Corn disease grading method and system and computer equipment

Abstract

The invention relates to a corn disease grading method, a corn disease grading system and computer equipment. The method comprises the following steps: S1, acquiring a normal corn leaf image and a diseased corn leaf image; S2, constructing a disease classification neural network, and initializing parameters of a convolutional layer and a full connection layer by adopting transfer learning; S3, establishing an area segmentation neural network, and initializing first two layers of network parameters by adopting transfer learning; s4, inputting the to-be-detected leaf original image into the corndisease classification neural network to obtain disease classification information; inputting the original image into an area segmentation neural network to obtain a binary image of a disease area and a leaf main body area, and calculating the area proportion of the disease area; and S5, grading disease conditions according to the disease classification information and the area ratio. The invention provides a method which adapts to pictures shot in a real environment, does not need to pick corn leaves and automatically grades common leaf disease conditions of corn, and lays a foundation for grading identification of corn diseases, especially early diagnosis, prediction and early warning.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a corn disease classification method, system and computer equipment. Background technique [0002] Corn is an important food crop and feed crop in my country. In recent years, due to the change of cultivation system, the variation of pathogenic bacteria species, and the unsound plant health care measures, the harm of corn diseases has increased year by year, and the types have also increased year by year. Common corn leaf diseases include: large spot disease and small spot disease , rust, round spot, Curvularia leaf spot, dwarf mosaic, etc. Therefore, the grading and identification of corn diseases is very important for corn planting, especially in early diagnosis and forecasting and early warning. Using intelligent means to detect disease types and classify the disease can improve farmers' awareness of diseases. awareness and take corresponding measures to control and improve a...

AI Technical Summary

Problems solved by technology

It has the following deficiencies: 1. the scope of use is limited

It has the following deficiencies: 1. It needs to provide a single background color when shooting. This technology uses white as the background when collecting images. This method has an algorithm to segment the background area, but it cannot adapt to the changing scenes in the real environment. Usually, there are a lot of background interference colors in photos; 2. The effect of threshold segmentation is not ideal. Threshold segmentation is an image segmentation algorithm commonly used in machine vision. Through threshold segmentation of RGB color channels, this method will eliminate abnormalities caused by leaf veins or other factors. The green part is calculated as a lesion, which affects the segmentation result and reduces the detection accuracy

Images