Vehicle detection using acoustic signatures

This paper deals with the problem of classification of vehicles based on their acoustic signatures. Each type of vehicle transmits a particular type of engine sound, which can be used as a basis of classification. The samples are first collected using a reliable recording device. The signals so obtained are de-noised using wavelet analysis. The frames to be analyzed are selected using a unique energy index method. The prominent features of the obtained frame are then extracted. A novel feature selection method based on mean and variance is used to select the required features for analysis. The paper then focuses on a fast and potent method for classification of vehicles using k-nearest neighbours algorithm (kNN) into three categories: Two wheelers, four wheelers and Heavy Transport Vehicles (HTVs). Thus the method achieves its required results by using expeditive algorithms.