Exploring Sound Signature for Vehicle Detection and Classification Using ANN

This paper attempts to explore the possibility of using sound signatures for vehicle detection andclassification purposes. Sound emitted by vehicles are captured for a two lane undivided road carryingmoderate traffic. Simultaneous arrival of different types vehicles, overtaking at the study location, sound ofhorns, random but identifiable back ground noises, continuous high energy noises on the back ground arethe different challenges encountered in the data collection. Different features were explored out of whichsmoothed log energy was found to be useful for automatic vehicle detection by locating peaks. Mel-frequency ceptral coefficients extracted from fixed regions around the detected peaks along with themanual vehicle labels are utilised to train an Artificial Neural Network (ANN). The classifier for fourbroad classes heavy, medium, light and horns was trained. The ANN classifier developed was able topredict categories well