Vehicle classification by estimation of the direction angle in a mixed traffic flow

The application of Intelligent Transportation System (ITS) is very important in developing societies nowadays. Vehicle monitoring is one of the primary tasks of ITS, where vehicles are classified by lanes for traffic management, especially in case of a mixed flow of motorcycles and other automobiles in the transport system of Vietnam. This paper proposes a new approach in vehicle classification, which is based on evaluation of the direction angle of the first primary axis of each coming vehicle detected in the captured video sequence and map into the predetermined database to mark it as motorcycle or automobiles instead of consideration of vehicle size. The experimental results show that the classification performance is promising, especially for motorcycles and cars, and therefore is applicable in detection of vehicle penalties moving in wrong lanes.