Vehicle Classification Using Visual Background Extractor and Multi-class Support Vector Machines
This paper describes a method to classify vehicle type using computer vision technology. In this study, Visual Background Extractor (ViBe) was used to extract the vehicles from the captured videos. The features of the detected vehicles were extracted using Histogram of Oriented Gradient (HOG). Multi-class Support Vector Machine (SVM) was used to recognise four classes of images: motorcycle, car, lorry and background (without vehicles). The results show that the proposed classifier was able to achieve an average accuracy of 92.3 %.
KeywordsVehicle classification Visual background extractor Support vector machines Histogram of oriented gradient
Specially thanks to Chih-Chung Chang and Chih-Jen Lin for sharing LIBSVM, which contributed greatly to this study.
- 1.Avely RP, Wang Y, Rutherford GS (2004) Length-based vehicle classification using images from uncalibrated video cameras. IEEE intelligent transportation systems conference, pp 737–742Google Scholar
- 3.Hjort M, Haraldsson M, Jansen JM (2008) Road wear from heavy vehicles. NVF committee vehicles and transportsGoogle Scholar
- 4.Daubaras A, Zilys M (2012) Vehicle detection based on magneto-resistive magnetic field sensor. Electron Electr Eng 118:27–32Google Scholar
- 5.Sun Z, Bebis G, Miller R (2004) On-road vehicle detection using optical sensors: a review. In: Proceedings of the 7th international IEEE conference on intelligent transportation systems, pp 585–590Google Scholar
- 7.Stauffer C, Grimson W (1999) Adaptive background mixture models for real-time tracking. In: IEEE computer society conference on computer vision and pattern recognitionGoogle Scholar
- 8.Barnich O, Van Droogenbroeck M (2011) ViBe: a universal background subtraction algorithm for video sequences. IEEE Trans Image Process: Publ IEEE Signal Process Soc 20:1709–1724Google Scholar
- 9.Razali MT, Jantan A (2008) Support vector machine for classify dynamic human/vehicle shapes. In: International conference on electronic design, pp 1–6Google Scholar
- 10.Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: IEEE computer society conference on computer vision and pattern recognitionGoogle Scholar
- 12.Teoh SS (2011) Development of a robust monocular-based vehicle detection and tracking system. Ph.D. Thesis, The University of Western AustraliaGoogle Scholar
- 13.Hsu CW, Lin CJ (2002) A comparison of methods for multiclass support vector machines. IEEE Trans Neural Networks 13:415–425Google Scholar