Abstract

License plate localization is an essential part of automated vehicle license plate recognition (ALPR). This paper proposes a two-step approach of the vehicle license plate (LP) localization which includes a novel attempt for LP detection and a new method for LP skew correction. The proposed approach begins with detecting the LP region while classifying the LP type in the image using single shot multi-box detector (SSD) and then further precisely determining the LP corner points in detected region by fitting LP borderlines according to the character contours based on the multi-level thresholding binarization. The experiment evaluated the performance of the proposed method on 600 vehicle images (~800*800) captured under realistic complex environment. The result shows that LPs in 579 images can be localized exactly (96.50%accuracy) and processing speed is up to 27 FPS on average.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.