基于数学形态学的交通标志自动识别

本文提出了基于数学形态学理论的交通标志自动识别方法，并研制了一个实验系统．运用数学形态学方法对交通标志的内核形状进行形态处理并提取其形态骨架函数，根据内核形状的形态骨架特征对交通标志进行分类，以达到识别交通标志的目的．实验结果表明形态骨架函数能有效地表达交通标志的内核形状特征，以形态骨架函数作为物体的形状特征进行模式匹配，能较好地识别交通标志．新的形态骨架匹配算法具有位移不变性质，使得识别具有良好的稳健性

Automatic recognition of traffic signs based on mathematical morphology

In this paper, a method for automatically recognizing traffic signs based on mathematical morphology is presented and an experimenatl system is developed. The kernel structure of the traffic signs is processed morphologically to obtain their skeletal functions. According to the skeletal features of the kernel structure, the traffic signs are classified and recognized. Experimental results show that it is suitable to choose the morphological skeletal function as the structural feature of the kernel structure of the traffic signs, and the new matching algorithm is robust because of its translational invariance.