License plate recognition in slow motion vehicles

Abstract

The recognition of license plate numbers represents one of the most efficient techniques to identify any individual vehicle. The principle of the system is that the detection of the license plate will be done with two techniques first you only look once (YOLO) and cascade classifier. Then after achive correct detection, the system will send the result (the image of the license plate) to Easy optical character recognition (OCR) library to read it and transform the image into text. In this paper, an analytical study of the surveillance system which affects by parallax due to camera movement has been done, by merging the OCR technique with the attached camera using python aided Raspberry Pi. The hardware system has been designed and implemented.