基于纹理特征的道路行人识别研究

道路行人识别已经成为智能车辆与车辆辅助驾驶系统的关键技术。研究一种基于纹理特征的行人特征提取和识别方法。纹理特征作为图像的重要特征在图像分析与识别有着广泛的应用。做出道路行人与非行人样本的纹理频谱,得出它们存在差异性的结论。首先用统计方法来计算样本图像的纹理特征,然后将支持向量机方法应用于道路行人的识别试验。试验结果表明,选择纹理特征作为道路行人特征具有很好的道路行人识别效果。

Study on pedestrian recognition based on texture feature

Pedestrian recognition has become a key technology in intelligent vehicles and intelligent assisted driving.A method of pedestrian feature extraction and recognition based on texture feature is proposed.Texture feature as an important feature of the image has been widely applied in image analysis and recognition.Meanwhile the texture spectrum of pedestrian and non-pedestrian sample was made,and the differences existence was concluded.First a statistical method was used to calculate the texture features of the sample images.Then the classification method of support vector machine for pedestrian recognition was tested,which results show that the effect of pedestrian recognition is better by selecting the texture features as the features of pedestrian.