Multi-vehicle Convoy Analysis Based On ANPR Data

The automatic number plate recognition (ANPR) system is an important component of Intelligent Transportation System. Research on automatic number plate recognition (ANPR) system involves many fields of techniques, such as data mining, computer vision, pattern recognition, artificial intelligence, automation, digital image processing and so on. The automatic number plate recognition (ANPR) is of great importance in intelligent traffic control system, and automatic number plate recognition (ANPR) technology has become both a very hot research field and an very important research field in transportation projects in recent years. And ANPR system has been widely used in our daily life. The amount of ANPR data captured daily by traffic cameras in the road networks is very substantial, then how to extract the interested information efficiently and with a highly correct rate, and how to reduce the amount of data processing and storage is the problem we need to solve. The interested information means that the information contains the cars which its behavior is unusual and its number plate is suspicious, then we detect such cars’information more clearly and help the police to investigate the related criminal activities, this is an important research field in vehicle convoy analysis and detection, and also a difficult research field.Based on the ANPR data provided by a lot of cameras in multiple road network, the aim of this paper focuses on how to efficiently and correctly find the vehicles’ information whose behavior is unusual and number plate is suspicious, and make full use of the availability and sensitivity of data, and the paper introduces a novel vehicle detection analysis method for use with correlation analysis of vehicle. The novel application of data mining techniques for convoy analysis of vehicles based on the automatic number plate recognition (ANPR) system, the ANRP data are first filtered to remove regular behavior or activities between vehicles, this is an important process as it helps to reduce the number of daily data from approximately200,000by98 percent; secondly, a sorting algorithm is applied to the remaining data to detect the multi-vehicle convoys.The method proposed in this paper can analysis the tremendous ANPR data and can extract the irregular behavious of vehicles. We apply data clustering techniques to extract relevant traffic patterns from the ANPR data to detect and identify unusual patterns and irregular behaviour of multi-vehicle convoy activities.Based on multi-vehicle convoy analysis, at the end of the paper introduced an integrated system, the system uses the ANPR data provided by police, extract the irregular behavious of vehicles, and help police in the investigation. Lots of experiments have done on the ANPR data, and the experiments are strongly proved the algorithm is available and accuracy.