Iranian License Plate Recognition using Deep Learning
Paper ID : 1070-MVIP2020
Atefeh Ranjkesh *1, Asadollah Shahbahrami2, Alireza Akoushideh3
2دانشگاه گیلان
3Shahid-Chamran College Technical and Vocational University Tehran, Iran
Automated License Plate Recognition (ALPR) is used in smart cities as its application ranges from parking to urban transport safety. The ALPR has three main steps, license plate localization, segmentation and Optical Character Recognition (OCR). Each step needs different techniques in real condition and each technique has its specific characteristics. License plate localization techniques detect the license plate after that segmentation algorithms should segment and isolate each character from each other. Finally, the OCR step is applied to recognize the separated characters. The final accuracy depends on the accuracy of each step. For example, the performance of the OCR step depends on the segmentation step. To improve the OCR step performance, we have combined both segmentation and OCR steps as a single-stage using deep learning techniques such as the YOLO framework. Our experimental results show that this proposed approach recognizes the Iranian license plate characters with accuracy over 99% compared to previous work.
optical character recognition; deep learning; YOLO; artificial neural network; support vector machine
Status : Paper Accepted (Oral Presentation)