Tobacco pathogenic fungus type rapid identification method based on NIR technology and ELM

Abstract

The invention discloses a tobacco pathogenic fungus type rapid identification method based on an NIR technology and ELM. A combination of a potato dextrose agar culture medium and a sterilized soda-lime glass culture dish is selected to culture a pure tobacco pathogenic fungus strain, the growth of tobacco pathogenic fungi is facilitated, the pretreatment steps are simplified, the risk of introducing different types of pollution can be reduced, spectral data are preprocessed through a discrete wavelet transform algorithm, interference factors can be eliminated, extraction of useful information is promoted, spectral characteristics become more obvious, accurate distinguishing of near infrared spectrums of different strains is achieved, and after a pure tobacco pathogenic fungus strain identification model is established through an ELM algorithm model, unknown pure tobacco pathogenic fungus strains can be rapidly identified. The method has the advantages of wide range of types of tobacco pathogenic fungi which can be covered by the method, difficulty in sample pollution, high strain identification accuracy, low implementation difficulty and high identification efficiency.

Description

technical field [0001] The present application relates to the technical field of identification of tobacco pathogenic fungi, in particular to a rapid identification method for tobacco pathogenic fungi based on NIR technology and ELM. Background technique [0002] Tobacco pathogenic fungi are widely distributed and various in variety, which can easily cause tobacco diseases during the growth process, such as black shank, red spot, anthracnose and root black rot, etc., resulting in reduced crop yield and quality decline. During the planting process, effective and rapid identification and monitoring of these pathogenic fungi should be carried out on a regular basis in order to more comprehensively guarantee the overall quality of tobacco. [0003] At present, the most widely used method for identifying fungal species is the morphological method, which uses the physiological and morphological differences displayed by fungi cultured on different media for identification. After is...

AI Technical Summary

Problems solved by technology

[0006] This application provides a rapid identification method of tobacco pathogenic fungi based on NIR technology and ELM, which is used to solve the technical problems of difficult identification of tobacco pathogenic fungi, low efficiency and low identification accuracy

Images