Space-Time Distance Calculation Method of Tram Obstacle Based on Instance Segmentation and Monocular Vision

doi:10.19678/j.issn.1000-3428.0068996

Abstract

Abstract:

Trams, owing to their operation in a shared right-of-way and reliance on driver visual observation, are more prone to collisions with intruding obstacles than urban rail transit systems, such as subways and maglev trains. Therefore, to ensure the operational safety of trams, a method for calculating the space-time distance of obstacles based on instance segmentation and monocular vision is proposed. First, the contour points of the obstacles and track area are extracted using an instance segmentation model. Subsequently, a monocular vision ranging model is established based on the principles of monocular vision. By incorporating a standard gauge length of 1.435 m as prior knowledge, the longitudinal distance between the obstacles and tram is calculated without camera calibration. Finally, the lateral distance between the obstacle and track area is calculated based on the point on the obstacle closest to the track area and the corresponding track endpoint. This method fills a research gap in the field of rail transit by calculating the space-time distance of obstacles using the standard gauge length of trams as prior knowledge. Additionally, by introducing an instance segmentation model, the key points for obstacle distance measurement are determined with pixel-level accuracy, enabling the precise calculation of the space-time distance of obstacles. The feasibility of the proposed method is verified using experimental data captured in real-world scenarios. The results show that the maximum positive and negative errors of longitudinal distance calculation are 1.60 m and 1.05 m, respectively, indicating a high level of accuracy in the distance calculation results.

Key words: tram, instance segmentation, monocular vision, ranging model, obstacle

摘要：

有轨电车因采用共享路权和司机目视行车的方式运行, 与地铁、磁悬浮等城市轨道交通相比, 更容易和入侵障碍物发生碰撞。因此, 为了保障有轨电车的运行安全, 提出一种基于实例分割和单目视觉的有轨电车障碍物时空距离计算方法。首先基于实例分割模型提取出障碍物和轨行区的轮廓点; 然后基于单目视觉原理建立单目视觉测距模型, 在引入有轨电车轨道标准轨距长度1.435 m作为先验知识后, 实现在相机无标定情况下障碍物与列车的纵向距离计算; 最后根据障碍物距离轨行区最近的点及对应的轨道端点计算障碍物与轨行区的横向距离。该方法通过引入有轨电车标准轨距长度作为先验知识计算障碍物的时空距离, 填补了轨道交通领域障碍物时空距离计算研究的空白, 并且通过引入实例分割模型, 以像素级精度确定障碍物测距关键点, 实现了障碍物时空距离的精准计算。通过从现实场景中拍摄的实验数据来验证所提方法的可行性, 实验结果表明, 该方法的纵向距离计算结果的最大正误差为1.60 m, 最大负误差为1.05 m, 距离计算结果具有较高的准确度。

关键词: 有轨电车, 实例分割, 单目视觉, 测距模型, 障碍物

HUANG Shize, QIN Jinzhe, TAO Ting, DONG Decun, SHEN Tuo. Space-Time Distance Calculation Method of Tram Obstacle Based on Instance Segmentation and Monocular Vision[J]. Computer Engineering, 2025, 51(4): 239-248.

黄世泽, 秦晋哲, 陶婷, 董德存, 沈拓. 基于实例分割和单目视觉的有轨电车障碍物时空距离计算方法[J]. 计算机工程, 2025, 51(4): 239-248.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0068996

https://www.ecice06.com/EN/Y2025/V51/I4/239

Figures/Tables 16

Fig.1 Overall technical roadmap of tram obstacle space-time distance calculation method based on instance segmentation and monocular vision

Fig.2 Schematic diagram of longitudinal distance and transverse distance

Fig.3 Schematic diagram of world coordinate system, camera coordinate system, image coordinate system, and pixel coordinate system

Fig.4 Schematic diagram of transformation from camera coordinate system to image coordinate system

Fig.5 Schematic diagram of transformation from image coordinate system to pixel coordinate system

Fig.6 Schematic diagram of world coordinate system transformation

Fig.7 Coordinate correspondence of rail endpoints

Fig.8 Longitudinal distance solving process

Fig.9 Coordinate correspondence between obstacle contour points and track contour points

Fig.10 Calculation process of lateral distance between obstacle and rail area

Fig.11 Industrial camera and lens

Fig.12 Experimental scene

Fig.13 Pedestrian and bicycle turn back on the left side of the rail area

Fig.14 Distance calculation results

References 25

1	马浩越. 前方车辆主动防撞预警系统关键技术研究[D]. 西安: 长安大学, 2019.
	MA H Y. Research on key technologies of active anti-collision warning system for front vehicles[D]. Xi'an: Changan University, 2019. (in Chinese)
2	刘童. 基于深度学习的ADAS智能图像识别测距技术研究[D]. 杭州: 杭州电子科技大学, 2019.
	LIU T. Research on ADAS intelligent image recognition and ranging technology based on deep learning[D]. Hangzhou: Hangzhou Dianzi University, 2019. (in Chinese)
3	CHENG G Y, GUO Y B, CHENG X C, et al. Real-time detection of vehicle speed based on video image[C]//Proceedings of the 12th International Conference on Measuring Technology and Mechatronics Automation. Washington D.C., USA: IEEE Press, 2020: 313-317.
4	BANERJEE S, DHAR M, SEN S. A novel technique to detect the number of ground vehicles along with respective speed of each vehicle from a given video[C]//Proceedings of the Emerging Trends in Electronic Devices and Computational Techniques. Washington D.C., USA: IEEE Press, 2018: 1-6.
5	BUI M T, DOSKOCIL R, KŘIVÁNEK V. Distance and angle measurement using monocular vision[EB/OL]. [2023-08-05]. https://ieeexplore.ieee.org/document/8624694.
6	马科伟, 张锲石, 康宇航, 等. 移动机器人中视觉里程计技术综述. 计算机工程, 2021, 47 (11): 1- 10. doi: 10.19678/j.issn.1000-3428.0060829
	MA K W , ZHANG Q S , KANG Y H , et al. Overview of visual odometry technology in mobile robots. Computer Engineering, 2021, 47 (11): 1- 10. doi: 10.19678/j.issn.1000-3428.0060829
7	毕嘉桢, 沈拓, 张轩雄. 基于机器视觉的轨道交通自动测距研究. 电子科技, 2022, 35 (9): 37- 43.
	BI J Z , SHEN T , ZHANG X X . A research on distance measurement between trains in rail transit based on machine vision. Electronic Science and Technology, 2022, 35 (9): 37- 43.
8	XU Z G, WANG L Y, WANG J. A method for distance measurement of moving objects in a monocular image[C]//Proceedings of the 3rd IEEE International Conference on Signal and Image Processing. Washington D.C., USA: IEEE Press, 2018: 245-249.
9	XU Z G, ZHOU Z H, WANG L Y. Research on monocular vision distance measurement algorithm based on reference target[C]// Proceedings of the 4th IEEE International Conference on Signal and Image Processing. Washington D.C., USA: IEEE Press, 2019: 192-197.
10	XU Z G, WANG L Y, WANG J. A method for distance measurement of moving objects in a monocular image[C]//Proceedings of the 3rd IEEE International Conference on Signal and Image Processing. Washington D.C., USA: IEEE Press, 2018: 102-110.
11	王志刚, 赵海良, 王星. 基于图像识别的模糊测距方法. 重庆理工大学学报(自然科学), 2020, 34 (1): 58- 63.
	WANG Z G , ZHAO H L , WANG X . Fuzzy ranging method based on image recognition. Journal of Chongqing University of Technology (Natural Science), 2020, 34 (1): 58- 63.
12	桑振, 周永华. 基于视频的移动车辆实时测距研究. 北京交通大学学报, 2020, 44 (5): 133- 140.
	SANG Z , ZHOU Y H . A video-based research on real-time distance measurement of mobile vehicles. Journal of Beijing Jiaotong University, 2020, 44 (5): 133- 140.
13	张欣, 魏志恒, 王兵. 单目视觉的前车位置识别及安全车距监测. 河北大学学报(自然科学版), 2022, 42 (3): 321- 326.
	ZHANG X , WEI Z H , WANG B . Position recognition of the vehicle ahead and safety distance monitoring based on monocular vision. Journal of Hebei University (Natural Science Edition), 2022, 42 (3): 321- 326.
14	沈志熙, 黄席樾. 基于数据回归建模的单目视觉测距算法. 计算机工程与应用, 2007, 43 (24): 15-18, 40.
	SHEN Z X , HUANG X Y . Monocular vision distance detection algorithm based on data regression modeling. Computer Engineering and Applications, 2007, 43 (24): 15-18, 40.
15	罗时光. 基于单目测距技术的道路障碍物检测方法. 软件, 2013, 34 (9): 100-102, 106.
	LUO S G . Method of obstacle detection for the road based on monocular measurement. Software, 2013, 34 (9): 100-102, 106.
16	ALI A A, HUSSEIN H A. Distance estimation and vehicle position detection based on monocular camera[EB/OL]. [2023-08-05]. https://ieeexplore.ieee.org/document/7759904.
17	高秀龙. 基于目标检测与单目视觉的车载智能测距算法研究[D]. 柳州: 广西科技大学, 2020.
	GAO X L. Research on vehicle intelligent ranging algorithm based on target detection and monocular vision[D]. Liuzhou: Guangxi University of Science and Technology, 2020. (in Chinese)
18	刘庆淼. 基于图像识别的物流车辆行驶中车距预警系统研究[D]. 济南: 山东财经大学, 2020.
	LIU Q M. Research on distance warning system of logistics vehicles based on image recognition[D]. Jinan: Shandong University of Finance and Economics, 2020. (in Chinese)
19	刘军, 后士浩, 张凯, 等. 基于单目视觉车辆姿态角估计和逆透视变换的车距测量. 农业工程学报, 2018, 34 (13): 70- 76.
	LIU J , HOU S H , ZHANG K , et al. Vehicle distance measurement with implementation of vehicle attitude angle estimation and inverse perspective mapping based on monocular vision. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34 (13): 70- 76.
20	王永森, 刘宏哲. 基于单目视觉的车辆下边沿估计和逆透视变换的车距测量. 计算机工程与科学, 2020, 42 (7): 1234- 1243.
	WANG Y S , LIU H Z . Vehicle distance measurement with vehicle lower edge estimation and inverse perspective mapping based on monocular vision. Computer Engineering & Science, 2020, 42 (7): 1234- 1243.
21	YE T , WANG B , SONG P , et al. Automatic railway traffic object detection system using feature fusion refine neural network under shunting mode. Sensors (Basel, Switzerland), 2018, 18 (6): E1916.
22	YE T , ZHANG Z H , ZHANG X , et al. Autonomous railway traffic object detection using feature-enhanced single-shot detector. IEEE Access, 2020, 8, 145182- 145193.
23	黄世泽, 杨玲玉, 陶婷, 等. 基于实例分割的有轨电车障碍物入侵检测及轨道识别方法. 上海公路, 2021 (2): 89-94, 103, 168.
	HUANG S Z , YANG L Y , TAO T , et al. A method of tram obstacle intrusion detection and track recognition based on instance segmentation. Shanghai Highways, 2021 (2): 89-94, 103, 168.
24	BOLYA D, ZHOU C, XIAO F Y, et al. YOLACT: real-time instance segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Washington D.C., USA: IEEE Press, 2019: 9156-9165.
25	高翔, 张涛, 刘毅, 等. 视觉SLAM十四讲: 从理论到实践. 北京: 电子工业出版社, 2017.
	GAO X , ZHANG T , LIU Y , et al. Fourteen lectures on visual SLAM: from theory to practice. Beijing: Publishing House of Electronics Industry, 2017.

[1]	LI Weikang, ZHANG Siquan. Mask Feature Fusion: New Paradigm of Instance Segmentation [J]. Computer Engineering, 2025, 51(2): 126-138.
[2]	ZHAO Nannan, GAO Feichen. Improved YOLOv8-based Algorithm for Instance Segmentation in Traffic Scenes [J]. Computer Engineering, 2025, 51(1): 198-207.
[3]	Mei ZENG, Yihan WANG, Zhiwei LEI, Xueyin LIU, Bailin LI. Grape Instance De-Overlapping Occlusion Algorithm Based on Self-Supervised Learning [J]. Computer Engineering, 2024, 50(8): 216-228.
[4]	ZHOU Qinyuan, DENG Yueping, ZHANG Lei, ZHANG Chen, LU Rirong, HU Xianzhe. Dynamic Visual SLAM System Integrating Optical Flow and Multi-View Geometry [J]. Computer Engineering, 2024, 50(5): 250-259.
[5]	Haijun DU, Su YU. Dynamic Obstacle Avoidance for Service Robots Based on Spatio-Temporal Graph Attention Network [J]. Computer Engineering, 2024, 50(2): 105-112.
[6]	LIU Hangbo, MA Li, LI Yang, MA Dongchao, FU Yingxun. Obstacle Avoidance Method Using DQN to Classify Obstacles in Unmanned Driving [J]. Computer Engineering, 2024, 50(11): 380-389.
[7]	ZHOU Yiyun, WAN Xinjun, HU Fuyuan, CHEN Hao. Instance Segmentation Algorithm Based on Joint Attention and Feature Association [J]. Computer Engineering, 2023, 49(6): 217-226.
[8]	Qiru LI, Xia GENG. Robot Path Planning Based on Improved DQN Algorithm [J]. Computer Engineering, 2023, 49(12): 111-120.
[9]	LIU Teng, LIU Hongzhe, LI Xuewei, XU Cheng. Improved Instance Segmentation Method Based on Anchor-Free Segmentation Network [J]. Computer Engineering, 2022, 48(9): 239-247,253.
[10]	YANG Zexue, WANG Achuan, LI Lu, LI Song. Visible Reverse View Field K-Nearest Neighbor Queries in Obstacle Environment [J]. Computer Engineering, 2022, 48(8): 258-265.
[11]	NING Xiaojuan, GONG Liang, ZHANG Jinlei. Detection Method of Passable Road Areas Based on Laser Point Clouds [J]. Computer Engineering, 2022, 48(4): 22-29.
[12]	WANG Kuan, YANG Huan, PAN Zhenkuan, SI Jianwei. Full-Reference Model for Stereoscopic Image Quality Assessment Based on Monocular and Binocular Visual Information [J]. Computer Engineering, 2022, 48(2): 207-214,223.
[13]	FAN Xinyue, BAO Hong, PAN Weiguo. Image Instance Segmentation Method Based on Class-imbalanced Dataset [J]. Computer Engineering, 2022, 48(12): 224-231.
[14]	YIN Song, CHEN Xueyun, BEI Xueyu. Improved Mask RCNN Algorithm and Its Application in Pedestrian Instance Segmentation [J]. Computer Engineering, 2021, 47(6): 271-276,283.
[15]	WU Haibin, XU Ruotong, WANG Aili, YU Xiaoyang, IWAHORI Yuji, ZHAO Lanfei, LIU He. Overview of 3D Human Lumen Reconstruction Technology Based on Computer Vision [J]. Computer Engineering, 2021, 47(10): 1-15.

Please choose a citation manager

Content to export