铁路视频序列的FOE的估计

doi:10.11896／j.issn.1002-137X.2018.07.039

计算机科学 ›› 2018, Vol. 45 ›› Issue (7): 226-229.doi: 10.11896／j.issn.1002-137X.2018.07.039

铁路视频序列的FOE的估计

胡燕花,唐鹏,金炜东,何正伟

西南交通大学电气学院成都610031

收稿日期:2017-05-04 出版日期:2018-07-30 发布日期:2018-07-30
作者简介:胡燕花(1991-),女,硕士生,主要研究方向为智能信息处理、图像处理,E-mail:hu_yanhua@126.com;唐鹏(1979-),男,博士,讲师,主要研究方向为图像处理;金炜东(1959-),男,博士,教授,主要研究方向为智能信息处理、模式识别,E-mail:wdjin@home.swjtu.edu.cn(通信作者);何正伟(1994-),男,硕士生,主要研究方向为图像处理。
基金资助:
本文受国家重点自然科学基金项目(61134002),中央高校基本科研业务费创新项目(2682014CX027),国家重点研发计划(2016YFB1200401-102F)资助。

Estimation of FOE for Railway Video Sequences

HU Yan-hua ,TANG Peng ,JIN Wei-dong, HE Zheng-wei

College of Electrical Engineering,Southwest Jiaotong University,Chengdu 610031,China

Received:2017-05-04 Online:2018-07-30 Published:2018-07-30

摘要/Abstract

摘要： 在结构化场景的轨道交通中,车载视频观测因相机平移运动而呈现出图像内容以某点为中心向四周扩散的现象,该点被称为FOE (Focus of Expansion)。当前计算FOE的算法对噪声敏感且计算量大,不能准确地计算铁路场景中的FOE。鉴于此,文中提出一种铁路视频序列的FOE估计方法。该方法首先利用金字塔光流法对检测的Harris角点进行跟踪和粗匹配,并在此基础上利用RANSAC算法进行精确的匹配,求得基础矩阵,然后提取图像中的极线束并计算FOE。实验结果表明,所提算法比Hough 直线求得的FOE误差小,适于实时应用。

关键词: FOE, RANSAC算法, 基础矩阵, 极线束, 金字塔光流法

Abstract: In the rail transit of the structured scene,due to the movement of camera,the objects in the image captured by the on-board camera will spread around the center of this image,which is called FOE (Focus of Expansion).In view of the current technology based on FOE,which is sensitive to noise and has a large amount of computation,it can not accurately calculate the FOE in the railway scene.This paper presented a method for estimating the FOE of railway video sequences.This method uses the Pyramid optical flow method to track and coarsely match the detected Harris corner points,and makes accurately matching with RANSAC algorithm based on the computation of fundamental matrix.Then the epipolar lines are extracted in the image,and the FOE is obtained at last.The experimental results show that the FOE error of this algorithm is smaller than that of the Hough line,and the proposed algorithm is suitable for real-time application.

Key words: Epipolar harness, FOE, Fundamental matrix, Pyramid flow algorithm, RANSAC algorithm

中图分类号:

U283.1

胡燕花,唐鹏,金炜东,何正伟. 铁路视频序列的FOE的估计[J]. 计算机科学, 2018, 45(7): 226-229. https://doi.org/10.11896／j.issn.1002-137X.2018.07.039

HU Yan-hua ,TANG Peng ,JIN Wei-dong, HE Zheng-wei. Estimation of FOE for Railway Video Sequences[J]. Computer Science, 2018, 45(7): 226-229. https://doi.org/10.11896／j.issn.1002-137X.2018.07.039

参考文献

[1]JAIN R.Direct computation of the focus of expansion[J].IEEE Trans Pattern Anal Mach Intell,1983,5(1):58-64.
[2]LI Q,REN S.A Real-Time Visual Inspection System for DisCreteSurface Defects of Rail Heads[C]∥IEEE Instrumentation and Easurement.2012:1-11.
[3]DUAN W J,ZHAO W,HUANG S L,et al.Automatic measurement method of the inclination of the contact net positioner based on computer vision.China Railway Science,2011,32(4):82-87.(in Chinese)
段汝娇,赵伟,黄松岭,等.基于计算机视觉的接触网定位器倾斜度自动测量方法[J].中国铁道科学,2011,32(4):82-87.
[4]WANG J K.Train wheelset detection system based on CCD ima-ging technology .Dalian:Dalian Jiaotong University,2007.(in Chinese)
王君可.基于CCD成像技术的火车轮对检测系统[D].大连:大连交通大学,2007.
[5]ZHANGL,KOCH R.Vanishing points estimation and line classification in a manhattanworld[M]∥Computer Vision-ACCV 2012.Springer Berlin Heidelberg,2013:38-51.
[6]GEIGER A,LENZ P,URTASUN R.Are we ready for Autonomous Driving?The KITTI Vision Benchmark Suite[C]∥Conference on Computer Vision and Pattern Recognition (CVPR).2012.
[7]BADINO H,HUBER D,KANADE T.The CMU Visual Localization Data Set[OL].http://3dvis.ri.cmu.edu/data-sets/locali-zation.
[8]DUAN W W.Detection and abnormal recognition of contact network equipment for surveillance video .Chengdu:Southwest Jiaotong University,2016.(in Chinese)
段旺旺.面向监控视频的接触网设备检测及异常识别[D].成都:西南交通大学,2016.
[9]HE Z M.Fast detection of vanishing points of the perspective text image .Shanghai University of Engineering and Technology,2012,23(2):229-232.(in Chinese)
贺志明.透视文本图像的灭点快速探测[J].上海工程技术大学报,2012,23(3):229-232.
[10]XIE W H,ZHANG Z X,ZHANG J Q.A new camera calibration method based on vanishing points.Journal of Harbin Institute of Technology,2003,35(11):1384-1387.(in Chinese)
谢文寒,张祖勋,张剑清.一种新的基于灭点的相机标定方法[J].哈尔滨工业大学学报,2003,35(11):1384-1387.
[11]HARRIS C,STEPHENS M.A combined corner and edge detector[C]∥Alvey Vision Conference.Manchester,UK,1988:147-151.
[12]LUCAS B D,KANADE T.An iterative image registration technique with an application to stereo vision[C]∥Proceedings of the 7th International Joint Conference on Artificial Intelligence.1981:674-679.
[13]JIANG Z J,YI H R.A feature tracking method based on image pyramid optical flow.Journal of Wuhan University(Information Science Edition),2007,32(8):680-683.(in Chinese)
奖志军,易华蓉.一种基于图像金字塔光流的特征跟踪方法[J].武汉大学学报(信息科学版),2007,32(8):680-683.
[14]HE M Y,DAI Y C.Three-Dimensional Measurement Algorithm for Uncalibrated Multi-view Images.Microelectronics & Computer,2010,27(9):181-185.(in Chinese)
何明一,戴玉超.多视角未标定图像三维测量算法[J].微电子学与计算机,2010,27(9):181-185.
[15]LIANGJ J,QIN A K,SUGANTHAN P N,et al.Comprehen-sive learning particle swarm optimizer for global optimization of multimodal functions[J].IEEE Transactionson Evolutionary Computation,2006,10(3):281-295.
[16]HU Z Y,WU F C.Calibration Method Based on Active Vision Camera.Chinese Journal of Computers,2002,25(11):1149-1156.(in Chinese)
胡占义,吴福朝.基于主动视觉摄像机标定方法[J].计算机学报,2002,25(11):1149-1156.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

铁路视频序列的FOE的估计

Estimation of FOE for Railway Video Sequences

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

Metrics

本文评价

推荐阅读 0

[1]	邢文博, 杜志淳. 数字图像复制粘贴篡改取证 Digital Image Forensics for Copy and Paste Tampering 计算机科学, 2019, 46(6A): 380-384.
[2]	邵进达, 杨帅, 程琳. 改进SIFT算法结合两级特征匹配的无人机图像匹配算法 UAV Image Matching Algorithm Based on Improved SIFT Algorithm and Two-stage Feature Matching 计算机科学, 2019, 46(6): 316-321. https://doi.org/10.11896/j.issn.1002-137X.2019.06.048
[3]	蔡涛,段善旭,李德华. 视觉基础矩阵估计方法的性能比较与分析 Performance Comparison and Analysis of Fundamental Matrix Estimating Methods for Computer Vision Applications 计算机科学, 2009, 36(12): 243-247.