道路交通标志检测分类方法的研究

研究了道路交通标志检测分类问题,针对颜色定位检测交通标志的缺陷,提出了一种基于形状边缘定位和颜色判别的交通标志检测分类方法。首先将原图像从RGB色彩空间转换到HSV,在饱和度S通道上用Canny算子检测边缘,计算边缘的形状参数(圆形度、矩形度以及推广得到的正三角形度)以判定边缘形状,定位出标志的位置;然后采用修正的HSV色彩空间分割模型判别标志颜色以进行分类,分类过程中筛除了非标志区域。实验证明,该方法具有良好的检测分类效果。

Study on road traffic sign detection and classification

This paper studies on the problem of road traffic sign detection. For the weakness of color positioning, it proposes a traffic sign detection method based on shape positioning and color discrimination. Firstly, it turns original RGB image to HSV color space, and uses Canny operator on saturation component for edge detection. It calculates the shape parameters (degree of circularity, rectangular degree and the extended regular trianglar degree) to judge the shape of edge and locate the position of signs. Then it uses the modified HSV color space partition model to judge color for classification and removes non-signs in the classification process. Experimental results prove that the method is effective in detection and classification.