A winter wheat yield prediction method and system

Abstract

The invention relates to a winter wheat yield prediction method and system, which calculates the cumulative average enhanced vegetation index of the area to be measured, the cumulative average enhanced vegetation index of the area to be measured and the technical yield of the area to be measured, and uses the multiple linear regression method of the least square method to establish the market. Yield estimation model and provincial yield estimation model, and use the two models to monitor and predict the yield of winter wheat. The yield estimation model established by the invention has high applicability to the estimation of crop per unit yield at municipal and provincial levels, the estimation result is accurate, and dynamic yield forecast can be realized, which is conducive to timely understanding of the growth status and changes of winter wheat.

Description

technical field [0001] The invention relates to the technical field of agricultural engineering, in particular to a winter wheat yield prediction method and system. Background technique [0002] Remote sensing agricultural monitoring has the advantages of large-scale observation, high time resolution, and objective and reliable data. It saves time and cost compared with the traditional yield estimation method with heavy workload, high cost, and low efficiency, and provides a scientific and effective means for crop yield estimation. [0003] There are three methods for crop yield estimation using remote sensing: remote sensing statistical yield estimation model, dry matter-yield model and crop model simulation. [0004] The remote sensing statistical production estimation model estimates production by establishing the relationship expression between remote sensing variables and production. [0005] The dry matter-yield model first estimates the aboveground biomass of crops b...

AI Technical Summary

Problems solved by technology

[0007] Among the three methods, the first method is the simplest and requires the least data. The latter two methods are more mechanical, but require a large amount of input data and the operation is more complicated.

Although the construction of remote sensing yield estimation model using vegetation has achieved great success, there are still some problems, such as NDVI is easily saturated in high vegetation coverage areas, there is a lag in response to disasters (such as drought, pests and diseases), and the accuracy of yield estimation Difficult to guarantee, cannot well reflect the trend of increasing yield brought about by agricultural management and scientific and technological progress

Images