Design of Ultra Low Power Vehicle Detector utilizing Discrete Wavelet Transform

Vehicle tra c is important information in the intelligent transport system (ITS). We have developed an acoustic vehicle detection system that relies on two microphones at a sidewalk. The system has already been confirmed that it successfully detected the vehicle and direction of travel with an F-measure of 0.92. However, the power consumption of the system is high, which puts power restrictions on deployment. We therefore propose a low power vehicle detection system utilizing a wake-up mechanism. The acoustic vehicle detection system, presented in our previous work, is activated when a newly developed ultra low power vehicle detector (ULP-VD) detects vehicles. Initial experimental evaluations reveal that the ULP-VD successfully detected vehicles with a precision of 0.94 and recall of 0.95.

[1]  S. Chen,et al.  Automatic traffic monitoring by intelligent sound detection , 1997, Proceedings of Conference on Intelligent Transportation Systems.

[2]  Grant R. Gerhart,et al.  Wavelet-based ground vehicle recognition using acoustic signals , 1996, Defense + Commercial Sensing.

[3]  Sergio A. Velastin,et al.  Vehicle localisation and classification in urban CCTV streams , 2009 .

[4]  Akira Fukuda,et al.  Design of Acoustic Vehicle Count System using DTW , 2016 .

[5]  Amir Averbuch,et al.  Wavelet-based acoustic detection of moving vehicles , 2009, Multidimens. Syst. Signal Process..

[6]  James F. Forren,et al.  Traffic monitoring by tire noise , 1997, Proceedings of Conference on Intelligent Transportation Systems.

[7]  Akira Fukuda,et al.  Design of Simple Vehicle Counter using Sidewalk Microphones , 2016 .

[8]  W. Jatmiko,et al.  ITS information source: Vehicle speed measurement using camera as sensor , 2012, 2012 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[9]  Shiping Chen,et al.  Traffic monitoring using digital sound field mapping , 2001, IEEE Trans. Veh. Technol..

[10]  P. Khosla,et al.  Vehicle sound signature recognition by frequency vector principal component analysis , 1998, IMTC/98 Conference Proceedings. IEEE Instrumentation and Measurement Technology Conference. Where Instrumentation is Going (Cat. No.98CH36222).