License plate location algorithm in complex scene

A novel license plate positioning algorithm is proposed in this paper. The algorithm is composed of two parts, which are candidate region position and pseudo-region elimination. First the texture features are used to find candidate plates areas. Then a sliding strip window, which scans image from top to bottom, is used to select connectivity regions in the binary image of the candidate area. And pseudo-plate regions are removed from the candidate plates by using plate syntactic features. License plates in complex scene and fuzzy image both can be solved effectively. Moreover, the proposed algorithm can provide priori knowledge for next segmentation and recognition process. Experimental results show that the location method is fast and has high accuracy and strong anti-interference ability for the complex scenes and the fuzzy image particularly.