WANG Liu-jun,ZHANG Ying.Development of a Machine-vision-based Vehicular Intelligence Patrol System for Railway Track[J].Journal of Chengdu University of Information Technology,2016,(02):185-189.
基于机器视觉的车载式铁路轨道智能巡检系统研究
- Title:
- Development of a Machine-vision-based Vehicular Intelligence Patrol System for Railway Track
- 文章编号:
- 2096-1618(2016)02-0185-05
- Keywords:
- railway transportation; railway track patrol; machine vision; defect detection; support of vector machine
- 分类号:
- TP23
- 文献标志码:
- A
- 摘要:
- 研究一种车载式铁路轨道智能巡检系统。高速运动状态下4台高分辨率摄像机实时采集、存储轨道图像,经预处理、分割定位、特征描述、数据降维及模式分类等方法实现钢轨表面擦伤和掉块、扣件断裂和缺失、轨枕裂纹和档肩掉块、轨下垫板厚度异常等4类主要轨道缺陷的机器识别。在缺陷图像数据库中进行二次识别和人工判别可提高缺陷检出的准确度,有利于提高轨道巡检和养护效率。
- Abstract:
- The study is done on developing a intelligence patrol system based on machine-vision for the inspection of railway track. Images of the railway track are captured and stored simultaneously by four high-resolution video cameras equipped in the patrol system. The four categories of railway track defects; burn and delamination of the wheel-rail, damage and lost of the rail fastening, crack or delamination of the sleepers and the abnormal thickness of rail pad could be recognized by the patrol system through preprocessing, tracking and location, features description, dimensionality reduction, model characterization and classification of the captured images. The combination of secondary recognition by the patrol system from defect image database and manually estimation could significantly increase the accuracy of defect detection and the efficiencies of railway inspection and maintenance.
参考文献/References:
[1] 柴雪松,杨风春,罗林. 轨道视频检查系统[J].中国铁路,2001.6:59-60.
[2] Singh M,Singh S,Jaiswal J,et al. Autonomous Rail Track Inspection Using Vision Based System[C].IEEE International Conference on Computational Intelligence for Homeland Security and Personal Safety. Alexandria: IEEE,2006: 56-59.
[3] Robin Clark. Rail Flaw Detection: Overview and Needs for Future Developments[J].NDT and E International,2004,37(2):111-118.
[4] 张未. 德国RAILCHECK光电式自动化钢轨检测系统在轨道检查车中的应用[J].哈尔滨铁道科技,2001,(4): 3-4.
[5] Stellae, Mazzeo P, Nitti M, et al. Visual recognition of missing fastening elements for railroad maintenance[C].IEEE 5th International Conference on Intelligent T. Singapore, 2002: 3-6.
[6] Mazzeo P L, Nicola A,Ettore S, et al. Visual recognition of hexagonal headed bolts by comparing ICA to wavelets[C].Proceedings of the 2003 IEEE International Symposi. Houston,Texas, 2003: 639-641.
[7] Mazzeo P, Nitti M, Stella E, et al. Visual recognition of fastening bolts for railroad maintennance[J]. Pattern Recognition Letters, 2004,(25): 669-677.
[8] Mazzeo P, Stella E, ANCONA N; et al. Visual detection of hexagonal headed bolts using method of frames and matching pursuit[C].Pattern Recognition and Image Analysis. Second Ibe. Estoril, Portugal, 2005: 277-284.
[9] Ruvo G D, Ruvo P D, MARINO F, et al. A FPGA-based architecture for automatic hexagonal bolts detection in railway maintenance[C].Proceedings of the seventh International Workshop, 2005:1-6.
[10] Marino F, Distante A, Mazzeo P L, et al. A real-time visual inspection system for railway maintenance-automatic hexagonal-headed bolts detection[J]. IEEE Transactions on Systems, Man,and Cybernetics; 2007, 37(3): 418-428.
[11] Marino P F. A GPU-based vision system for real time detection of fasteningelements in railway inspection[C]. International Conference on Image Processing,IC.Cairo,Egypt, 2009: 2333-2336.
[12] 谢凤英,吴叶芬,周世新. 基于互信息的铁路轨枕扣件自动定位算法[J]. 中国体视学与图像分析, 2013, 18(2).
[13] 李永波,李柏林,熊鹰. 基于 HOG 特征的铁路扣件状态检测[J]. 传感器与微系统,2013,32(10):110-113.
[14] 范宏.基于图像的铁路扣件缺陷检测算法研究[D].成都:西南交通大学,2012.
[15] 张洋.巡道车视频处理系统关键技术的研究[D].北京:北京交通大学,2008.
[16] 常江.巡道车视频检测系统的设计及关键技术的实现[D].北京:北京交通大学,2009.
[17] 刘丽,谢毓湘,魏迎梅等. 局部二进制模式方法综述[J]. 中国图象图形学报,2014,19( 12):1696-1720.
备注/Memo
收稿日期:2015-07-30 基金项目:四川省科技支撑计划资助项目(2016GZ0194)