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

Trademark retrieval method and system based on attention mechanism, computer equipment and storage medium

A technology of attention and trademarks, applied in computer components, calculations, neural learning methods, etc., can solve problems such as inaccurate key features of trademark images, and achieve the effect of optimizing losses

Pending Publication Date: 2021-03-19
GUANGDONG UNIV OF TECH
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the defect of inaccurate extraction of key features of trademark images in the prior art, the present invention provides a trademark retrieval method, system, computer equipment and storage medium based on attention mechanism

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
  • Trademark retrieval method and system based on attention mechanism, computer equipment and storage medium
  • Trademark retrieval method and system based on attention mechanism, computer equipment and storage medium
  • Trademark retrieval method and system based on attention mechanism, computer equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] This embodiment provides a trademark retrieval method based on attention mechanism, such as figure 1 As shown, the method includes the following steps:

[0046] S1: Obtain a trademark data set; the trademark data set is a METU trademark data set.

[0047] S2: Build channel attention network module;

[0048] S3: The preprocessing module preprocesses the trademark image in the trademark data set to obtain the preprocessed trademark image x i , i=1,2,...,n, n is the number of preprocessed trademark images, and it is input into the channel attention network module;

[0049]S4: Channel attention network module according to feature channel domain weight ω i For the preprocessed trademark image x i Train and distribute to obtain the preprocessed trademark image x i Corresponding initial trademark features v i ,i=1,2,...n, the initial trademark feature v i Form the initial trademark feature set V, and input the initial trademark feature set V into the instance distinguis...

Embodiment 2

[0069] This embodiment provides a trademark retrieval system based on attention mechanism, such as image 3 As shown, the system includes a preprocessing module, a channel attention network module, an instance distinguishing module and a retrieval module;

[0070] The preprocessing module preprocesses the trademark image in the trademark data set to obtain the preprocessed trademark image;

[0071] The channel attention network module trains and distributes the preprocessed trademark images to obtain an initial trademark feature set;

[0072] The instance distinguishing module calculates the loss and optimizes the initial trademark features in the initial trademark feature set, obtains trademark feature training data to form a training data set, uses the training data set to train an unsupervised network, and obtains a feature extraction network;

[0073] The retrieval module measures the similarity between the trademark image features to be retrieved and the trademark featur...

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 provides a trademark retrieval method and system based on an attention mechanism, computer equipment and a storage medium, and the method comprises the steps: firstly carrying out the preprocessing of trademark images in a trademark data set, and carrying out the more reasonable weight distribution of initial trademark image features through a channel attention network module; the difference of the initial trademark image features is learned through the example distinguishing module, the loss of the initial trademark image features is optimized, the obtained training data set isused for training the unsupervised network, a feature extraction network is formed, and the feature extraction network can accurately extract key features of trademark images.

Description

technical field [0001] The present invention relates to the field of image retrieval, more specifically, to a trademark retrieval method, system, computer equipment and storage medium based on attention mechanism. Background technique [0002] As an important symbol of commodities and manufacturers, trademarks are increasingly valued by various industries. The core technology in trademark retrieval is the extraction and measurement of trademark features. The accuracy of trademark feature extraction directly affects the subsequent search results. In the past, trademark management departments searched in the form of "classification numbers" and used manual judgment of the similarity between images. This method has the problems of heavy workload and low efficiency. In order to solve the problem of retrieval work, researchers at home and abroad began to use content-based image retrieval methods to capture more accurate trademark feature information and avoid deviations due to ...

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): G06F16/583G06F16/55G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06F16/583G06F16/55G06N3/08G06V10/44G06N3/048G06N3/045G06F18/241Y02D10/00
Inventor 曹江中黄云飞
Owner GUANGDONG UNIV OF TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com