Distance Metrics based Vehicle Object Identification in Dynamic Vision

Vehicle object identification is a challenging issue in the visual surveillance. In recent years, video monitoring and surveillance system have been widely utilized for traffic monitoring and management. In this paper, we propose an algorithm to identify the moving objects from the sequence of video frames which contains dynamically changing backgrounds in the noisy environment. Reference to our previous surveillance skeletanization works, here we have proposed a methodology to identify an object using distance metrics such as Hamming and Euclidian distance metrics. The recent vehicle recognition methods could fail to identify the objects and produce more false acceptance rate (FAR) or false rejection rate (FRR) however our research recommends a method for object identification using weighted distance to extract features in order to obtain robust identification in the noisy environment. KeywordsDistance transformations, Euclidian Distance, false acceptance rate, false rejection rate, Hamming Distance, Skeletanization, Traffic video sequences, weighted distance.