Wheat heavy disease prediction method based on sequential multi-index element depth characteristic

Abstract

The invention discloses a wheat heavy disease prediction method based on a sequential multi-index element depth characteristic. The method comprises following steps: step one, obtaining basic data; step two, establishing a wheat heavy disease prediction model; step three, training a sequential information storage network and a deep convolutional neural network at the same time; step four, obtaining information data of an image to be predicted and environment to be predicted; and step five, predicting a wheat heavy disease. The time for matching a single address is reduced from 2 minute or so to about 2.2 seconds by the method. The matching result is more balanced on matching degree and precision. The method has a high application value on building of smart cities. The method can automatically learn and acknowledge the degrees of wheat diseases in different periods from the data sequences and thus predicts the wheat heavy diseases. By analyzing the current factors, the method can predict the development of wheat diseases.

Description

technical field [0001] The invention relates to the technical field of agricultural plant protection prediction, in particular to a method for predicting severe wheat diseases based on multi-time series attribute element depth features. Background technique [0002] At present, agricultural big data is driving the transformation of agricultural production to precision and intelligence, and data has gradually become an emerging factor of production in modern agricultural production. The model research on big data representation, identification and prediction of wheat diseases in farmland environment is still in its infancy, and it is not perfect either in theory or in algorithm. In particular, the traditional wheat disease identification technology can only identify or predict diseased and non-diseased wheat, but cannot judge the degree of wheat disease. In practical applications, the prediction of severe disease plays an important role in the early control of wheat diseases....

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a wheat severe disease prediction method based on the depth characteristics of multi-time series attribute elements, which solves the problem that the existing technology cannot analyze the dependence relationship between the time observation of wheat disease data, and has not realized the time series prediction of modeling, so that it cannot pass the existing Deficiency of Information Predicting Severe Disease

Images