A method for identifying crop leaf diseases based on improved convolution neural network model AlexNet

A convolutional neural network and disease identification technology, which is applied in the field of crop leaf disease identification based on the improved convolutional neural network model AlexNet, can solve the problems of poor robustness and poor recognition effect, and achieve the effect of improving the recognition effect

Inactive Publication Date: 2018-12-25
JIANGSU UNIV
View PDF1 Cites 54 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In order to solve technical problems such as poor recognition effect and weak

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method for identifying crop leaf diseases based on improved convolution neural network model AlexNet
  • A method for identifying crop leaf diseases based on improved convolution neural network model AlexNet
  • A method for identifying crop leaf diseases based on improved convolution neural network model AlexNet

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The present invention is described in further detail in conjunction with accompanying drawing and specific embodiment:

[0027] A crop leaf disease identification method based on the improved convolutional neural network model AlexNet. It is characterized in that it comprises the following steps:

[0028] (1) Acquisition and processing of experimental data;

[0029] (2) Construct a convolutional neural network model;

[0030] (3) Train the convolutional neural network and test the results;

[0031] Described process (1) specifically comprises the following steps:

[0032] The number of data sets needs to be considered when selecting test samples. If the number of data sets is too small, the network will be undertrained and overfitting will occur. Overlearning with a small number of samples will make the model very sensitive to image distortion; if If the sample of the data set is too large, the training speed of the network will be very slow, and the convergence of ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A method for identifying crop leaf diseases based on an improved convolution neural network model AlexNet is disclosed. Firstly, the expanded and balanced data set is divided into a training set and atest set according to the ratio of 4: 1, and the picture data set is normalized, averaged and so on. Then the iterative input samples are trained to set and optimize the relevant parameters of the network. Finally, the trained network is tested using the test set. It is found that the average classification accuracy of the improved model is 99.3% and the loss rate is only 2%. On this basis, the invention also conducts a comparative experiment based on the gray image and the segmented image training network to explore the influence of the background and the color on the crop disease identification system. The analysis result shows that the disease identification method of the invention has high accuracy rate and low loss rate for disease identification classification of crops, and featuressuch as color and background of the fused image can improve the identification effect.

Description

technical field [0001] The invention relates to an image recognition and classification method, in particular to a crop leaf disease recognition method based on an improved convolutional neural network model AlexNet. Background technique [0002] Agriculture is the foundation of national economic development, and crop diseases seriously affect the yield and quality of agricultural production, causing significant economic losses. Crop disease identification is an important task, especially the disease diagnosis based on the diseased leaves of crops is particularly critical. If the type of crop disease cannot be correctly diagnosed, pesticides cannot be used correctly, not only will the crop disease not be diagnosed and treated in time, but it may also lead to environmental pollution and other problems. Therefore, finding a fast and accurate method for identifying crop leaf diseases has received extensive attention and research. [0003] The traditional methods of crop disea...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/04
CPCG06V20/68G06N3/045G06F18/214
Inventor 孙俊汪龙芦兵谭文军武小红沈继锋戴春霞
Owner JIANGSU UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap