An automatic analysis method for wheat powdery mildew disease severity in the middle and late stage

Abstract

The invention relates to an automatic analysis method for wheat powdery mildew disease severity in the middle and late stages. The method comprises the following steps: (1) collecting hyperspectral data of leaf samples. (2) Calculate the percentage of diseased spots in the leaf area according to the number of pixels in the entire leaf and diseased spot area, and obtain the disease severity a0 of the leaf sample. (3) Perform dimensionality reduction processing on the test data. (4) After dimensionality reduction, select m1 samples among m samples as training samples, use d-dimensional features as independent variables, and input corresponding categories as dependent variables into the probabilistic neural network model for training to obtain a prediction model; The remaining m-m1 samples are used as test samples to verify the accuracy of the model, and the d-dimensional features of these m-m1 samples are input into the probabilistic neural network model as independent variables to obtain the prediction result a1 of the severity of the disease of the leaf sample, and compare a1 with a0. The invention can monitor and analyze the severity of wheat powdery mildew.

Description

technical field [0001] The invention relates to the technical field of monitoring wheat powdery mildew, in particular to an automatic analysis method for the severity of middle and late stages of wheat powdery mildew. Background technique [0002] The occurrence of wheat powdery mildew mainly harms the leaves of wheat. The leaf damage caused by the disease will cause damage to the physiological structure of the leaves, which will indirectly affect its photosynthesis and nutrient supply. The changes in spectral absorption will be intuitively displayed, and for wheat plants In general, the leaf is the organ with the largest proportion observed at the canopy scale and aerial scale. In the middle and late stages of disease onset, the DI of many leaves is already higher than 50%. At this time, it is not meaningful to classify and grade and give control guidance, which is mainly to provide reference for disaster loss assessment. [0003] Traditional disaster assessment mainly obt...

AI Technical Summary

Problems solved by technology

The above studies on crops based on hyperspectral are mainly focused on the classification and discrimination of different diseases. There are few studies on the monitoring of disease severity, and there are few studies on the discrimination of the severity of a single disease of wheat powdery mildew in the critical period.

Images