G. Chen, Z. Li, W. Zhang, Y.W. Lu, X.F. Zhu

Pages: 143-150

Abstract
For the problem of automatic vehicle recognition for traffic flow detection at home and abroad, the recognition method based on Support Vector Machine(SVM) and Hidden Markov Model(HMM) algorithm is proposed through extracting linear prediction coefficient(LPC) characteristic from vehicle acoustic. LPC algorithm is learned from voice technology by common between vehicle acoustic and voice. By contrasting other recognition method, simulation indicates vehicle targets are recognized correctly form acoustic signals and this method is effective in vehicle target recognition.

Keywords: traffic flow; vehicle target recognition; linear prediction coefficient(LPC); recognition