Wheat yield estimation method based on unmanned aerial vehicle multispectral images

Abstract

The invention discloses a wheat yield estimation method based on unmanned aerial vehicle multispectral images. According to the wheat yield estimation method based on the unmanned aerial vehicle multispectral images, wheat canopy multispectral image data and yield data are combined, and a wheat yield prediction model is constructed by adopting different modeling methods from single-factor and multi-factor modeling perspectives. According to the method, the formation process mechanism of the yield and the influence of different growth periods on wheat yield estimation are fully considered, andthe wheat yield under different treatments can be accurately and nondestructively estimated through screening of sensitive vegetation indexes, combination of the multiple growth periods and the application of a machine learning method. With the method adopted, a new thought is provided for a wheat yield accurate estimation method, and an important theoretical basis and technical support are provided for regional-scale accurate agricultural management and decision making.

Description

Technical field[0001]The invention belongs to the technical field of crop growth monitoring, in particular to a method for estimating wheat yield based on drone multi-spectral images.Background technique[0002]Wheat is the main food crop in many parts of the world and feeds nearly 40% of the world's population. my country's consumption demand for wheat is second only to rice, and its importance is self-evident. Yield is one of the final indicators of crop production. The ability to accurately estimate the final grain yield of crops is of great significance to ensuring high and stable yields.[0003]Crops have spectral characteristics and can absorb, reflect and radiate different spectra. Various sensors of UAV remote sensing technology used in agriculture can detect these spectral characteristics of plants. Different wavelengths of light have different effects on crop growth. The sensors mounted on the drone can collect crop images of different wavebands and extract different features....

AI Technical Summary

Problems solved by technology

But in general, there is still a lack of systematic comparative evaluation of various modeling methods, including simple linear regression methods, machine learning methods, etc.

Images