Construction and application of spectral disease index for identification of wheat scab-infected grains

Abstract

The invention particularly relates to the construction of a spectral disease index for identification of wheat scab-infected grains, comprising the following steps: (A) selecting healthy and infected wheat grains as samples, and collecting and obtaining two average spectral curves; The average spectral curve is subjected to first-order differential processing to obtain two first-order differential spectral curves; (C) solve the wavelength λ according to the two average spectral curves 1 , according to two first-order differential spectral curves to solve the wavelength λ 2 (D) calculate the spectral disease index of each wheat grain; (E) draw its frequency histogram respectively according to the spectral disease index of healthy and infected wheat grain; (F) carry out curve fitting to two frequency histograms, take The intersection point of the two fitting curves is the judgment threshold; a method for detecting wheat scab by using the model is also disclosed. The spectral disease index constructed by the invention has the application advantages of non-destructiveness, fast calculation speed, high precision and good stability.

Description

technical field [0001] The invention relates to the technical field of wheat scab disease detection, in particular to the construction and application of a spectral disease index for identifying wheat scab-infected grains. Background technique [0002] Wheat is one of the food crops with the most extensive planting area and the highest total output in the world, and more than 1 / 3 of the world's population takes wheat products as staple food. Therefore, wheat plays an important role in society and economy, and its quality and safety issues have become a hot research topic at home and abroad. Fusarium head blight (FHB) is a common high incidence of wheat, which can produce 25 kinds of mycotoxins, among which deoxynivalenol (DON) and its acetylated derivatives (3ADON, 15ADON) are the most common. DON, also known as vomitoxin, interferes with the activity of ribosomal peptidyl transferase, hinders ribosomal circulation, inhibits protein synthesis, and causes symptoms such as he...

AI Technical Summary

Problems solved by technology

However, spectral data has the characteristics of multiple bands, large data volume, large redundancy, and strong correlation. It takes a lot of time to use traditional machine learning classification methods (such as support vector machines, partial least squares linear discriminant methods, etc.)

Generally speaking, the current detection methods for wheat scab-infected grains cannot take into account both the accuracy and efficiency indicators, so the development and application of a recognition method with a light algorithm and a high recognition rate has become a technical problem that needs to be solved urgently

Images