MIM gear defect detection based on transfer learning and feature reuse

ZHAO Tong; LEI Baozhen; WANG Xunwei; QI Guanghao

doi:10.13873/J.1000-9787(2021)10-0129-03

Chinese

您当前的位置：

首页 >

文章列表页 >

MIM gear defect detection based on transfer learning and feature reuse

更新时间：2025-09-11

- MIM gear defect detection based on transfer learning and feature reuse
- Issue 10, Pages: 129-131+135(2021)
- 作者机构：
  
  1. 北京联合大学北京市信息服务工程重点实验室
  2. 北京联合大学北京市智能机械创新设计服务工程技术研究中心
- 作者简介：
- 基金信息：
- DOI：10.13873/J.1000-9787(2021)10-0129-03
  CLC： TP183;TH132.41
- Published：2021
- 稿件说明：
移动端阅览
ZHAO Tong, LEI Baozhen, WANG Xunwei, et al. MIM gear defect detection based on transfer learning and feature reuse[J]. 2021, (10): 129-131+135.
DOI：

ZHAO Tong, LEI Baozhen, WANG Xunwei, et al. MIM gear defect detection based on transfer learning and feature reuse[J]. 2021, (10): 129-131+135. DOI： 10.13873/J.1000-9787(2021)10-0129-03.

摘要

针对金属注射成型(MIM)工业生产零件由于缺陷细微导致的肉眼识别率低和人力资源耗费严重的问题

设计了一种基于卷积神经网络(CNN)的缺陷检测识别方法。利用迁移学习解决当前由于训练数据少而导致识别率低的问题

并提出了一种轻量化特征重用网络模型

对MIM工艺生产的小模数齿轮进行缺陷检测。实验结果表明:训练模型在MIM齿轮缺陷检测的准确率达到98%以上

可对零件实现快速、准确的检测

该检测识别技术可以应用于同类的零件制造业完成分类和检测任务

在工业上具有重要的研究价值和实践意义。

Abstract

Aiming at the problems of low visual recognition rate and serious human resource consumption of metal injection molding（MIM）industrial production parts due to the fine defects

a defect detection and recognition method based on convolution neural network（CNN） is designed.The problem of low recognition rate caused by the lack of training data is solved by using transfer learning.A lightweight feature reuse network model is proposed to detect the defects of small module gears produced by MIM process.The experimental results show that the accuracy rate of the training model in MIM gear defect detection is more than 98 %

which can realize the rapid and accurate detection of parts.The detection and recognition technology can be applied to similar parts manufacturing industry to complete the classification and detection tasks

which has important research value and practical significance in industry.

关键词

Keywords

references

Views

373

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Arrester segmentation algorithm based on transfer learning and improved ResNet

Traffic sign recognition algorithm based on double-channel CNN

Flying object recognition algorithm based on region CNN

Research on restoration technology of motion blur image

Quantity identifying technology for static target of ultra-wideband through-wall radar based on CNN

Related Author

LEI Yingjia

HU Junwei

ZHANG Xianghong

HE Haiyang

DONG Erbao

KONG Yueyao

YAN Qun

YAO Jianmin

Related Institution

Institute of Advanced Technology,University of Science and Technology of China

School of Engineering Science,University of Science and Technology of China

College of Physics and Information Engineering,Fuzhou University

School of Electronic and Electrical Engineering,Shanghai University of Engineering Science

School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology

⁰