《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (1): 273-279.DOI: 10.11772/j.issn.1001-9081.2021020333

• 前沿与综合应用 • 上一篇    

基于双重检测的气门识别方法

佘维1,2, 郑倩2, 田钊1, 刘炜1,2, 李英豪1,2()   

  1. 1.郑州大学 软件学院,郑州 450002
    2.互联网医疗与健康服务河南省协同创新中心(郑州大学),郑州 450052
  • 收稿日期:2021-03-05 修回日期:2021-04-16 接受日期:2021-04-22 发布日期:2021-04-27 出版日期:2022-01-10
  • 通讯作者: 李英豪
  • 作者简介:佘维(1977—),男,湖南常德人,教授,博士,CCF会员,主要研究方向:区块链、信息安全、可信分布式系统
    郑倩(1996—),女,河南许昌人,硕士研究生,主要研究方向:图像识别、机器学习
    田钊(1985—),男,河南荥阳人,讲师,博士,主要研究方向:人工智能、信息安全
    刘炜(1981—),男,河南安阳人,副教授,博士,CCF会员,主要研究方向:区块链、无线网络
    李英豪(1987—),男,河南郑州人,讲师,博士,CCF会员,主要研究方向:数字图像处理、模式识别、机器学习。
  • 基金资助:
    国家重点研发计划项目(2020YFB1712401);河南省科技攻关项目(212102310039);中国铁路北京集团公司科技研究开发计划(2021AY03)

Valve identification method based on double detection

Wei SHE1,2, Qian ZHENG2, Zhao TIAN1, Wei LIU1,2, Yinghao LI1,2()   

  1. 1.Collage of Software,Zhengzhou University,Zhengzhou Henan 450002,China
    2.Henan Collaborative Innovation Center of Internet Medical and Health Services (Zhengzhou University),Zhengzhou Henan 450052,China
  • Received:2021-03-05 Revised:2021-04-16 Accepted:2021-04-22 Online:2021-04-27 Published:2022-01-10
  • Contact: Yinghao LI
  • About author:SHE Wei, born in 1977, Ph. D., professor. His research interests include blockchain, information security, trusted distributed system.
    ZHENG Qian, born in 1996, M. S. candidate. Her research interests include image recognition, machine learning.
    TIAN Zhao, born in 1985, Ph. D., lecturer. His research interests include artificial intelligence, information security.
    LIU Wei, born in 1981, Ph. D., associate professor. His research interests include blockchain, wireless network.
    LI Yinghao, born in 1987, Ph. D., lecturer. His research interests include digital image processing, pattern recognition, machine learning.
  • Supported by:
    National Key Research and Development Program(2020YFB1712401);Science and Technology Project of Henan Province(212102310039);Science and Technology Research and Development Plan of China Railway Beijing Group Company Limited(2021AY03)

摘要:

针对目前工业中的气门识别方法存在重叠目标漏检率高、检测精度较低、目标包裹度差、圆心定位不准的问题,提出了一种基于双重检测的气门识别方法。首先,运用数据增强对样本进行轻量扩充;其次,以深度卷积网络为基础,加入空间金字塔池化层(SPP)和路径聚合网络(PAN),同时调整先验框,改进损失函数,从而提取气门预测框;最后,以霍夫圆变换(CHT)方法对预测框中的气门进行二次识别,从而达到精准识别气门区域的目的。把所提方法和原YOLOv3、YOLOv4、传统CHT方法进行对比,并采用精确率、召回率、交并比联合进行检测效果评估。实验结果表明,所提出的方法在检测精度和召回率上分别达到了97.1%和94.4%,相较原YOLOv3方法分别提高了2.9个百分点和1.8个百分点;且该方法使目标包裹度更好,目标中心点的定位更准确,其矫正框和真实框的交并比(IOU)达到了0.95,与传统CHT方法相比提高了0.05。所提方法在提高模型识别准确率的同时提高了目标抓取的成功率,在实际应用中有一定的实用价值。

关键词: 目标检测, 气门识别, YOLO方法, 霍夫圆变换, 二次识别

Abstract:

Aiming at the problems that current valve identification methods in industry have high missed rate of overlapping targets, low detection precision, poor target encapsulation degree and inaccurate positioning of circle center, a valve identification method based on double detection was proposed. Firstly, data enhancement was used to expand the samples in a lightweight way. Then, Spatial Pyramid Pooling (SPP) and Path Aggregation Network (PAN) were added on the basis of deep convolutional network. At the same time, the anchor boxes were adjusted and the loss function was improved to extract the valve prediction boxes. Finally, the Circle Hough Transform (CHT) method was used to secondarily identify the valves in the prediction boxes to accurately identify the valve regions. The proposed method was compared with the original You Only Look Once (YOLO)v3, YOLOv4, and the traditional CHT methods, and the detection results were evaluated by jointly using precision, recall and coincidence degree. Experimental results show that the average precision and recall of the proposed method reaches 97.1% and 94.4% respectively, 2.9 percentage points and 1.8 percentage points higher than those of the original YOLOv3 method respectively. In addition, the proposed method improves the target encapsulation degree and location accuracy of target center. The proposed method has the Intersection Over Union (IOU) between the corrected frame and the real frame reached 0.95, which is 0.05 higher than that of the traditional CHT method. The proposed method improves the success rate of target capture while improving the accuracy of model identification, and has certain practical value in practical applications.

Key words: target detection, valve identification, You Only Look Once (YOLO) method, Circle Hough Transform (CHT), secondary identification

中图分类号: