Surface Defect Detection Algorithm Using a Lightweight Convolutional Block Attention Transformer
DOI:
CSTR:
Author:
Affiliation:

1.School of Artificial Intelligence, Nanjing University of Information Science and Technology;2.School of Computer Science, Nanjing University of Information Science and Technology

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    针对缺陷语义分割任务中缺陷与背景差异小以及同类缺陷间差异大的问题,本文提出了一种基于轻量化卷积块注意力Transformer的表面缺陷检测算法,即LCBAFormer,该算法旨在提高不同类型缺陷的分割准确率。本文首先设计了轻量化卷积块注意力模块LCBAM,通过结合通道注意力模块和空间注意力模块,提取有效的通道信息和空间信息,使模型更专注于局部缺陷特征信息,增强不同缺陷间的特征差异,减少同类缺陷间的差异。其次,本文提出轻量化的语义注入模块SIM,以逐步融合多尺度特征信息,提升网络对不同缺陷的定位和区分能力。实验结果表明,在NEU-Seg钢带缺陷数据集和MT-Defect磁瓦缺陷数据集上,本文提出的算法平均交并比(mIoU)分别为84.75%和79.46%mIoU,平均召回率(mRec)分别为92.29%和87.50%,平均F1分数(mF1)分别为91.52%和88.08%mF1,此外,算法的计算复杂度较低,每秒十亿次浮点运算次数(GFLOPs)分别为1.03和2.65。与主流算法相比,本文方法在参数量和检测性能上做到了较好的平衡,具有更小参数量和更好的分割结果。

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:June 05,2024
  • Revised:September 04,2024
  • Adopted:September 05,2024
  • Online:
  • Published:
Article QR Code

Address:No. 219, Ningliu Road, Nanjing, Jiangsu Province

Postcode:210044

Phone:025-58731025