Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (6): 1824-1829.DOI: 10.11772/j.issn.1001-9081.2019111926

• Virtual reality and multimedia computing • Previous Articles     Next Articles

Real-time segmentation algorithm for bubble defects of plastic bottle based on improved Fast-SCNN

FU Lei, REN Dejun, WU Huayun, GAO Ming, QIU Lyu, HU Yunqi   

  1. School of Mechanical Engineering, Sichuan University, Chengdu Sichuan 610065, China
  • Received:2019-11-13 Revised:2020-01-13 Online:2020-06-10 Published:2020-06-18
  • Contact: REN Dejun, born in 1971, Ph. D., associate professor. His research interests include machine vision, image processing, embedded control system.
  • About author:FU Lei, born in 1995, M. S. candidate. His research interests include machine vision, image processing.REN Dejun, born in 1971, Ph. D., associate professor. His research interests include machine vision, image processing, embedded control system.WU Huayun, born in 1993, M. S. candidate. His research interests include machine vision, image processing.GAO Ming, born in 1996, M. S. candidate. His research interests include machine vision, image processing.QIU Lyu, born in 1996, M. S. candidate. His research interests include machine vision, image processing.HU Yunqi, born in 1995, M. S. candidate. His research interests include machine vision, image processing.

基于改进Fast-SCNN的塑瓶气泡缺陷实时分割算法

付磊, 任德均, 吴华运, 郜明, 邱吕, 胡云起   

  1. 四川大学 机械工程学院,成都 610065
  • 通讯作者: 任德均(1971—)
  • 作者简介:付磊(1995—),男,江西丰城人,硕士研究生,主要研究方向:机器视觉、图像处理.任德均(1971—),男,四川成都人,副教授,博士,主要研究方向:机器视觉、图像处理、嵌入式控制系统.吴华运(1993—),男,河南信阳人,硕士研究生,主要研究方向:机器视觉、图像处理.郜明(1996—),男,安徽阜阳人,硕士研究生,主要研究方向:机器视觉、图像处理.邱吕(1996-),女,四川广安人,硕士研究生,主要研究方向:机器视觉、图像处理.胡云起(1995-),男,江西赣州人,硕士研究生,主要研究方向:机器视觉、图像处理.

Abstract: When the bubbles of medical plastic bottles are detected, the arbitrariness of the bubble position in the bottle body, the uncertainty of the bubble size, and the similarity between the bubble characteristics and the bottle body characteristics increase the difficulty of detecting the bubble defects. In order to solve the above problems in the detection of bubble defects, a real-time segmentation algorithm based on improved Fast Segmentation Convolutional Neural Network (Fast-SCNN) was proposed. The basic framework of the segmentation algorithm is the Fast-SCNN. In order to make up for the lack of robustness of the original network segmentation scale, the ideas of the usage of the information between the channels of Squeeze-and-Excitation Networks (SENet) and the multi-level skip connection were adopted. Specifically, the deep features were extracted by further down-sampling of the network, the up-sampling operation was merged with SELayer module in the decoding stage, and the skip connections with the shallow layer of the network were increased two times at the same time. Four sets of experiments were designed for comparison on the bubble dataset with the Mean Intersection over Union (MIoU) and the segmentation time for single image of the algorithm used as evaluation indicators. The experimental results show that the comprehensive performance of the improved Fast-SCNN is the best, this network has the MIoU of 97.08%, the average segmentation time for a medical plastic bottle of 24.4 ms, and the boundary segmentation accuracy 2.3% higher than Fast-SCNN, which improves the segmentation ability of tiny bubbles, and this network has the MIoU improved by 0.27% and the time reduced by 7.5 ms compared to U-Net, and the comprehensive detection performance far better than Fully Convolutional Networks (FCN-8s). The proposed algorithm can effectively segment smaller bubbles with unclear edges and meet the engineering requirements for real-time segmentation and detection of bubble defects.

Key words: semantic segmentation, image processing, Fast Segmentation Convolutional Neural Network (Fast-SCNN), Squeeze-and-Excitation Networks (SENet), defect detection

摘要: 在医用塑瓶的瓶身气泡检测时,瓶身气泡位置的任意性、气泡大小的不确定性以及气泡特征与瓶身特征之间的相似性增加了气泡缺陷的检测难度。针对上述气泡缺陷检测难点问题,提出了一种基于改进快速分割卷积神经网络(Fast-SCNN)的实时分割算法。该分割算法的基础框架为Fast-SCNN,而为弥补原有网络分割尺寸的鲁棒性不足,借鉴了SENet的通道间信息的利用与多级跳跃连接的思想,具体为网络进一步下采样提取深层特征,在解码阶段将上采样操作融合SELayer模块,同时增加两次与网络浅层的跳跃连接。设计四组对比实验,在气泡数据集上以平均交并比(MIoU)与算法单张分割时间作为评价指标。实验结果表明,改进Fast-SCNN的综合性能最好,其MIoU为97.08%,其预处理后的医用塑瓶的平均检测时间为24.4 ms,其边界分割准确率较Fast-SCNN提升了2.3%,增强了对微小气泡的分割能力,而且该网络的MIoU相较现有的U-Net提升了0.27%,时间上降低了7.5 ms,综合检测性能远超过全卷积神经网络(FCN-8s)。该算法能够有效地对较小的、边缘不清晰的气泡进行分割,满足对气泡缺陷实时分割检测的工程要求。

关键词: 语义分割, 图像处理, 快速分割卷积神经网络(Fast-SCNN), SENet, 缺陷检测

CLC Number: