Unlock instant, AI-driven research and patent intelligence for your innovation.

Crop disease and pest identification method based on deep learning

A recognition method and deep learning technology, applied in the field of image recognition of pests and diseases, can solve problems such as difficulty in ensuring recognition accuracy

Pending Publication Date: 2021-11-05
XUZHOU NORMAL UNIVERSITY
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The detection of crop diseases and insect pests is inherently variable, and the state presented in the image is also complex. It is difficult to guarantee the accuracy of image recognition with a single algorithm.

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
  • Crop disease and pest identification method based on deep learning
  • Crop disease and pest identification method based on deep learning
  • Crop disease and pest identification method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The technical solutions in the embodiments of the present invention will be specifically described below in conjunction with the accompanying drawings.

[0031] A method for identifying pests and diseases based on deep learning technology described in this embodiment comprises the following steps:

[0032] Step 1 Obtain real images of crop diseases and insect pests. Specifically, the data set of crop diseases and insect pests used in the invention is the public data set of the 2018 AI_Challenger crop diseases and insect pests detection competition. The total number of training images is 31,718, including a total of 61 categories (by "species-disease-degree " points), 10 crop types, 27 diseases (24 of which are divided into two levels of general and severe), and 10 health categories.

[0033] Step 2 divides the obtained sample data into training set, verification set and test set, and divides them according to the ratio of 8:1:1. The total number of training images is 3...

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

The invention discloses a crop disease and pest identification method based on deep learning, and relates to the technical field of disease and pest identification. The method comprises the following steps: acquiring a crop disease and insect pest image data set, and preprocessing images, including image enhancement, data enhancement and normalization processing; training the ResNet101 in combination with an FPN (Fabry-Perot Networks) network by using a weight parameter pre-trained by a cock data set; and generating a bounding box through an RPN network from the extracted feature map, removing an overlapping box through NMS, gathering region features through RoIAlign, and finally outputting coordinates and classification of plant diseases and insect pests through a full connection layer. According to the crop disease and pest identification method based on deep learning, accurate positioning of the disease target leaf is realized, accurate disease and insect pest position information and disease types are provided, and the disease and insect pest image recognition rate is improved.

Description

technical field [0001] The invention relates to a method for identifying crop diseases and insect pests based on deep learning, which belongs to the technical field of image identification of diseases and insect pests. Background technique [0002] The traditional detection technology of agricultural diseases and insect pests depends entirely on people's subjective experience of crops. This method is slow and the accuracy cannot be effectively guaranteed, resulting in low overall detection efficiency. In recent years, with the introduction and development of the concept of deep learning, the recognition rate and speed of image targets have been significantly improved. Using this technology to detect crop pests can not only improve the accuracy of pest identification, but also effectively reduce labor costs. [0003] The detection of crop diseases and insect pests is inherently variable, and the state presented in the image is also complex. It is difficult to guarantee the a...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/32G06K9/46G06K9/62G06N3/04G06N3/08G06Q50/02
CPCG06N3/08G06Q50/02G06N3/045G06F18/2415
Inventor 刘笑意段纳
Owner XUZHOU NORMAL UNIVERSITY