计算机应用 ›› 2021, Vol. 41 ›› Issue (1): 275-279.DOI: 10.11772/j.issn.1001-9081.2020060886

所属专题: 第八届中国数据挖掘会议(CCDM 2020)

• 第八届中国数据挖掘会议(CCDM 2020) • 上一篇    下一篇

基于多尺度卷积神经网络和类内mixup操作的磁瓦表面质量识别

张京爱1, 王江涛1,2   

  1. 1. 淮北师范大学 物理与电子信息学院, 安徽 淮北 235000;
    2. 淮北师范大学 信息学院, 安徽 淮北 235000
  • 收稿日期:2020-05-30 修回日期:2020-07-13 出版日期:2021-01-10 发布日期:2020-11-12
  • 通讯作者: 王江涛
  • 作者简介:张京爱(1980-),女,山东潍坊人,实验师,硕士,主要研究方向:数字图像处理、缺陷检测;王江涛(1979-),男,山东昌乐人,教授,博士,CCF会员,主要研究方向:模式识别、深度学习、目标检测。
  • 基金资助:
    安徽省高校自然科学研究项目(KJ2018ZD038,KJ2019B15);安徽省质量工程项目(2019jxtd142)。

Magnetic tile surface quality recognition based on multi-scale convolution neural network and within-class mixup operation

ZHANG Jing'ai1, WANG Jiangtao1,2   

  1. 1. School of Physics and Electronic Information, Huaibei Normal University, Huaibei Anhui 235000, China;
    2. School of Information, Huaibei Normal University, Huaibei Anhui 235000, China
  • Received:2020-05-30 Revised:2020-07-13 Online:2021-01-10 Published:2020-11-12
  • Supported by:
    This work is partially supported by the Natural Science Research Project for Colleges in Anhui Province (KJ2018ZD038, KJ2019B15), the Anhui Provincial Quality Engineering Project (2019jxtd142).

摘要: 铁氧体磁瓦由于形状的不规则性和表面缺陷的多样性给基于计算机视觉的表面质量识别带来很大的挑战。针对该问题,将深度学习技术引入到磁瓦表面质量识别中,提出一种基于卷积神经网络的磁瓦表面质量识别系统。首先将磁瓦目标从采集到的图像中分割出来并进行旋转从而得到标准图像,然后把改进后的多尺度ResNet18作为骨干网络来设计识别系统。训练时,设计一种新颖的类内mixup操作来提高系统对样本的泛化能力。为了更加贴近实际应用场景,在考虑到光线变化、姿态差异等因素的前提下构建了磁瓦缺陷数据集。在自建的数据集中进行实验的结果表明,该系统可以达到97.9%的识别准确率,为磁瓦缺陷的自动识别提供了可行的思路。

关键词: 磁瓦, 表面缺陷检测, 卷积神经网络, mixup, ResNet18

Abstract: The various shapes of ferrite magnetic tiles and the wide varieties of their surface defects are great challenges for computer vision based surface defect quality recognition. To address this problem, the deep learning technique was introduced to the magnetic tile surface quality recognition, and a surface defect detection system for magnetic tiles was proposed based on convolution neural networks. Firstly, the tile target was segmented from the collected image and was rotated in order to obtain the standard image. After that, the improved multiscale ResNet18 was used as the backbone network to design the recognition system. During the training process, a novel within-class mixup operation was designed to improve the generalization ability of the system on the samples. To close to the practical application scenes, a surface defect dataset was built with the consideration of illumination changes and posture differences. Experimental results on the self-built dataset indicate that the proposed system achieves recognition accuracy of 97.9%, and provides a feasible idea for the automatic recognition of magnetic tile surface defects.

Key words: magnetic tile, surface defect detection, convolution neural network, mixup, ResNet18

中图分类号: