基于Hough和Camshift的铁轨异物入侵检测
1. 大连交通大学, 计算机与通信工程学院电子与通信工程系, 辽宁大连 116028
2. 大连交通大学, 自动化与电气工程学院铁道信号与控制工程系, 辽宁大连 116028
| 摘 要： | 随着中国铁路的飞速发展，铁路运输总量得到了飞速增加和铁路运行速度也得到的快速提升，这对整个铁路网的可靠性和安全性提出了更高的要求。异物侵入铁路对铁路系统可靠性和安全性造成了极大的威胁，为了有效避免此类现象的发生，本文基于图像处理的原理，给出了一种铁轨异物入侵检测的方法。首先利用车头的摄像头从列车前方获取视频数据，并运用灰度化、中值滤波和边缘处理的图像处理技术将其转换为清晰的图像，以更加准确地展示铁轨的位置。使用Otsu方法进行图像的阈值分割，利用Hough变换去除残杂线条，突出轨线条轮廓。使用OpenCV库中背景减除法对运动目标进行检测，Camshift算法对图像跟踪，并根据被追踪的目标与铁轨的重合程度，判断出是否有可疑物体入侵铁轨。GUI视图给出了被跟踪目标区域和铁轨区域的重叠比例，判断物体是否对行车造成影响。仿真验证了本文方法的有效性。 |
| 关 键 词： | 铁轨异物; 霍夫变换; Otsu方法; 目标跟踪; Camshift算法 |
| DOI： | 10.57237/j.jsts.2024.01.004 |
View:
Downloads:
收稿日期: 2024-01-01
接受日期: 2024-02-05
在线出版日期: 2024-03-11
Intrusion Detection of Foreign Objects in Railway Tracks Based on Hough and Camshift
1. Department of Electronic and Communication Engineering, School of Computer and Communication Engineering, Dalian Jiaotong University, Dalian 116028, China
2. Department of Railway Signal and Control Engineering, School of Automation and Electrical Engineering, Dalian Jiaotong University, Dalian 116028, China
| Abstract: | With the rapid development of China's railways, the total amount of railway transportation has increased rapidly and the operating speed of railways has also been rapidly improved, which puts higher requirements on the reliability and safety of the entire railway network. Foreign object intrusion into railways poses a great threat to the reliability and safety of railway systems. In order to effectively avoid such phenomena, this article proposes a method for detecting foreign object intrusion in railway tracks based on the principle of image processing. Firstly, video data is obtained from the front of the train using the front camera of the train, and image processing techniques such as grayscale, median filtering, and edge processing are used to convert it into clear images to more accurately display the position of the railway tracks. The Otsu method is used for image threshold segmentation, and the Hough transform is used to remove residual impurities line and highlight the outline of the rail line. The background subtraction method in OpenCV library is used for detecting the moving targets, and the Camshift algorithm is used to track images, and determine whether suspicious objects have invaded the rail based on the degree of overlap between the tracked target and the rail. The GUI view provides the overlap ratio between the tracked target area and the rail area to determine whether the object has affected the driving. The effectiveness of the method proposed in this paper was verified through simulation. |
| Keywords: | Foreign Objects on Railway Tracks; Hough Transform; Otsu Method; Target Tracking; Camshift Algorithm |
基金
| 1. | 大连市重大科市重点科技局研发计划 (2022YF11GX008) |
参考文献
| [1] | 胡行涛, 刘大明, 虞发桐. 基于FCN不确定性特征的铁路入侵异物检测算法 [J]. 计算机应用与软件, 2023, 40(04): 141-146. |
| [2] | 孟彩霞, 王兆楠, 石磊等. 改进YOLOv5s的铁路异物入侵检测算法 [J/OL]. 小型微型计算机系统: 1-10 (2023-12-06). |
| [3] | 管岭, 贾利民, 谢征宇. 融合注意力机制的轨道入侵异物检测轻量级模型研究 [J]. 铁道学报, 2023, 45(05): 72-81. |
| [4] | 王文杰. 铁路周界异物侵入的机器视觉智能检测技术 [D]. 四川: 西南交通大学, 2022. |
| [5] | 赵文钰. 高速铁路周界入侵多尺度目标检测算法研究 [D]. 北京: 北京交通大学, 2023. |
| [6] | Zheng S Y, Jin W Y, Dong Y. Rail Detection Based on LSD and the Least Square Curve Fitting [J]. International Journal of Automation and Computing, 2020, 18(1). |
| [7] | 董昱, 郭碧. 基于Hu不变矩特征的铁路轨道识别检测算法[J]. 铁道学报, 2018, 40(10): 64-70. |
| [8] | 张丽艳, 韩熠, 赵艺璇等. 霍夫变换和Faster-RCNN铁轨上行人检测方法 [J]. 大连交通大学学报, 2020, 41(02): 100-103. |
| [9] | 王文豪, 高利. 一种基于OpenCV的车道线检测方法 [J]. 激光杂志, 2019, 40(01): 44-47. |
| [10] | 侯涛, 宝才文, 陈燕楠. 基于自适应高斯混合模型的铁轨异物入侵检测研究 [J]. 光电子•激光, 2022, 33(04): 403-413. |
| [11] | 牛晗, 伍希志, 任桂芹等.基于OTSU与CANNY算法的竹片缺陷图像检测 [J]. 森林工程, 2022, 38(06): 75-81. |
| [12] | 刘萌雅, 张丽艳, 费继友. 基于视觉的道路信息识别技术研究 [J]. 大连交通大学学报, 2016, 37(03): 103-106+111. |
| [13] | 廖逍, 张弛, 应国德等.基于背景差分法的电网巡检运动目标检测技术 [J]. 沈阳工业大学学报, 2023, 45(01): 78-83. |
| [14] | 李文举, 王子杰, 崔柳. 基于SIFT和感知哈希改进的CamShift跟踪算法 [J]. 电子测量技术, 2023, 46(04): 184-192. |
| [15] | 敖邦乾, 杨莎, 刘小雍等. 基于改进CamShift的目标跟踪系统算法 [J]. 探测与控制学报, 2021, 43(06): 92-100. |
| [16] | 邱晓欢, 郑尚坡, 刘俊峰等.基于改进Kalman滤波与Camshift结合的嵌入式跟踪系统设计 [J]. 计算机时代, 2023(11): 41-45+51. |
| [17] | 凌晨昱, 李文其, 余雷. 基于彩色图像与深度图像信息融合的改进Camshift算法研究 [J]. 电工技术, 2021(18): 33-35. |
