Deep learning-based car plate optical character recognition
In the field of intelligent transport systems, recent years have witnessed the application of deep learning techniques to both car plate detection and recognition. The latter stage, known as optical character recognition (OCR), is more challenging as it requires an accurate prediction of the entire...
| Main Author: | |
|---|---|
| Format: | Final Year Project / Dissertation / Thesis |
| Published: |
2022
|
| Subjects: | |
| Online Access: | http://eprints.utar.edu.my/4953/ http://eprints.utar.edu.my/4953/1/3E_1806581_Final_report_%2D_ZHEN_BO_CHOO.pdf |
| _version_ | 1848886286098628608 |
|---|---|
| author | Choo, Zhen Bo |
| author_facet | Choo, Zhen Bo |
| author_sort | Choo, Zhen Bo |
| building | UTAR Institutional Repository |
| collection | Online Access |
| description | In the field of intelligent transport systems, recent years have witnessed the application of deep learning techniques to both car plate detection and recognition. The latter stage, known as optical character recognition (OCR), is more challenging as it requires an accurate prediction of the entire license numbers. One of the widely used OCR engines is the Tesseract, which uses long short-term memory (LSTM). However, the drawback of this approach is the time-consuming image preprocessing techniques. This project aims to design an accurate yet lightweight OCR solution by exploring the bidirectional LSTM, connectionist temporal classification (CTC) and ResNet. The training datasets comprise two public synthetic datasets and one self-collected dataset, which is specific to the Malaysian car plate format. The trained models are subsequently optimized via OpenVINO for faster inference time. Results show that the proposed solution is 10x faster than the Tesseract OCR while still having more than a 2x increase in accuracy. In a case study of vehicle surveillance, a local webserver is established to host the newly developed OCR solutions in combination with a pre-trained YOLOv4 car plate detection. Results show that the end-to-end solution can process video streams at a rate of 20 frames per second (FPS). |
| first_indexed | 2025-11-15T19:36:04Z |
| format | Final Year Project / Dissertation / Thesis |
| id | utar-4953 |
| institution | Universiti Tunku Abdul Rahman |
| institution_category | Local University |
| last_indexed | 2025-11-15T19:36:04Z |
| publishDate | 2022 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | utar-49532022-12-23T09:12:35Z Deep learning-based car plate optical character recognition Choo, Zhen Bo TK Electrical engineering. Electronics Nuclear engineering In the field of intelligent transport systems, recent years have witnessed the application of deep learning techniques to both car plate detection and recognition. The latter stage, known as optical character recognition (OCR), is more challenging as it requires an accurate prediction of the entire license numbers. One of the widely used OCR engines is the Tesseract, which uses long short-term memory (LSTM). However, the drawback of this approach is the time-consuming image preprocessing techniques. This project aims to design an accurate yet lightweight OCR solution by exploring the bidirectional LSTM, connectionist temporal classification (CTC) and ResNet. The training datasets comprise two public synthetic datasets and one self-collected dataset, which is specific to the Malaysian car plate format. The trained models are subsequently optimized via OpenVINO for faster inference time. Results show that the proposed solution is 10x faster than the Tesseract OCR while still having more than a 2x increase in accuracy. In a case study of vehicle surveillance, a local webserver is established to host the newly developed OCR solutions in combination with a pre-trained YOLOv4 car plate detection. Results show that the end-to-end solution can process video streams at a rate of 20 frames per second (FPS). 2022 Final Year Project / Dissertation / Thesis NonPeerReviewed application/pdf http://eprints.utar.edu.my/4953/1/3E_1806581_Final_report_%2D_ZHEN_BO_CHOO.pdf Choo, Zhen Bo (2022) Deep learning-based car plate optical character recognition. Final Year Project, UTAR. http://eprints.utar.edu.my/4953/ |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Choo, Zhen Bo Deep learning-based car plate optical character recognition |
| title | Deep learning-based car plate optical character recognition |
| title_full | Deep learning-based car plate optical character recognition |
| title_fullStr | Deep learning-based car plate optical character recognition |
| title_full_unstemmed | Deep learning-based car plate optical character recognition |
| title_short | Deep learning-based car plate optical character recognition |
| title_sort | deep learning-based car plate optical character recognition |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| url | http://eprints.utar.edu.my/4953/ http://eprints.utar.edu.my/4953/1/3E_1806581_Final_report_%2D_ZHEN_BO_CHOO.pdf |