基于字袋模型的交通标志识别方法研究Recognition of traffic signs based on Bag-of-Words model
胡晓光,朱欣焰,柳林,李卉,李德仁
摘要(Abstract)：
交通标志识别是智能交通中一个重要的研究领域,实现辅助驾驶和自动导航等应用正获得越来越多的关注。本文提出一种自然场景下交通标志的识别方法:提取标志的局部特征以提高获取效率也使其较少受到遮挡的影响,通过字袋模型量化后获得标志的本征特征,基于该本征特征训练支持向量机,最后使用得到的判别分类器进行交通标志的分类。实验结果表明,和基于Hu矩的识别方法相比,提出的方法提高了识别率和鲁棒性,显示了对交通标志识别的有效性。
关键词(KeyWords)： 交通标志识别;字袋模型;本征特征;支持向量机
基金项目(Foundation): 武汉大学测绘遥感信息工程国家重点实验室自主探索课题资助项目;; 武汉大学测绘遥感信息工程国家重点实验室专项基金资助项目;; 武汉大学测绘遥感信息工程国家重点实验室开放研究基金资助项目
作者(Author): 胡晓光,朱欣焰,柳林,李卉,李德仁
DOI: 10.16251/j.cnki.1009-2307.2012.06.028
参考文献(References)：
- [1]Estevez L,N Kehtarnavaz.A real-time histographic ap-proach to road sign recognition[C]//Proceedings of the IEEE Southwest Symposium on Image Analysis and Inter-pretation,1996.
- [2]A de la Escalera,J M A.,M A Salchs,et al.Road traffic sign detection and classification[C]//IEEE Transactions on Industrial Electronics,1997,44(6):848-859.
- [3]S H Hsu,C L H.Road sign detection and recognition u-sing matching persuit method[J].Image and Vision Computing,2001,19(3):119-129.
- [4]Zhu,S D,Y Zhang,X F Lu.Detection for triangle traffic sign based on neural network[C]//IEEE International Conference on Vehicular Electronics and Safety,2005.
- [5]Maldonado-Bascon,S,et al.Road-Sign Detection and Rec-ognition Based on Support Vector Machines[J].Intelli-gent Transportation Systems,IEEE Transactions on,2007,8(2):264-278.
- [6]Kus,M,M Gokmen,S Etaner-Uyar.Traffic sign recognition using Scale Invariant Feature Transform and color classifi-cation[J].Computer and Information Sciences,2008.
- [7]Maldonado Bascón,S,et al.An optimization on pictogram identification for the road-sign recognition task using SVMs[J].Computer Vision and Image Understanding,2010,114(3):373-383.
- [8]Muller,M,et al.Design of an automotive traffic sign recog-nition system targeting a multi-core SoC implementation[C]//Automation&Test in Europe Conference&Exhi-bition(DATE),2010.
- [9]Lowe,D G.Distinctive Image Features from Scale-Invari-ant Keypoints[J].International Journal of Computer Vi-sion2004,60(2):91-110.
- [10]Hsu,C.-W.,C.-C.Chang,C.-J.Lin,A Practical Guide to Support Vector Classication,2010,Department of Comput-er Science National Taiwan University:Taiwan.
- [11]Sivic,J,et al.Discovering objects and their location in im-ages[C]//Tenth IEEE International Conference on Com-puter Vision,2005.
- [12]Fei-Fei,L,PPerona.A Bayesian hierarchical model for learning natural scene categories[C]//Computer Vision and Pattern Recognition,2005.
- [13]Gang,W,Z Ye,F-F Li.Using Dependent Regions for Ob-ject Categorization in a Generative Framework[C]//Com-puter Vision and Pattern Recognition,IEEE Computer So-ciety Conference,2006.
- [14]Fei-Fei L.CVPR2007_tutorial_bag_of_words[EB/OL].(2007).http://vision.cs.princeton.edu/documents/CVPR2007_tutorial_bag_of_words.ppt.
- [15]北京市公安局公安交通管理局.国家标准GB5768-1999,《道路交通标志和标线》[S].http://www.bjjtgl.gov.cn/publish/portal0/tab113/.2010.
- [16]Ming-Kuei,H.Visual pattern recognition by moment invar-iants[J].Information Theory,IRE Transactions on,1962,8(2):179-187.