WANG Qi,JIN Xiaofeng*.Research on method of vehicle detection and sample acquisition[J].Journal of Yanbian University,2015,41(02):164-169.
车辆识别与样本自动采集方法的研究
- Title:
- Research on method of vehicle detection and sample acquisition
- Keywords:
- vehicle detection and tracking; Blob analysis; CamShift; vehicle sample acquisition; vehicle sample library
- 分类号:
- TP391.41
- 文献标志码:
- A
- 摘要:
- 为解决车辆样本采集困难的问题,在研究车辆识别与跟踪的基础上提出了样本的自动采集方法.首先,采用Blob分析技术从视频流中检测出车辆; 其次,结合Blob和CamShift跟踪算法跟踪运动车辆; 再次,通过分析车辆的运动轨迹判定其停驶状态; 最后,控制云台变焦摄像机获取车辆的细节图像,以此作为车辆的样本.实验结果表明,本文提出的方法实时性高,对车辆的识别与跟踪、停驶判断具有较高的准确性,获取的车辆样本图像细节丰富,能够满足车辆样本库建设的基本要求.
- Abstract:
- In order to solve the problem of vehicle sample collection difficulty, this paper proposed a method of automatic data collection of vehicle samples on the basis of vehicle detection and tracking. Firstly, the Blob analysis technology was employed to detect vehicles from video. Secondly, moving vehicles were tracked by the Blob method and CamShift tracking algorithm. Again, after analyzing the movement of vehicles, the vehicle states of movement and stop were detected. Finally, the detail image obtained by controlling camera pan was used as the vehicle samples. Experimental results show that the proposed method is of high real-time performance and accuracy in detecting and tracking of vehicles and the judgment of vehicles states. Meanwhile, the obtained vehicle sample images with rich details was able to meet the basic requirements of vehicle sample library construction.
参考文献/References:
[1] Sidla O, Paletta L, Lypetskyy Y, et al. Vehicle recognition for highway lane survey[C]//The 7th International IEEE Conference on Intelligent Transportation Systems. Graz, Austria: IEEE, 2004:531-536.
[2] Sun Z, Bebis G, Miller R. Improving the performance of on-road vehicle detection by combining gabor and wavelet features[C]//The 5th International IEEE Conference on Intelligent Transportation Systems. Singapore: IEEE, 2002:130-135.
[3] 文学志,赵英男,郑钰辉,等.车辆识别中样本自动化准备方法研究[J].计算机工程与应用,2010,46(15):25-27.
[4] Khammari A, Nashashibi F, Abramson Y, et al. Vehicle detection combining gradient analysis and adaboost classification[C]//Intelligent Transportation Systems, 2005. IEEE, 2005:1084-1089.
[5] 刘星.基于co-training学习的车辆检测算法及场景自适应分析[D].北京工业大学,2013:66-70.
[6] 王相海,秦钜鳌,方玲玲.基于感兴趣区域AdaBoost分类器的视频车辆检测研究[J].辽宁师范大学学报,2014,37(1):52-62.
[7] Vosters L, Shan C, Gritti T. Real-time robust background subtraction under rapidly changing illumination conditions[J]. Image and Vision Computing, 2012,12(30):1004-1015.
[8] 施小成,郝丽超,张伟,等.基于Blob分析和贝叶斯决策的水下目标提取方法[J].计算机应用,2012,32(11):3214-3217.
[9] Collins R T. Mean shift blob tracking through scale space[C]// The 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Madison, USA, 2003:234-240.
[10] 李伟生,汪钊.基于时空背景模型的自适应运动目标检测方法[J].计算机应用,2014,34(12):3515-3520.
[11] LIU Chao, HUI Jing. Object tracking algorithm based on improved CamShift[J]. Computer Engineering and Applications, 2014,50(11):149-153.
[12] 黄园刚,桑楠,郝宗波,等.改进CamShift算法的眼动跟踪方法[J].计算机应用研究,2014,31(4):1220-1224.
[13] 骆且,丁盼盼,杜宇人.一种基于多特征融合CamShift的运动车辆跟踪方法[J].扬州大学学报:自然科学版,2014,17(3):54-58.
[14] Jazayeri A, Cai H, Zheng J. Vehicle detection and tracking in car video based on motion model [J]. IEEE Transactions on Intelligent Transportation Systems, 2011(2):583-595.
[15] Cheng Y. Mean shift, mode seeking and clustering[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1995,17(8):790-799.
[16] Wu J, Cui Z, Chen J, et al. A survey on video-based vehicle behavior analysis algorithms [J]. Journal of Multimedia, 2012,7(3):223-230.
备注/Memo
收稿日期: 2015-03-25 *通信作者: 金小峰(1972—),男,副教授,研究方向为智能信息处理.基金项目: 吉林省科技厅自然科学基金资助项目(20140101225JC)