German License Plate Recognition System Using the YoloV8 Model and EasyOCR
License Plate Recognition (LPR) is a highly valuable technology for the modern road-infrastructure, which can accommodate wide-range features. This paper focuses on developing a German License Plate Recognition (LPR) using the technology of the YOLOv8 model and EasyOCR recognition technology. The Python programming language and OpenCV image-processing library are used to develop this system whereas the EasyOCR library is used as an OCR for recognizing characters. The proposed system developed to detect different license plates for a video input and at the same time recognizes both characters and numbers combination for these detected plates. This hybrid mechanism improves performance and accuracy of the detection process by the combination of simultaneous detection and recognition of each license plate for each frame. The over 4000 images of vehicles and German license plates are used as a dataset of this research. By using the YOLOv8 model and EasyOCR recognition model, a license plate detection rate of 95% and a recognition rate of 94% have been achieved on the test results. These results indicate the success of the proposed detection and recognition process on the German Car license Plate.
Authors
- Atul Chandra Nath
- Hasan Aljzaere
- Wolfram Hardt
Institutions
- Chemnitz University of Technology (DE)
Publication Details
- Published
- 2024-11-13
- DOI
- https://doi.org/10.1109/icsj62869.2024.10804754
- Citations
- 6
- Primary Topic
- Vehicle License Plate Recognition
- Type
- article
- Field-Weighted Citation Impact
- 1.92