Logo image detection method and system based on feature fusion

A technology of feature fusion and image detection, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problems of insufficient model feature extraction capabilities, lack of long-distance information association, etc., to achieve rich and enhanced detailed information and semantic information The ability of remote information extraction and the effect of accurate detection results

Pending Publication Date: 2022-05-27
SHANDONG NORMAL UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above problems, this disclosure proposes a Logo image detection method and system based on feature fusion, which optimizes and solves the problems of insufficient feature extraction capabilities of existing models and lack of long-distance information association, so as to improve the accuracy of Logo detection and be more applicable for real-world applications and scenarios

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
  • Logo image detection method and system based on feature fusion
  • Logo image detection method and system based on feature fusion
  • Logo image detection method and system based on feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] In the technical solutions disclosed in one or more embodiments, such as figure 1 As shown, a Logo image detection method based on feature fusion includes the following steps:

[0034] Step 1: Perform feature extraction on the Logo image to be detected;

[0035] Step 2: Input the extracted features into a feature pyramid network that fuses multiple heterogeneous information paths for feature fusion;

[0036] Step 3: Input the fused feature map into the region generation network constructed by involution to generate a region suggestion box;

[0037] Step 4: Classify and locate the regional proposal box through the fully connected layer, and calculate the category and final position coordinates of the proposal box.

[0038] In this embodiment, the effect of feature fusion is enhanced by using a combination of feature pyramid network and neural network structure search, so that the extracted detailed information and semantic information are more abundant, and the detecti...

Embodiment 2

[0081] Based on Embodiment 1, this embodiment provides a logo image detection system based on feature fusion, including:

[0082] Feature extraction module: configured to perform feature extraction on the Logo image to be detected;

[0083] Feature fusion module: configured to input the extracted features to a feature pyramid network that fuses multiple heterogeneous information paths for feature fusion;

[0084] Suggestion box generation module: It is configured to input the fused feature map into the region generation network constructed by involution to generate region suggestion boxes;

[0085] Classification and positioning module: It is configured to classify and locate the regional proposal box through the fully connected layer, and calculate the category and final position coordinates of the proposal box.

Embodiment 3

[0087] This embodiment provides an electronic device, including a memory, a processor, and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps described in the method of Embodiment 1 are completed.

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 Logo image detection method and system based on feature fusion. The detection method comprises the following steps: carrying out feature extraction on a to-be-detected Logo image; inputting the extracted features into a feature pyramid network fusing a plurality of heterogeneous information paths for feature fusion; inputting the fused feature map into a region generation network constructed in an involute mode, and generating a region suggestion box; the regional suggestion boxes are classified and positioned through a full connection layer, and the categories and final position coordinates of the suggestion boxes are obtained through calculation. The method optimizes and solves the problems that an existing model is insufficient in feature extraction capability and short of long-distance information association, so that the precision of Logo detection is improved, and the method is more suitable for practical applications and scenes.

Description

technical field [0001] The present disclosure relates to the related technical field of image processing, to target detection technology in computer vision, and in particular, to a method and system for detecting a Logo image based on feature fusion. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] With the booming development of the Internet, information in social media has exploded, and the ways of conveying information have become diverse. As the most common, intuitive and efficient information transmission medium, pictures play a role that cannot be ignored. Logo is the visual image of enterprise comprehensive information. Its diversification and uniqueness make the importance of Logo detection technology in practical applications gradually appear. Logo detection is used in commodity image recognition, vehicle logo intelligent recogni...

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): G06V10/40G06V10/25G06K9/62G06N3/04G06N3/08G06V10/774G06V10/82G06V10/80
CPCG06N3/08G06N3/047G06N3/045G06F18/253
Inventor 侯素娟李星卓邵文静刘文洁贾伟宽郑元杰
Owner SHANDONG NORMAL 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