AI Driven Smart Number Plate Identification for Automatic Identification

Number Plates are necessary for every vehicle. Nowadays, they are recognized for identifying persons violating traffic rules, accused and so on. Number Plate recognition suffers from several challenges. One among them is identifying the numbers with different styles. The Government of Tamilnadu has suggested the format for two-wheeler number plate for easy recognition. Other regional characters or different styles are not allowed in the number plates. But most of the people are not adhere to the format. The objective of this paper is to reduce the complexity of the Number Plate Recognition (NPR) by font style/Size/Color and background recognition. Hence, in this research, the format of the two-wheeler number plate is matched, and the unformatted ones are identified. A message is sent via Short Message Service (SMS) to the Registered Mobile Number (RMN). A new dataset is created from real-time videos and experiments are conducted in it. The proposed method achieved maximum accuracy of above 99% on the collected dataset.