MULTI DATASET BASED LICENSE PLATE RECOGNITION USING YOLO DETECTORS

Detecting the licence plate is the most reliable and economical technique for identifying cars. The methodologies and approaches vary based on several parameters, including but not limited to: picture quality, the vehicle in fixed places, lighting circumstances, and single images. The variations in licence plates across different countries and states must to be able to handle it as well. Additionally, the technique must be able to function properly when there are many characters in the collected photos that have different plate sizes. This endeavor's aim is developing and creating a software for Licence Plate Recognition (LPR), which may be used for e-challan, in car parking, and vehicle identification for smart garage applications through a design by thinking approach. The main focus will be recognising and identifying several automobiles with licence plates from a single picture. Two processes make up the suggested system: plate number identification and identification. The plate number identification procedure identifies the number plate from the photograph, and the segmented plate is then sent to the plate recognition phase in the second phase to determine the characters and numbers.