic perception of food security. The purpose of this paper is to analyze the monitoring data of soybean growth under ecosystem by remote sensing technology. On the soybean scale, based on the difference of reflectance caused by the change of water structure, a method for screening and monitoring the sensitive characteristics of soybean growth was proposed. By measuring the spectral data of soybean growth potential samples, based on the characteristics of surface albedo, vegetation index and detail, and combined with correlation analysis and SVM and GASVM, the growth monitoring model on soybean scale was established. The characteristics of sensitivity to soybean growth and significant difference were screened out, it includes three characteristic bands of 340-380, 480-580 and 750-1000 nm, and three vegetation indices of MSR, NDVI and SIPI, WF01 and WF02 are two wavelet features. The experimental results show that in all models, the monitoring model established by MSR and GASVM has the highest monitoring accuracy, which is 75%.