Investigating the Misidentification of Automatic Number Plate Recognition Systems

The aim of this study is to analyze the patterns of an automatic number plate recognition system, AVI (Automatic Vehicle Identification) misidentification errors. ITS performance evaluation results show that the performance level of the equipments becomes lower as they are old. Also, AVI is more suitable for day time traffic data collection than night time because of its visibility. Most errors come from similar shaped letter such as consonant, vowel or number, and different plate types such as old fashioned plate, plate of vehicle kept for business and new model (long-width type) plate types. In the results and discussion part, we talk about these types with figures. In conclusion, it is a necessary to maintain the equipments regularly to assure the ITS reliability.