中文 | English
Hefei University of Technology Logo J. Hefei Univ. Tech. (Nat. Sci.)

Navigation

Research on surface defect detection method of aluminum-plastic blister package based on improved DeepLabv3+

Download Full Text

基于改进 DeepLabv3 $ ^{+} $的铝塑泡罩包装表面缺陷检测方法研究

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Oct.2024, Vol.47 No.10: 1314-1320, 1327

DOI: 10.3969/j.issn.1003-5060.2024.10.003

Authors

WANG Junfeng, LIU Mingzhou, WANG Xiaoqiao, GONG Yu, WANG Ziruo

(School of Mechanical Engineering, Hefei University of Technology, Hefei 230009, China)

Abstract and Keywords

Abstract: Aiming at solving the problems of low accuracy of small-scale defect detection, poor edge segmentation effect in complex texture background and slow detection speed of traditional semantic segmentation model for aluminum-plastic blister package defect detection, this paper proposes an improved surface defect detection method for aluminum-plastic blister package. Firstly, the backbone feature extraction network replaces the original Xception network with the lightweight MobileNetV2 network, which significantly reduces the number of model parameters. Secondly, the efficient channel attention (ECA) module is cascaded in the feature extraction module and atrous spatial pyramid pooling (ASPP) module to accelerate the global feature fusion and reduce the loss of detail information, thus improving the segmentation accuracy of the model for small-scale defects. Finally, a boundary refinement module is added to the decoder of DeepLabv3+ to improve the segmentation accuracy of the model on the edges of the defect region under the complex texture background of the aluminum-plastic surface. Experimental verification is carried out on the self-built capsule board image dataset, and the

Keywords: aluminum-plastic blister package; defect detection; semantic segmentation; attention mechanism; boundary refinement

作者信息

汪俊峰，刘明周，王小巧，龚宇，王子若

（合肥工业大学机械工程学院，安徽合肥230009）

摘要和关键词

摘要: 针对传统语义分割模型用于铝塑泡罩包装缺陷检测存在小目标缺陷检测精度低、复杂纹理背景下边缘分割效果差、检测速度慢的问题，文章提出一种改进的铝塑泡罩包装表面缺陷检测方法。主干特征提取网络采用轻量级 MobileNetV2 网络替代原始的 Xception 网络，大幅减少模型参数量；在特征提取模块和空洞空间金字塔池化（atrous spatial pyramid pooling，ASPP）模块中串联高效通道注意力（efficient channel attention，ECA）模块，加快全局特征融合，降低细节信息丢失，从而提高模型对小目标缺陷的分割精度；最后，在 DeepLabv3+ 的解码器中加入边界细化模块，提升模型在铝塑表面复杂纹理背景下对缺陷区域边缘的分割精度。在自建药板图片数据集上进行的验证实验结果表明，所提方法模型与传统 DeepLabv3+、PSPNet、HRNet 等模型相比，平均交并比（mean intersection over union，MIoU）最大提高 14.50%，单图预测时间最大减少 92.71 ms，参数量最大减少 47.67 MiB。文章方法具有较高的识别正确率和效率，可以实现铝塑泡罩包装表面缺陷的快速检测，具有较强的应用性。

关键词: 铝塑泡罩包装；缺陷检测；语义分割；注意力机制；边界细化

Funding

安徽省科技攻关计划资助项目（JZ2016AKKG0837）

Previous Next
Back to Contents

Account

Submission Login Reviewer Login Editor Login