Proposal of symbol mark detection using one-dimensional histogram and Polytope figure extraction method
Yue Liu, Yuhki Kitazono, Shota Nakashima, Lifeng Zhang, Seiichi Serikawa · 2010
Recently, researches and developments in vehicle safety system has been intensively conducted. In the research field of intelligent transportation systems (ITS), various methods have been made to concern road signs, which can be automatically recognized by driver-support systems. Human characteristics are important factors of traffic accidents. However, research is only focused on the road signs, not in the driver information analysis. Because of this, we describe a new method to extract symbol mark using one-dimensional histogram and Polytope. The method utilizes the Polytope method, which is one of minimization algorithms. For extraction of figures, one-dimensional histogram is proposed. Analysis of the results,we found that, this algorithm is simple, need less memory space, and processing quickly.