Unsupervised Vehicle-Classifier Learning Method using It-means and Support Vector Machine

摘要

Vehicle classification is an essential function of vehicle detectors that provide real-time traffic information to traffic control system. This investigation proposed an unsupervised vehi-cle classification method which combined k-means clustering algorithm support vector machine. This method deals with features derived from FMCW radar signal and able to classify two vehicle classes. The numerical example shows a significant result of less than 2% classification error.