A Comparative Analysis of Deep Learning Approach for Automatic Number Plate Recognition

Automatic number plate detection and analysis is a general monitoring strategy used by a large number of city vehicles to enhance traffic management, routing, traffic control, toll collection, and regulation and protection of highway law. ANPR approach can be applied according to different methodologies. This job can be scanned, executed and compared. This proposed work is carried out in real-time application using YOLO v3 for the identification and recognition of plate numbers. In this study, a comparative method for ANPR has been demonstrated. Traditional approaches were focused on contouring, segmentation, edge detection processes which gave less accuracy but here tried to implement YOLO v3 technique that will give more accurate results for Indian license plate detection in real-time.