A method for rapidly diagnosing crop diseases by using volatiles

Abstract

The invention discloses crop volatile matter detection by a metal oxide sensor array. The method comprises the following steps: 1) putting a crop sample in a sampling device, allowing the sample to stand for 30-60 minutes, when top air is balanced, introducing the headspace gas in the container into a sensor array reaction chamber by a sampling pump, allowing the sensor to react with the gas so as to obtain a corresponding response signal which is converted into digits and is inputted into a computer by an acquisition card; 2) performing feature selection and feature extraction of a feature matrix of electronic nose raw data; 3) establishing a mathematical model of the relationship between sensor response signals and different degrees of sample diseases by neural network with the raw dataand the treated data being used as the sensor response signals respectively. The invention has simple operation, greatly improved reliability and repeatability, and increased detection efficiency.

Description

technical field [0001] The invention relates to a method for detecting crop diseases using volatile matter. Background technique [0002] Crop diseases are one of the most important factors restricting the stable development of agricultural production. In my country, the losses caused by diseases to agriculture amount to billions of RMB every year, among which the damage caused by the diseases transmitted by seedlings is the greatest. According to statistics, the annual occurrence of diseases, insect pests and weeds in my country reaches 236 million hectares, and the loss of about 15% of grain, 20% to 25% of cotton, and more than 25% of fruits and vegetables is lost every year. [0003] The traditional method of diagnosing crop diseases is pathological analysis or empirical judgment by professionals, but the speed of pathological analysis is slow, the real-time performance is poor, and empirical judgment is highly subjective and has large errors. With the development of sc...

AI Technical Summary

Problems solved by technology

These methods have made great contributions to the diagnosis of crop diseases, but they have some shortcomings: it is time-consuming and laborious to diagnose rice diseases by using disease symptoms, and they rely heavily on experts; satellite remote sensing technology is more suitable for large-scale crops Disease detection still has some difficulties in the management of small areas; and computer vision technology, although one of the more advanced technologies, still has a certain distance to achieve real-time monitoring of crop diseases

Images