System Locating License Plates with Shadow Based on Self-adaptive Window Size Technique.

Jingyu Dun,Sanyuan Zhang
DOI: https://doi.org/10.1007/978-981-10-3969-0_15
2016-01-01
Abstract:Most of the existing license plate localization algorithms have a parameter that is related to the size of the license plate. There is no parameter that is suitable for all the cases. In this paper, an algorithm is proposed to automatically compute the size-related parameter. Then a hierarchical system based on the self-adaptive parameter is proposed to locate license plates. Both connected component based methods and vertical edge based methods are used. The parameter is first used as the local window size to suppress the shadow. Then it is used to connect the discrete vertical edges to form a license plate region. The proposed system is used to locate license plates with shadow, and experiments are taken on images with different resolutions. The total localization accuracy achieves 94.40%. It can compete with the state-of-the-art methods and need not determine the optimal parameter by trial and error.
What problem does this paper attempt to address?