Detecting License Plate Number Using OCR Technique and Raspberry Pi 4 With Camera

In this paper, a system of car license plate number detection and recognition is proposed. The implemented system consists of four main stages: image processing, segmentation, noise removal and deep learning. A Raspberry Pi 4 device-aided python language and camera are used. After image pre-processing and segmentation, the system will detect the License plate from an image of a vehicle using YOLOv4 algorithm and Cascade Classifier with accuracy in YOLOv4 technique is 0.999, in Cascade Classifier with database of Russian numbers is 0.982 and in the Cascade Classifier with Indian numbers is 0.906. Then the characters of license plate have been extracted. The facility of an easy OCR library has been used to convert this picture to text characters. Then, the process of license plate recognition has been implemented step by step, using python, OpenCV, Numpy, and easy OCR libraries.