摘要： 针对交通场景图像中交通标志因颜色失真、形状失真和尺度变化导致漏检严重的问题,提出一种颜色特 征、形状特征和尺度特征的多特征协同方法。从交通场景图像计算颜色增强图,利用多阈值分割方法和闭合轮廓 曲率直方图链码表达,并对提取的形状轮廓的曲率直方图进行尺度归一化处理,融合颜色特征和归一化后的形状 特征构成区域的特征向量,采用支持向量机分类获得检测结果。实验结果表明,该算法在较低时间复杂度下,能有 效提高交通标志检测精度。

关键词: 交通标志, 多特征协同, 多阈值, 曲率链码, 尺度归一化, 支持向量机分类

Abstract: A new method is presented to improve traffic sign detection with cooperation of color,shape and scale features,especially under conditions of color distortion,shape deformation and scale variance. Color enhancement maps are generated from traffic scene images. Regions of interest are then extracted from the color enhancement maps using multiple thresholds of color, chain codes of the curvature histograms of closed contours are calculated and scale normalized for the contours. The Support Vector Machine(SVM) classifier is applied to classify the chain codes of the extracted traffic signs and the template signs. Experimental results demonstrate that this method is capable of improving traffic sign detection,with low time complexity.

Key words: traffic sign, multiple features cooperation, multiple thresholds, curvature chain code, scale normalization, Support Vector Machine(SVM) classification

中图分类号: 

• TP391