作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2019, Vol. 45 ›› Issue (12): 237-242. doi: 10.19678/j.issn.1000-3428.0054915

• 图形图像处理 • 上一篇    下一篇

一种基于YOLOv3的共享单车违规停放检测方法

盛宜华1, 武友新1, 姚磊岳2   

  1. 1. 南昌大学 信息工程学院, 南昌 330031;
    2. 江西科技学院 协同创新中心, 南昌 330098
  • 收稿日期:2019-05-14 修回日期:2019-06-25 发布日期:2019-07-12
  • 作者简介:盛宜华(1994-),男,硕士研究生,主研方向为计算机视觉、数据挖掘、人工智能;武友新、姚磊岳,教授。
  • 基金资助:
    江西省科技厅科技计划专项"基于自然语音交互模式的行车安全辅助系统"(20171BBE50060);南昌市科技局科技计划项目"基于移动互联网的‘人-车’语音交互系统"(2016-ZCJHCXY-013)。

An Illegal Parking Detection Approach for Shared Bicycles Based on YOLOv3

SHENG Yihua1, WU Youxin1, YAO Leiyue2   

  1. 1. School of Information Engineering, Nanchang University, Nanchang 330031, China;
    2. Center of Collaboration and Innovation, Jiangxi University of Technology, Nanchang 330098, China
  • Received:2019-05-14 Revised:2019-06-25 Published:2019-07-12

摘要: 为解决共享单车随意停放给交通管理带来的困难,提出一种基于计算机视觉的共享单车违规停放检测方法。通过多尺度检测训练以及k-means维度聚类改进YOLOv3网络,在此基础上获取共享单车在图片上的特征矩阵,根据特征矩阵计算当前场景下共享单车的运行状态并进行状态统计。在交通监控视频数据集上的测试结果表明,该方法的检测准确率达到87%以上,能够实现共享单车违规停放的有效检测并给出实时预警。

关键词: 共享单车, 停放检测, YOLOv3网络, 计算机视觉, 状态统计

Abstract: To solve the transportation management difficulties caused by the random parking of shared bicycles,this paper proposes an illegal parking detection method based on computer vision for shared bicycles.The method uses multi-scale detection training and k-means dimensional clustering to improve YOLOv3 network,and accordingly obtains the feature matrix of the shared bicycle on the image.Based on the obtained feature matrix,the running state of the target shared bicycle in a certain scenario can be computed for statistics.Testing results on the dataset of transportation monitoring videos show that the detection accuracy rate of the proposed method is above 87%,which means it can effectively detect illegal parking of shared bicycles and give real-time alarms.

Key words: shared bicycles, parking detection, YOLOv3 network, computer vision, state statistics

中图分类号: