[1]张燕昆,洪初阳,WANG Charles.实时矩形交通限速标志识别系统[J].智能系统学报,2010,5(06):540-544.
 ZHANG Yan-kun,HONG Chu-yang,WANG Charles.A real time rectangular speed limit sign recognition system[J].CAAI Transactions on Intelligent Systems,2010,5(06):540-544.
点击复制

实时矩形交通限速标志识别系统(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第5卷
期数:
2010年06期
页码:
540-544
栏目:
出版日期:
2010-12-25

文章信息/Info

Title:
A real time rectangular speed limit sign recognition system
文章编号:
1673-4785(2010)06-0540-05
作者:
张燕昆 洪初阳 WANG Charles
哈曼国际中国研究中心 核心技术研发部,上海 200233
Author(s):
ZHANG Yan-kun HONG Chu-yang WANG Charles
Harman International, Shanghai RD Center Corporate Technology Group, Shanghai 200233, China
关键词:
交通限速标志辅助驾驶系统局部二值模式主元分析二叉树支持向量机
Keywords:
speed limit sign driver assistant system local binary pattern PCA binary tree support vector machine
分类号:
TP273
文献标志码:
A
摘要:
交通限速标志识别系统是汽车辅助驾驶系统的一项重要组成部分,本文提出了一种实时矩形交通标志识别系统.首先采用多尺度多区域的局部二值模式(LBP)特征训练Adaboost分类器进行交通限速标志的检测,然后利用线性预测的算法进行标志跟踪.识别预处理阶段,首先采用投影分析的方法对交通标志进行旋转校正,然后采用基于积分图的自适应二值化方法将图像进行二值化,再利用连通区域标记方法得到包含限速标志数字的最小矩形区域.识别时首先采用主元分析(PCA)进行特征向量提取,然后用聚类的方法构建二叉树的线性支持向量机进行分类识别.在普通笔记本电脑系统配置下,通过大量的实际道路场景的视频数据测试,系统取得了98.3%的正确识别率,平均处理速度达16帧/s.
Abstract:
Speed limit sign recognition is one of the important components for a driver assistance system. An efficient real time rectangular speed limit sign recognition system was proposed. The system framework design considered the computation load and hardware resources for a driver assistance system. First, multiscale overlapping local binary pattern (LBP) image features were used to train an AdaBoost cascade classifier for sign detection. Then a simple linear prediction method was used to do the tracking task. In the recognition stage, the projection method was used to correct the rotation angle and then the integral image based adaptive threshold algorithm was applied to segment the speed limit number, and then the principal component analysis (PCA) was used for feature vector extraction. Finally, a clustering based binary tree of a linear support vector machine was designed for the classification task. The system achieved a 98.3% recognition rate with an approximate frame rate of 16fps in video files for the laptop computer system during actual road tests.

参考文献/References:

[1]BARGETON A, MOUTARDE F, NASHASHIBI F, BRADAI B. Improving panEuropean speedlimit signs recognition with a new “global number segmentation before digit recognition[C]//IEEE Intelligent Vehicles Symposium. Eindhoven, Netherlands, 2008: 349354.
[2]KELLER C G, SPRUNK C, BAHLMANN C, GIEBEL J, BARATOFF G. Realtime recognition of US speed signs[C]//IEEE Intelligent Vehicles Symposium. Eindhoven, Netherlands, 2008: 518523.
 [3]LOY G, BARNES N. Fast shapebased road sign detection for a driver assistance system[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Sendai, Japan, 2004: 7075.
[4]BAHLMANN C, ZHU Ying, RAMESH V, PELLKOFER M, KOEHLER T. A system for traffic sign detection, tracking, and recognition using color, shape, and motion information[C]//IIEEE Intelligent Vehicles Symposium. Las Vegas, USA, 2005: 255260.
[5]VIOLA P, JONES M. Robust realtime object detection[J]. International Journal of Computer Vision, 2004, 57(2): 137154.
[6]陈春燕, 章品正, 罗立民. 基于粒特征和连续Adaboost的人脸检测[J]. 智能系统学报, 2009, 4(5): 446452.
CHEN Chunyan, ZHANG Pinzheng, LUO Limin. Face detection using real Adaboost on granular features[J]. CAAI Transactions on Intelligent Systems, 2009, 4(5): 446452.
[7]OJALA T, PIETIK〖AKA¨〗INEN M, M&ENP T. Multiresolution grayscale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971987.
[8]ZHANG Lun, CHU Rufeng, XIANG Shiming, LI S Z. Face detection based on multiblock LBP representation[C]//Proceedings of IAPR/IEEE International Conference on Biometrics. Seoul, Korea, 2007: 1118.
[9]BRADLEY G, ROTH G. Adaptive thresholding using integral image[J]. Journal of Graphics Tools, 2007, 12(2): 1321.
[10]FEI B, LIU J. Binary tree of SVM: a new fast multiclass training and classification algorithm[J]. IEEE Trans on Neural Networks, 2006, 17(3): 696704.
[11]张铭钧, 尚云超, 杨杰. 基于灰度梯度不变矩的水下目标识别系统[J].哈尔滨工程大学学报, 2009, 6: 653657. 
ZHANG Mingjun, SHANG Yunchao, YANG Jie. Recognition system for underwater objects based on gray and grads invariant moments[J].Journal of Harbin Engineering University, 2009, 6: 653657.

备注/Memo

备注/Memo:
收稿日期:2009-12-17.
通信作者:张燕昆.E-mail:James.Zhang@harman.com.
作者简介:
张燕昆,男,博士.主要研究方向为计算机视觉、人脸检测与识别、三维计算机视觉;发表学术论文7篇. 
洪初阳,男,硕士,主要研究方向为图像处理、图像分割和机器视觉等.
Wang Charles, 博士,主要研究方向为视频分析、模式识别等.发表论文多篇,同时拥有多项专利.
更新日期/Last Update: 2011-03-03