Preparation of papers in a two-column format for the 2015 15th international conference on control, automation and systems (iccas 2015) night time vehicle detection by using color information based on tail-light

In this paper presents an algorithm, aims at practical applications, for the high speed processing and performance enhancement of vehicle detection base on vision processing system. As a preprocessing for high speed vehicle detection in the night, the vanishing line estimation and the optimal extraction of ROI (region of interest) for can be processed to reduction of detection region in which high speed processing is enabled. In order to get better performance the image need pre-processing. Legally vehicle tail light must be red color, so we use the red plane image with simultaneous processing of noise reduction and edge enhancement using the adv morphology function. And then give Dilation and Erosion eliminating the noise through the process of Morphology. Finally, using the Projection-Histogram information get the center of mass by restricting threshold detects a vehicle in the ROI. This method is more efficient than simply detecting and sliding the entire screen. Finally, the various real road experimental results are presented to evaluate the effectiveness of the proposed method.