模糊核空间下高速公路路面裂缝智能识别与评价

Intelligent Recognition and Evaluation of Highway Pavement Cracks in Fuzzy Kernel Space

为实现大规模高速公路路面裂缝快速精确识别和评价,提出了模糊核空间下的高速公路路面裂缝识别与评价模型。将裂缝的影像栅格数据空间转换到核空间,在高维空间中,构建了模糊识别策略,根据裂缝和路面的不同特征设计了加权模糊SVM二分类识别方法,最后设计了基于核理论的裂缝损伤评价指标(KPCI)。理论分析和试验结果显示在模糊核空间中进行路面裂缝的识别和评价更客观,识别精度更高。

In order to obtain quickly and accurately identify and evaluate the highway pavement cracks, a new identification and evaluation model for highway pavement cracks is designed in fuzzy kernel space. The cracks image data is transformed from pixel space to kernel space. In the high dimensional space, the fuzzy identification strategy is designed to get the weighted fuzzy SVM binary classification identification method according to the different features of cracks and pavement. Finally, the crack damage evaluation index is designed based on fuzzy kernel space （KPCI）. Both the theoretical analysis and experimental results show that it is more objective to pavement crack identification and evaluation the fuzzy kernel space, and the recognition accuracy is higher.