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

Alfalfa field weed identification method based on CNN and GraphSAGE

A recognition method, a technology of weeds, applied in the field of image recognition, can solve problems such as the recognition of weeds in unaffected alfalfa fields, and achieve a good generalization effect

Pending Publication Date: 2022-08-09
NANJING FORESTRY UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In recent years, deep learning has been gradually applied to the field of weed identification. However, based on the research status at home and abroad, there is no research on the identification of weeds in alfalfa fields.

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
  • Alfalfa field weed identification method based on CNN and GraphSAGE
  • Alfalfa field weed identification method based on CNN and GraphSAGE
  • Alfalfa field weed identification method based on CNN and GraphSAGE

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0039] Weed identification method in alfalfa field based on CNN and GraphSAGE, using such as figure 1 The shown weed recognition model combining CNN and GraphSAGE network models can fully utilize the semantic information of weed images and can identify weeds well from alfalfa fields with only a few labeled samples. The specific implementation process is as follows:

[0040] Step 1: Input the weed image and use CNN to extract the weed image features;

[0041] CNN has strong feature representation ability. The first CNN used is the fine-tuned ResNet-101 algorithm model. The fine-tuning method is: remove the fully connected layer, and only retain the convolution pooling part for weed image feature extraction, such as figure 2 As shown, input the original weed image, the image is subjected to downsampling calculation and pooling of the convolut...

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 alfalfa field weed recognition method based on CNN and GraphSAGE, and the method comprises the steps: 1, inputting a weed image, and extracting the features of the weed image through the CNN; 2, constructing a weed feature matrix undirected similarity graph based on the Euclidean distance of the weed image features; step 3, inputting the weed feature matrix undirected similarity graph into a graph neural network GraphSAGE to carry out propagation among weed image nodes to obtain fusion features; and step 4, inputting the fusion features into a classifier, and based on feature similarity, obtaining label information of unmarked weed samples from neighbor nodes in the graph to realize weed and alfalfa classification. According to the method, the feature relevance between the weed samples can be fully utilized, and a relatively high weed recognition rate is obtained under the condition that only a small number of marked sample images exist.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to an alfalfa field weed recognition method based on CNN and GraphSAGE. Background technique [0002] Alfalfa is a perennial herb of the genus Leguminosae and is one of the main forage grasses in my country. Planting alfalfa can improve soil, maintain soil and water, and play an important role in protecting the ecological environment. Alfalfa is rich in minerals and a large amount of crude protein, and is considered to be an important feed crop for livestock and the best feed for dairy cows. [0003] Alfalfa grows slowly in the seedling stage, and weeds are a major threat in the production process of alfalfa. They will compete with alfalfa for nutrients, space, sunlight and water. Long-term damage will cause weak alfalfa seedlings, difficult root growth, and difficulty in forming a community. , thereby reducing forage yield and nutritional value. In addition...

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): G06V20/10G06V10/40G06V10/74G06V10/764G06V10/80G06V10/82G06N3/04G06N3/08
CPCG06V20/188G06V10/40G06V10/806G06V10/761G06V10/764G06V10/82G06N3/082G06N3/045Y02A40/70
Inventor 杨婕陈勇于佳琳
Owner NANJING FORESTRY UNIV