Traffic incident acoustic recognition method based on wavelet decomposition and support vector machine
-
摘要: 分析了现有交通事件自动检测和识别方法, 提出了应用小波分解与支持向量机相结合的交通事件声频识别方法。将车辆行驶的声音信号进行小波分解, 以不同频段的重构信号能量作为特征向量, 对由多个支持向量机构成的交通事件分类器进行训练, 并对正常行驶、刹车和碰撞事件的声音信号进行识别。试验结果表明: 利用车辆声音信号能够正确识别不同的交通事件, 识别准确率达95%, 识别方法可行。Abstract: The existing automatic detection and recognition methods of traffic incidents were analyzed, a recognition method with vehicle acoustic signals was proposed based on wavelet decomposition(WD) and support vector machine(SVM). Vehicle acoustic signals were decomposed with WD, the powers in different frequencies were regarded as different incident eigenvectors, and the traffic incident classifier composed of multiple SVMs was trained. The acoustic signals of normal driving, braking and crash incidents were recognized. Test result shows that various traffic incidents can be recognized with vehicle acoustic signals, the recognition rate reaches 95%, so the proposed method is feasible.
-
Key words:
- traffic information processing /
- traffic incident /
- wavelet decomposition /
- SVM
-
图 3 刹车事件声音信号小波重构结果
Figure 3. Wavelet reconstruction results of braking acoustic signal
表 1 交通事件分类器训练样本
Table 1. Training samples of traffic incident classifiers
样本序号 训练样本的特征向量元素 类别标签 事件类型 1 0.135 1, 0.167 8, 0.159 3, 0.159 4, 0.174 5, 0.124 5, 0.079 4 1 正常行驶 2 0.154 5, 0.135 2, 0.124 7, 0.146 3, 0.168 2, 0.155 7, 0.115 4 1 正常行驶 3 0.154 6, 0.136 5, 0.124 3, 0.148 3, 0.164 6, 0.157 5, 0.114 2 1 正常行驶 4 0.146 0, 0.166 2, 0.131 9, 0.145 5, 0.156 6, 0.127 6, 0.126 2 1 正常行驶 5 0.059 6, 0.016 7, 0.022 0, 0.115 6, 0.290 0, 0.359 4, 0.136 7 2 刹车事件 6 0.048 4, 0.012 2, 0.011 1, 0.211 0, 0.345 1, 0.249 7, 0.122 5 2 刹车事件 7 0.060 2, 0.013 7, 0.030 3, 0.192 2, 0.299 7, 0.249 2, 0.154 7 2 刹车事件 8 0.060 4, 0.009 9, 0.010 9, 0.213 9, 0.328 6, 0.253 0, 0.123 3 2 刹车事件 9 0.034 7, 0.053 2, 0.164 4, 0.145 2, 0.195 7, 0.250 7, 0.156 1 3 碰撞事件 10 0.031 4, 0.057 8, 0.179 4, 0.126 5, 0.146 7, 0.303 5, 0.154 7 3 碰撞事件 11 0.027 0, 0.061 7, 0.196 7, 0.148 4, 0.169 5, 0.266 8, 0.129 9 3 碰撞事件 12 0.020 3, 0.058 6, 0.187 0, 0.139 1, 0.154 3, 0.283 5, 0.157 2 3 碰撞事件 
下载: 导出CSV
-
[1] 庄斌, 杨晓光, 李克平. 道路交通拥挤事件判别准则与检测算法[J]. 中国公路学报, 2006, 19(3): 82-86. doi: 10.3321/j.issn:1001-7372.2006.03.015ZHUANG Bin, YANG Xiao-guang, LI Ke-ping. Criterion and detection algorithmfor road traffic congestion incidents[J]. China Journal of Highway and Transport, 2006, 19(3): 82-86. (in Chinese) doi: 10.3321/j.issn:1001-7372.2006.03.015 [2] 何杰, 胡如夫, 李传志, 等. 基于无线定位终端的公路事件检测方法研究[J]. 系统仿真学报, 2009, 21(12): 3828-3832. https://www.cnki.com.cn/Article/CJFDTOTAL-XTFZ200912077.htmHEJie, HURu-fu, LI Chuan-zhi, et al. Study on freeway incident detection using integrated wireless position terminal[J]. Journal of System Simulation, 2009, 21(12): 3828-3832. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XTFZ200912077.htm [3] 汪勤, 黄山, 张洪斌, 等. 基于视频图像处理的交通事件检测系统[J]. 计算机应用, 2008, 28(7): 1886-1889. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY200807080.htmWANG Qin, HUANG Shan, ZHANG Hong-bin, et al. Traffic incident detection system based on video image processing[J]. Computer Applications, 2008, 28(7): 1886-1889. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY200807080.htm [4] 张存保, 杨晓光, 严新平. 基于浮动车的高速公路交通事件自动判别方法研究[J]. 武汉理工大学学报: 交通科学与工程版, 2006, 30(6): 973-975, 983. doi: 10.3963/j.issn.2095-3844.2006.06.013ZHANG Cun-bao, YANG Xiao-guang, YAN Xin-ping. An automatic incident detection methodology for freeway using floating cars[J]. Journal of Wuhan University of Technology: Transportation Science and Engineering, 2006, 30(6): 973-975, 983. (in Chinese) doi: 10.3963/j.issn.2095-3844.2006.06.013 [5] 郭艳玲, 吴义虎, 黄中祥. 基于小波分析和SOM网络的交通事件检测算法[J]. 系统工程, 2006, 24(10): 100-104. https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200610020.htmGUO Yan-ling, WU Yi-hu, HUANG Zhong-xiang. An algorithmfor traffic incidents detection based on wavelet analysis and SOM network[J]. Systems Engineering, 2006, 24(10): 100-104. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200610020.htm [6] 陈斌. 基于支持向量机的高速公路意外事件检测模型[J]. 中国公路学报, 2006, 19(6): 107-112. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200606020.htmCHEN Bin. Freeway accident detection model based on support vector machine[J]. China Journal of Highway and Transport, 2006, 19(6): 107-112. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200606020.htm [7] 张敬磊, 王晓原. 交通事件检测算法研究进展[J]. 武汉理工大学学报: 交通科学与工程版, 2005, 29(2): 215-218. https://www.cnki.com.cn/Article/CJFDTOTAL-JTKJ200502014.htmZHANG Jing-lei, WANG Xiao-yuan. Research progress of traffic incident automatic detection algorithms[J]. Journal of Wuhan University of Technology: Transportation Scienceand Engineering, 2005, 29(2): 215-218. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JTKJ200502014.htm [8] WANG Wei, CHEN Shu-yan, QU Gao-feng. Incident detection algorithm based on partial least squares regression[J]. Transportation Research Part C: Emerging Technologies, 2008, 16(1): 54-70. doi: 10.1016/j.trc.2007.06.005 [9] CHEN S, SUN Z P, BRIDGE B. Automatic traffic monitoring by intelligent sound detection[C]∥ ITSC. Proceedings of IEEE Intelligent Transportation Systems Conference. Boston: IEEE, 1997: 171-176. [10] BALRAJ N. Automated accident detection in intersections via digital audio signal processing[D]. Starkville: Mississippi State University, 2003. [11] 熊烈强, 商蕾, 高孝洪. 基于噪声和振动的快速路交通事件检测方法[J]. 武汉理工大学学报: 交通科学与工程版, 2005, 29(2): 238-241. doi: 10.3963/j.issn.2095-3844.2005.02.021XIONG Lie-qiang, SHANG Lei, GAO Xiao-hong. A study of AID for urban expressway based on traffic noise and road vibration[J]. Journal of Wuhan University of Technology: Transportation Science and Engineering, 2005, 29(2): 238-241. (in Chinese) doi: 10.3963/j.issn.2095-3844.2005.02.021 [12] 陈强. 高速公路交通流特征参数被动声学检测技术研究[D]. 长春: 吉林大学, 2005.CHEN Qiang. Research on the technology of passive acoustics detection of expressway traffic flow characteristic parameters[D]. Changchun: Jilin University, 2005. (in Chinese) [13] MALLAT S G. Atheoryfor multiresolution signal decomposition: the wavelet representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1989, 11(7): 674-693. [14] MILLET-ROIGJ, VENTURA-GALI ANO R, CHORRO-GASCO F J, et al. Support vector machine for arrhythmia discrimination with wavelet transform-based feature selection[J]. Computers in Cardiology, 2000, 27(1): 407-410. [15] HSU C W, LI N CJ. Acomparison of methods for multiclass support vector machines[J]. IEEE Transactions on Neural Networks, 2002, 13(2): 415-425. [16] OLI VIER C, VLADI MIR V, OLI VIER B, et al. Choosing multiple parameters for support vector machines[J]. Machine Learning, 2002, 46(1): 131-159. -
点击查看大图
图(3) / 表(1)
计量
- 文章访问数: 513
- HTML全文浏览量: 103
- PDF下载量: 698
- 被引次数: 0
