基于颜色与形状特征的交通标志检测方法

Traffic Sign Detection Method Based on Color and Shape Features

现今社会,交通事故频发,每年因交通事故而造成的伤亡人数正在逐年递增,如何提高道路交通的安全性,已然成为迫在眉睫的问题。文中所研究的交通标志识别技术,可以及时将道路两旁交通标志反馈给驾驶员,从而减少或避免由于驾驶员的疏忽引发的交通事故。交通标志主要包括,警告标志、禁令标志和指示标志三种,而不同类型的交通标志均具有不同的颜色和形状特征。针对交通标志的上述特点,文中提出了一种将颜色和形状特征相结合的交通标志识别系统。在该方法中,首先采用HSV颜色特征对整幅图像进行粗提取,而后进行形态学滤波并基于标志的几何形状特征对整幅图像实现精细分割,完成标志检测。实验结果表明,该方法运算量小,对外界环境变化不敏感,能够准确、快速地检测出视线范围内的交通标志。

Nowadays, traffic accidents happening every day, the number of casualties caused by traffic accidents is increasing. How to im- prove road traffic safety has become a pressing problem. Traffic signs recognition mentioned gets useful traffic information for drivers through computer processing and feedback to the driver timely, to reduce and avoid traffic accidents caused by the drivers＇ negligence. The road traffic signs includes warning signs, prohibition signs and directional signs. They all have specific colors and shapes. According to the characteristics of the traffic signs,present a traffic sign recognition system based on color-shape feature. Firstly,use HSV color model to extract the whole image for traffic signs region, then utilize morphological methods to get filtration and apply geometric shape characteristic feature to get split, finally extracting the region which are satisfied both color and shape conditions. Experimental results show that the algorithm has a small amount of computation, not sensitive to changes in the external environment, can accurately and quickly detect the road traffic signs within sight.