基于图像特征及改进支持向量机算法的交通标志识别

Traffic sign recognition based on image feature and improved support vector machine algorithm

研究一种基于图像特征以及改进支持向量机算法的交通标志识别方法。使用颜色以及形状特征对交通标志图像特征进行提取,使用Gabor滤波方法增强处理交通标志图像。针对支持向量机识别算法的精度在很大程度上受到基本参数的影响,通常根据经验来进行参数的选取等问题,使用模拟退火算法对支持向量机的参数进行优化选择。研究结果表明,使用研究的交通标志检测方法的检测精度高于其他三种方法,使用颜色和形状特征提取的检测精度要高于单独使用颜色或形状特征提取方法。

A traffic sign recognition method based on image feature and improved support vector machine （SVM） algorithm is studied in this paper. The color feature and shape feature are used to extract the image feature of the traffic sign. The Gabor filtering method is adopted to perform the enhancement processing of the traffic sign image. Since the accuracy of the recognition algorithm based on SVM is affected by the basic parameters to a great extent, and the parameters are selected according to the experience usually, the simulated annealing algorithm is used to select the parameters of the support vector machine optimally. The research results show that the detection accuracy of the traffic sign detection method is higher than that of the other three methods, and the detection accuracy of color and shape features extraction method is higher than that of the single color feature extraction method or shape feature extraction method.