基于深度学习与多级匹配机制的港区人员轨迹提取

doi:10.16097/j.cnki.1009-6744.2023.04.008

交通运输系统工程与信息 ›› 2023, Vol. 23 ›› Issue (4): 70-79.DOI: 10.16097/j.cnki.1009-6744.2023.04.008

• 智能交通系统与信息技术 • 上一篇下一篇

基于深度学习与多级匹配机制的港区人员轨迹提取

陈信强^1a，王美琳^1a，李朝锋^1a，杨洋^*2，梅骁峻^1b，周亚民^1a

1. 上海海事大学，a. 物流科学与工程研究院，b. 信息工程学院，上海 201306； 2. 北京航空航天大学，交通科学与工程学院，北京 100191

收稿日期:2022-11-28 修回日期:2023-02-13 接受日期:2023-03-29 出版日期:2023-08-25 发布日期:2023-08-21
作者简介:陈信强(1987- )，男，江西南昌人，副教授
基金资助:
国家自然科学基金(52102397)；中国博士后科学基金(2022M712027)；上海市科学技术委员会重点项目(23010502000)

Port Staff Trajectory Extraction Based on Deep Learning and Multi-level Matching Mechanism

CHEN Xin-qiang^1a, WANG Mei-lin^1a, LI Chao-feng^1a, YANG Yang^*2, MEI Xiao-jun^1b, ZHOU Ya-min^1a

1a. Institute of Logistics Science and Engineering, 1b. College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China; 2. School of Transportation Science and Engineering, Beihang University, Beijing 100191, China

Received:2022-11-28 Revised:2023-02-13 Accepted:2023-03-29 Online:2023-08-25 Published:2023-08-21
Supported by:
National Natural Science Foundation of China (52102397)；China Postdoctoral Science Foundation (2022M712027)；The Shanghai Committee of Science and Technology, China (23010502000)

摘要/Abstract

摘要： 针对港口环境空间布局复杂，集装箱堆场、起重机械、装卸运输设备等复杂背景干扰下港区工作人员难以被准确跟踪的问题，本文面向港口监控视频提出一种基于 Faster-RCNN(Faster Region Convolutional Neural Networks) 检测算法和改进 Deep SORT(Deep Simple Online and Realtime Tracking)跟踪算法的港区工作人员轨迹提取框架(FRIMDS)。本框架加入自适应高斯降噪和直方图均衡化算法，融合图像增强技术和行人重识别网络(Person Re-identification，ReID)提取港航图像特征信息，以提高港区工作人员轨迹提取的快速性和准确度。通过前置特征提取网络、候选区域建议网络、感兴趣区域池化和全连接层联合输出港区工作人员图像序列检测结果，采用级联匹配和匈牙利算法匹配港区工作人员位置信息，最后利用卡尔曼滤波预测得到港区工作人员运动轨迹。结果显示，本文所提方法在各典型港口场景中面对不同光照变化、低能见度、阴影干扰等挑战均表现出良好的性能，EIDF1 、EIDR 、ERCLL 、EMOTA 指标平均值分别为98％、97％、97％、95％。结论表明，本文提出的FRIMDS框架具有一定的精确性和稳定性，可为自动化码头安全监管提供技术支撑。

关键词: 交通工程, 自动化码头, Faster-RCNN算法, Deep SORT跟踪算法, 港区工作人员轨迹

Abstract: Due to the complex spatial layout of the port environment, the difficulty of accurate tracking of port staff exists under the interference of complex backgrounds such as container yards, lifting machinery, loading, unloading, and transportation equipment. This study proposes a trajectory extraction framework based on a Faster-RCNN detection algorithm and an improved Deep SORT tracking algorithm for port surveillance video. In this framework, an adaptive Gaussian noise reduction and histogram equalization algorithm were added, and the image enhancement technology and Person Re-identification network were integrated to extract the feature information of port images, to improve the rapidity and accuracy of the track extraction of port staff. The detection results of the port staff image sequence were output through the pre-feature extraction network, the candidate region suggestion network, the pool of interest area, and the full connection layer. The location information of port staff was matched by cascade matching and the Hungarian algorithm. Finally, the motion trajectory of port staff was predicted by the Kalman filter. The results show that the proposed method has good performance in the face of challenges such as different light changes, low visibility, and shadow interference in each typical port scene. The average values of EIDF1 , EIDR , ERCLL , and EMOTA are 98%, 97%, 97%, and 95%, respectively. The conclusion shows that the FRIMDS framework proposed in this study has certain accuracy and stability, and can provide technical support for the safety supervision of automated terminals.

Key words: traffic engineering, automatic terminal, Faster-RCNN algorithm, Deep SORT tracking algorithm, track of port staff

中图分类号:

陈信强, 王美琳, 李朝锋, 杨洋, 梅骁峻, 周亚民. 基于深度学习与多级匹配机制的港区人员轨迹提取[J]. 交通运输系统工程与信息, 2023, 23(4): 70-79.

CHEN Xin-qiang, WANG Mei-lin, LI Chao-feng, YANG Yang, MEI Xiao-jun, ZHOU Ya-min. Port Staff Trajectory Extraction Based on Deep Learning and Multi-level Matching Mechanism[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(4): 70-79.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2023.04.008

http://www.tseit.org.cn/CN/Y2023/V23/I4/70

参考文献

[1] 戈艳艳, 王姗姗. 考虑碳排放的港口全要素生产率及影响因素分析[J]. 交通运输系统工程与信息, 2021, 21 (2): 22-29. [GE Y Y, WANG S S. Analysis of port total factor productivity and influencing factors considering carbon emissions[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(2): 22-29.]

[2] 张永锋, 龚建伟, 殷明. 新冠肺炎疫情对中国港航业的影响及其对策[J]. 交通运输工程学报, 2020, 20(3): 159-167. [ZHANG Y F, GONG J W, YIN M. The impact of the new crown pneumonia epidemic on China's port and shipping industry and its countermeasures[J]. Journal of Traffic and Transportation Engineering, 2020, 20(3): 159-167.]

[3] 游峰, 梁健中, 曹水金, 等. 面向多目标跟踪的密集行人群轨迹提取和运动语义感知[J]. 交通运输系统工程与信息, 2021, 21(6): 42-54. [YOU F, LIANG J Z, CAO S J, et al. Trajectory extraction and motion semantic perception of dense rows for multi-object tracking[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(6): 42-54.]

[4] WANG Z H, LI M, LU Y, et al. Effective multiple pedestrian tracking system in video surveillance with monocular stationary camera[J]. Expert Systems With Applications, 2021, 178(2021): 114992.

[5] ZHU Y P, WANG T, ZHU S Q. Adaptive multi�pedestrian tracking by multi-sensor: Track-to-track fusion using monocular 3D detection and MMW radar[J]. Remote Sensing, 2022, 14(8): 1837.

[6] FU T, HU W C, MIRANDA-MORENO L, et al. Investigating secondary pedestrian-vehicle interactions at non-signalized intersections using vision-based trajectory data[J]. Transportation Research Part C: Emerging Technologies, 2019, 105(2019): 222-240.

[7] CHEN X Q, LING J, WANG S Z, et al. Ship detection from coastal surveillance videos via an ensemble Canny-Gaussian-morphology framework[J]. Journal of Navigation, 2021, 74(6): 1252-1266.

[8] YANG L, HU G H, SONG Y H, et al. Intelligent video analysis: A Pedestrian trajectory extraction method for the whole indoor space without blind areas[J]. Computer Vision and Image Understanding, 2020, 196(2020): 102968.

[9] 韩明, 王景芹, 王敬涛, 等. 级联特征融合孪生网络目标跟踪算法研究[J]. 计算机工程与应用, 2022, 58(6): 208- 218. [HAN M, WANG J Q, WANG J T, et al. Research on target tracking algorithm of cascading feature fusion twin network[J]. Computer Engineering and Applications, 2022, 58(6): 208-218.]

[10] PANETTA K, KEZEBOU L, OLUDARE V, et al. Comprehensive underwater object tracking benchmark dataset and underwater image enhancement with GAN[J]. IEEE Journal of Oceanic Engineering, 2021, 47(1): 59- 75.

[11] PRABHU B V B, LAKSHMI R, ANKITHA R, et al. RescueNet: YOLO-based object detection model for detection and counting of flood survivors[J]. Modeling Earth Systems and Environment, 2022, 8(4): 1-8.

[12] DURVE M, TIRIBOCCHI A, BONACCORSO F, et al. DropTrack: Automatic droplet tracking with YOLOv5 and Deep SORT for microfluidic applications[J]. Physics of Fluids, 2022, 34(8): 082003.

[13] XING Z Y, ZHANG Z Y, YAO X W, et al. Rail wheel tread defect detection using improved YOLOv3[J]. Measurement, 2022, 203(2022): 111959.

[14] LI Z Y, JIANG X Q, SHUAI L Y, et al. A Real-time detection algorithm for sweet cherry fruit maturity based on YOLOX in the natural environment[J]. Agronomy, 2022, 12(10): 2482.

[15] ZHAO S, QIAO K, YU W D, et al. D2ADet: Dual-activation domain RetinaNet based on a deformable feature pyramid network for ship detection in SAR images [J]. Remote Sensing Letters, 2022, 13(11): 1153-1163.

基于深度学习与多级匹配机制的港区人员轨迹提取

Port Staff Trajectory Extraction Based on Deep Learning and Multi-level Matching Mechanism

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	张铎, 饶红玉, 刘佳琦, 王俊骅, 孙剑. 基于自然驾驶数据的驾驶人跟驰行为内在异质性预测与建模[J]. 交通运输系统工程与信息, 2023, 23(5): 33-44.
[2]	高云峰, 席建伟, 孙科. 面向客货分离交叉口的网联货车队列车速引导与信号优先组合优化[J]. 交通运输系统工程与信息, 2023, 23(4): 88-101.
[3]	管德永, 徐越, 王可. 基于动态灵敏度的感应控制优化方法[J]. 交通运输系统工程与信息, 2023, 23(4): 124-133.
[4]	柏强, 江瑛, 李毅帆, 徐誌蔓. 基于改进时空轨迹理论的机场出发层车道边通行能力评估方法研究[J]. 交通运输系统工程与信息, 2023, 23(4): 228-236.
[5]	郝嘉田, 吴忠广, 田万利, 韩峰, 刘博文. 高速公路改扩建中央分隔带开口长度通用计算模型构建与验证[J]. 交通运输系统工程与信息, 2023, 23(4): 262-269.
[6]	秦严严, 肖腾飞, 贺正冰. 雾天场景下高速公路通行能力分析及提升策略[J]. 交通运输系统工程与信息, 2023, 23(3): 39-47.
[7]	王殿海, 黄莉莎, 曾佳棋, 蔡正义. 降雨对城市路网交通状态的影响机理[J]. 交通运输系统工程与信息, 2023, 23(3): 48-55.
[8]	葛慧敏, 臧文铠, 董磊, 周礼军. 大型多环道涡轮环交车道级信号控制方法[J]. 交通运输系统工程与信息, 2023, 23(3): 123-132.
[9]	付锐, 许清津, 葛振振, 陈桃, 吴付威, 袁伟. 弯道路侧标志梯度提示策略研究[J]. 交通运输系统工程与信息, 2023, 23(3): 133-144.
[10]	李思贤, 邢增勇, 钱大琳, 李鹏程, 袁梦. 危险货物道路运输车辆驾驶人风险倾向性分类研究[J]. 交通运输系统工程与信息, 2023, 23(3): 161-173.
[11]	杨洋, 贺昆, 王云鹏, 陈垚, 袁振洲. 面向动态交通流的高速公路事故风险模型空间移植研究[J]. 交通运输系统工程与信息, 2023, 23(3): 174-186.
[12]	康柳江, 李浩, 孙会君, 吴建军. 复杂山区工程建设物资运输无人机巡航模型构建与实证研究[J]. 交通运输系统工程与信息, 2023, 23(3): 290-299.
[13]	钱国敏, 任璐, 季青原, 冯远静, 章立辉, 王殿海. 考虑车流通行效率的混行交叉口时空资源配置[J]. 交通运输系统工程与信息, 2023, 23(2): 67-73.
[14]	刘志伟, 刘建荣. 数字鸿沟背景下老年人对自动驾驶汽车接受度[J]. 交通运输系统工程与信息, 2023, 23(2): 168-175.
[15]	唐铁桥, 钟静燃, 袁小婷. 考虑社交距离的火车站检票过程建模研究[J]. 交通运输系统工程与信息, 2023, 23(2): 225-232.