Research on few-shot learning of metal defect recognition based on fusion distribution metric strategy

Home > Archive>Volume 46, Issue 4, 2023 >107-113

Research on few-shot learning of metal defect recognition based on fusion distribution metric strategy
DOI:
                        
                    
CSTR:
                        [cstr]
                    
Author:
                        Tang DonglinTang Donglin
1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China；3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
Liu MingxuanLiu Mingxuan
1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China；3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
He Yuanyuan2He Yuanyuan
SIchuan Special Equipment Inspection Institute,Chengdu 610100, China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
Yang ZhouYang Zhou
1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China；3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
Cheng HengCheng Heng
1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China；3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
Ding ChaoDing Chao
1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China；3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site

                    
Affiliation:1.School of Mechanical and Electrical Engineering, Southwest Petroleum University，Chengdu 610500, China; 2.SIchuan Special Equipment Inspection Institute,Chengdu 610100, China; 3.Key Laboratory of Petroleum and Natural Gas Equipment of Ministry of Education，Southwest Petroleum University,Chengdu 610500,China
Clc Number:TP391
Fund Project:

Article

Figures

Metrics

Reference

Cited by

Materials

Comments

Abstract:

In view of the current metal surface defects classification, data is scarce and tagging process is cumbersome and expensive. This paper introduces the few-shot learning into the metal surface defect classification, proposes a few-shot learning network model with more informative detail descriptor to represent image features: Through adding spartial attention mechanism to screen local information and introducing the fusion measurement to class metric. Experiment results show that our model has better metric effect on MiniImageNet.We gains 6.34%, 5.78% and 1.25% improvements over RelationNet, CovaMNet and DN4 algorithms on the 5-way 5-shot task.The average accuracy of 5-way 5-shot on NEU-DET was improved by 2.87%, 3.34%, and 2.5% respectively.

Key words:classification of metal defects;metric learing;few-shot learning;spartial attention mechanism; detail descriptor

Get Citation

Copy

Article Metrics

Abstract:
PDF:
HTML:
Cited by:

History

Received:
Revised:
Adopted:
Online: February 22,2024
Published:

Home

Introduction

Editorial Committee

Policy

Contact Us

中文版

Get Citation

Share

Article Metrics

History

Article QR Code