Reeal Time Featur Based Vehicle Detection and Classification from On-Road Videos

Abstract

Vision Based vehicle detection and classification has become an active area of research for intelligent transportation system. But this task is very difficult and challenging due to the dynamic condition of roads. In the proposed method, a feature based cost effective detection and classification method is proposed that is suitable for real time applications, provide satisfactory accuracy and computationally cheap. The proposed method uses haar-like image features and AdaBoost classifier for detection. To reduce false positive rate, we propose to use two virtual detection lines (VDL). In order to predict the class of a vehicle, we propose a two level classifier where first classifier separates bigger (bus, truck) vehicles from the smaller one (car, CNG, rickshaw) based on some shape information of vehicles. For the second classifier, we propose to use bag of features (BOF) model which uses the feature efficiently and generates bag of visual words (BOVW). Shape based features are used for first classifier and texture based feature (SURF) is used for second classifier. Error correcting output code (ECOC) framework is used to achieve multi class prediction with SVM to predict the class. Extensive experiments have been carried out on different local traffic data of varying environments to evaluate the detection and classification performance of the proposed method. Experimental results demonstrate that the proposed two level classifier achieves a significant improvement in classification of heterogeneous vehicles in terms of accuracy with a considerable execution time as compared to other methods.