基于LBP算子的路面类型识别研究

Research on Road Terrain Recognition Based on Improved LBP Operator

不同的路面类型会对车辆的制动、加速、变道等决策产生不同影响,因此实时获取路面类型信息对于提高智能汽车的安全性、舒适性等有着重要意义。论文提出一种基于LBP算子的路面类型识别方法,首先采集了四种车辆行驶常见路面的图像信息,并对图像进行了增广处理;然后使用LBP算子提取出路面图像的纹理特征,再采用PCA方式对纹理特征进行降维;最后通过分类器对数据进行训练与分类。实验结果表明该方法的最高分类准确率可以达到98.5%,有效提升了当前路面类型识别的精度。

Different pavement types will have diverse impacts on vehicle braking,acceleration,lane change and other decisions.It is of great significance to obtain pavement type information in time for improving the safety and comfort of intelligent vehicles.A road terrain recognition method based on LBP operator is proposed.Firstly,this paper collects the image information of four common road surfaces,and augments the images.LBP operator is used to extract the texture features of the pavement image,and PCA is used to reduce the dimension of the texture features.Finally,the classifier is used to train and classify the data.Experimental results show that the highest classification accuracy of this method can reach 98.5%,which effectively improves the accuracy of current pavement type recognition.