[1]蔡英凤,王海.视觉车辆识别迁移学习算法[J].东南大学学报(自然科学版),2015,45(2):275-280.[doi:10.3969/j.issn.1001-0505.2015.02.015]
 Cai Yingfeng,Wang Hai.Vision based vehicle detection transfer learning algorithm[J].Journal of Southeast University (Natural Science Edition),2015,45(2):275-280.[doi:10.3969/j.issn.1001-0505.2015.02.015]
点击复制

视觉车辆识别迁移学习算法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
45
期数:
2015年第2期
页码:
275-280
栏目:
计算机科学与工程
出版日期:
2015-03-20

文章信息/Info

Title:
Vision based vehicle detection transfer learning algorithm
作者:
蔡英凤1王海2
1江苏大学汽车工程研究院, 镇江 212013; 2江苏大学汽车与交通工程学院, 镇江 212013
Author(s):
Cai Yingfeng1 Wang Hai2
1Research Institute of Automotive Engineering, Jiangsu University, Zhenjiang 212013, China
2School of Automotive and Traffic Engineering, Jiangsu University, Zhenjiang 212013, China
关键词:
车辆识别 迁移学习 样本自标注 概率神经网络 先进驾驶辅助系统
Keywords:
vehicle recognition transfer learning sample self-marking probability neural network advanced driver assistant system
分类号:
TP391.4
DOI:
10.3969/j.issn.1001-0505.2015.02.015
摘要:
针对采用大样本离线训练的车辆识别分类器在新场景中性能显著下降的问题,提出了一种具有样本自标注能力的车辆识别迁移学习算法,并采用概率神经网络(probability neural network, PNN)进行分类器训练.首先,提出一种基于多细节先验信息的样本标注策略,融合复杂度、垂直平面和相对速度等先验信息实现新样本的自动标注;然后,充分利用PNN训练速度快以及增加新样本时只需分类器进行局部更新的特点,将其引入到分类器训练模型中,取代传统机器学习算法中的Adaboost分类器.实验结果表明:该算法在新场景下的新样本标注准确率高达99.76%.通过迁移学习,新场景的车辆识别分类器性能较通用分类器在检测率和误检率指标上均有显著提升.
Abstract:
Existing classifiers for vehicle recognition are mainly trained offline with a large number of samples, of which the performance may decline dramatically in a new scenario. In order to solve the problem, a sample self-marking transfer learning algorithm for vehicle recognition based on the probabilistic neural network(PNN)is proposed. First, a sample self-marking strategy is proposed based on multi-cue prior knowledge including complexity, vertical plane and relative velocity. Then, instead of traditional classifiers such as Adaboost, PNN is used to establish the transfer learning model by utilizing its features such as high architecture flexibility, fast training speed and no retraining requirement when new samples are added. Experimental results demonstrate that this algorithm can mark new samples with high accuracy(99.76%). Besides, new classifier trained in a new scenario with transfer learning performs better in true positive rate and false detection rate than traditional general classifiers.

参考文献/References:

[1] Trivedi M M, Gandhi T, McCall J. Looking-in and looking-out of a vehicle: computer-vision-based enhanced vehicle safety [J]. IEEE Transactions on Intelligent Transportation Systems, 2007, 8(1): 108-120.
[2] Wang H, Zhang W, Cai Y. Design of a road vehicle detection system based on monocular vision [J]. Journal of Southeast University: English Edition, 2011, 27(2): 169-173.
[3] 许庆,高峰,徐国艳.基于Haar特征的前车识别算法[J].汽车工程,2013,35(4):381-384.
  Xu Qing, Gao Feng, Xu Guoyan. An algorithm for front-vehicle detection based on Haar-like feature [J]. Automotive Engineering, 2013, 35(4): 381-384.(in Chinese)
[4] 文学志,方巍,郑钰辉.一种基于类Haar特征和改进AdaBoost分类器的车辆识别算法[J].电子学报,2011,39(5):1121-1126.
  Wen Xuezhi, Fang Wei, Zhen Yuhui. An algorithm based on Haar-like features and improved AdaBoost classifier for vehicle recognition [J]. Acta Electronica Sinica, 2011, 39(5): 1121-1126.(in Chinese)
[5] 马雷,臧俊杰,张润生.不同光照条件下前方车辆识别方法[J].汽车工程,2012,36(4):360-366.
  Ma Lei, Zang Junjie, Zhang Runsheng. Front vehicle identification under different lighting conditions [J]. Automotive Engineering, 2012, 36(4): 360-366.(in Chinese)
[6] Sun Z, Bebis G, Miller R. On-road vehicle detection: a review [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(5): 694-711.
[7] Sivaraman S, Trivedi M M. Looking at vehicles on the road: a survey of vision-based vehicle detection, tracking, and behavior analysis [J]. IEEE Transactions on Intelligent Transportation Systems, 2013, 14(4): 1773-1795.
[8] Dalal N, Triggs B. Histograms of oriented gradients for human detection [C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA, 2005, 1: 886-893.
[9] Sun Z, Bebis G, Miller R. Monocular precrash vehicle detection: features and classifiers [J]. IEEE Transactions on Image Processing, 2006, 15(7): 2019-2034.
[10] Viola P, Jones M. Rapid object detection using a boosted cascade of simple features [C]//2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Hawaii, USA, 2001, 1: I-511-I-518.
[11] Zhang X, Zheng N, He Y, et al. Vehicle detection using an extended hidden random field model [C]//14th International IEEE Conference on Intelligent Transportation Systems. Washington, DC, USA, 2011: 1555-1559.
[12] Lin B F, Chan Y M, Fu L C, et al. Integrating appearance and edge features for sedan vehicle detection in the blind-spot area [J]. IEEE Transactions on Intelligent Transportation Systems, 2012, 13(2): 737-747.
[13] Yuan Q, Thangali A, Ablavsky V, et al. Learning a family of detectors via multiplicative kernels [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(3): 514-530.
[14] Caraffi C, Vojir T, Trefny J, et al. A system for real-time detection and tracking of vehicles from a single car-mounted camera [C]//15th International IEEE Conference on Intelligent Transportation Systems. Anchorage, AK, USA, 2012: 975-982.
[15] Ludwig O, Nunes U. Improving the generalization properties of neural networks: an application to vehicle detection [C]//11th International IEEE Conference on Intelligent Transportation Systems. Beijing, China, 2008: 310-315.
[16] Fossati A, Schönmann P, Fua P. Real-time vehicle tracking for driving assistance [J]. Machine Vision and Applications, 2011, 22(2): 439-448.
[17] Kapoor A, Grauman K, Urtasun R, et al. Gaussian processes for object categorization [J]. International Journal of Computer Vision, 2010, 88(2): 169-188.
[18] Hoiem D, Efros A A, Hebert M. Recovering surface layout from an image [J]. International Journal of Computer Vision, 2007, 75(1): 151-172.
[19] Specht D F. Enhancements to probabilistic neural networks [C]//International Joint Conference on Neural Networks. Baltimore, MD, USA, 1992, 1: 761-768.
[20] Sivaraman S, Trivedi M M. A general active-learning framework for on-road vehicle recognition and tracking [J]. IEEE Transactions on Intelligent Transportation Systems, 2010, 11(2): 267-276.

备注/Memo

备注/Memo:
收稿日期: 2014-09-10.
作者简介: 蔡英凤(1985—),女,博士,讲师,caicaixiao0304@126.com.
基金项目: 国家自然科学基金资助项目(61403172,51305167,61203244)、交通运输部信息化资助项目(2013364836900)、江苏省自然科学基金资助项目(BK20140555)、中国博士后科学基金资助项目(2014M561592)、江苏省“六大人才”高峰资助项目(2014-DZXX-040)、江苏省博士后基金资助项目(1402097C)、江苏大学高级专业人才科研启动基金资助项目(12JDG010,14JDG028).
引用本文: 蔡英凤,王海.视觉车辆识别迁移学习算法[J].东南大学学报:自然科学版,2015,45(2):275-280. [doi:10.3969/j.issn.1001-0505.2015.02.015]
更新日期/Last Update: 2015-03-20