Improved hybrid attention module-based crop pest and disease damage fine-grained identification method

A recognition method and crop technology, applied in the field of computer vision, can solve the problems of unstable effect of attention module and difficulty in improving the accuracy rate, and achieve the effect of low application cost and reduced mapping interval

Active Publication Date: 2020-11-24
SOUTH CHINA AGRI UNIV
View PDF4 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Especially in fine-grained classification tasks, this kind of interference due to "serial connection" will m

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
  • Improved hybrid attention module-based crop pest and disease damage fine-grained identification method
  • Improved hybrid attention module-based crop pest and disease damage fine-grained identification method
  • Improved hybrid attention module-based crop pest and disease damage fine-grained identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] Such as figure 1 and 2 As shown, this embodiment provides a method for fine-grained identification of crop diseases and insect pests based on the improved mixed attention module, including the following steps:

[0053] S1), input the crop disease and insect pest picture of RGB three channels, it is preprocessed, specifically comprises the following steps:

[0054] S101), randomly adopt four methods of bilinear interpolation, nearest neighbor method, bicubic interpolation and area interpolation to process the pictures of crop diseases and insect pests, and scale them to 224×224 operations;

[0055] S102), flipping the picture randomly horizontally and vertically, so as to achieve the purpose of data enhancement.

[0056] S2), the image processed in step S1) is extracted through the convolutional layer features, and then the maximum pooling layer with a small scale of 3×3 is used for further feature abstraction, namely:

[0057]

[0058]

[0059] where x out Ind...

Embodiment 2

[0082] The data set used in this example selects pictures of some crop diseases and insect pests jointly created by Shanghai Xinke Technology and Innovation Works. Pests and diseases are further divided into two types: "general" and "serious" (for example: "apple scab is normal" and "apple scab is severe"), with a total of 61 categories. The data set used in this example has a total of 36,258 pictures of crop diseases and insect pests, and the complete labels of the data set are shown in Table 1.

[0083] Table 1 Classification labels of crop diseases and insect pests dataset

[0084]

[0085]

[0086]

[0087] On this basis, this embodiment divides the pictures into a training set and a test set, wherein the training set contains 31718 pictures, the test set contains 4540 pictures, and the pixel size of most pictures in the data set is about 300×400. Some pixels are pictures with a resolution of 2kd; therefore, in order to meet the input size of the convolutional ne...

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 an improved hybrid attention module-based crop pest and disease damage fine-grained identification method. The method comprises the following steps of: firstly, inputting a crop disease and insect pest picture, performing feature extraction through a convolution layer after preprocessing, and taking a feature map F obtained by the convolution layer as input of attention I _CBAM by using an Inception thought in combination with a residual connection structure in a forward propagation process to obtain weights MC (F) and MS (F); and finally, obtaining a feature map F2, and generating a final prediction probability by using a softmax function. In order to improve the accuracy of a disease and pest identification model and detect diseases and pests in time, the hybridattention CBAM is improved; through a parallel connection structure of channel attention and space attention, the problem of interference generated by serial connection of channel attention and spaceattention is solved, and the direct generalization of I _ CBAM in different models is ensured while the improvement of the accuracy of a pest and disease damage fine-grained identification model afterattention adding is more stable.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method for fine-grained identification of crop diseases and insect pests based on an improved hybrid attention module. Background technique [0002] Among the various natural disasters of plants in our country, crop diseases and insect pests occupy a very important position. Therefore, it is necessary to predict and monitor them in time to prevent the occurrence of major disasters. [0003] However, the occurrence of pest and disease disasters is closely related to factors such as planting system, crop layout, and climate trends. It is very difficult for ordinary farmers to predict pests and diseases in small-scale planting. There are often more than ten kinds of pesticides to choose from for the control of a disease and insect pest, and the environmental pollution and greenhouse gas emissions caused by each pesticide are also different. At the same time, the different...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/0409G06N3/08G06V20/13G06N3/045G06F18/2415G06F18/214
Inventor 王美华吴振鑫周祖光
Owner SOUTH CHINA AGRI 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