中文摘要：

提出一种鲁棒的交通标志检测与分类方法。根据交通标志的颜色与形状特点进行突出颜色转换，以提取交通标志中的显著颜色，并利用基于仿射不变矩的形状测度有效地度量区域形状。先根据颜色转换特征进行粗分割，再利用形态学方法去除面积较小的区域，并填补区域中的孔洞，然后根据形状测度，检测交通标志，并对其进行粗分类。实拍图像的实验结果表明，该方法在不同尺寸、光照变化、旋转变形的情况下，依然能够表现出良好的交通标志检测和粗分类性能。

英文摘要：

A robust traffic sign detection and classification method is proposed,which combines color and shape characteristics of traffic sign. Firstly,dominant color transform is used to extract the prominent color in traffic signs. Then, small areas are removed and holes are filled using mathematical morphology methods. Thirdly, four shape measures based on affine invariant moment is defined to depict the shape of the segmented color region effectively. Finally, the traffic sign regions are confirmed and classified by dominant color and shape measure combinations. Experiment results of real images show high performance of the proposed method in detecting and classifying the traffic sign under different size, variant illumination, rotation and deformation circumstances.