Two-band hyperspectral index and prediction model for estimating yield and shoot dry weight of soybean

Abstract

The invention belongs to the field of agricultural remote sensing application, and provides a model and a method for estimating the yield and shoot dry weight of soybean based on a spectral index. Four soybean yield models based on a two-band sensitive vegetation index are selected by comparing and analyzing the single-band, two-band and three-band vegetation indexes and a known vegetation index as follows: an exponential model based on NDVI (normalized difference vegetation index) (938, 642) at an R2 stage (a full-bloom stage), an R4 stage (a full-pod stage) and an R5 stage (pod-filling initial stage) of the soybean, an exponential model based on the NDVI (938, 642) at the R4 and R5 stages, an exponential model y=2.57e7.88x based on the NDVI (938, 642) at the R5 stage, and a power function model y=6280.97*6.78, and verification on the models is carried out. In addition, a common core waveband of the yield and the shoot dry weight of the soybean is determined, and the predictability of the vegetation index based on the common core waveband for the shoot dry weight can be further confirmed. By adopting the method, the yield and the shoot dry weight of the soybean can be rapidly and nondestructively estimated. The method and the model are applicable to the large-scale breeding of the soybean and particularly applicable to the high-yield breeding of the soybean in ecological areas of the middle and lower reaches of Yangtze River.

Description

technical field [0001] The invention belongs to the field of agricultural remote sensing applications, and relates to an estimation model for measuring soybean yield and aboveground dry weight by two-band spectral index and a method for estimating soybean yield and aboveground dry weight by using the model. technical background [0002] In recent years, China has changed from a net soybean exporter to the world's largest soybean importer, relying too much on foreign imports. There are two main reasons for this unfavorable situation. One is that due to the impact of low-price dumping of foreign genetically modified soybeans in the Chinese market, the price of domestic soybeans has been in a downturn, which has seriously dampened the enthusiasm of domestic soybean farmers, causing the domestic soybean planting area to decrease year by year. As a result, the total soybean output continues to decrease, which cannot meet the growing domestic demand for soybeans; second, the low l...

AI Technical Summary

Problems solved by technology

[0005] However, the amount of information in soybean canopy hyperspectral data is huge. How to fully mine the spectral index constructed in the form of NDVI and RVI in the full spectral range of 350-2500nm, and obtain a band combination with better prediction effect on target traits is a current research topic. problems encountered

Images