摘要
针对路面中常见裂缝识别分类难度大的问题,提出了一种基于二维经验模态分解和支持向量机结合的路面裂缝识别方法。通过对裂缝图像进行二维经验模态分解,提取其本征模函数并获得每个分量的瞬时幅值均值,放入特征向量,采用支持向量机对裂缝图像分类。实验表明,算法对指定样本进行测试,横向裂缝准确率为97.14%,纵向裂缝准确率为94.28%,龟裂裂缝准确率为94.28%,块状裂缝准确率为88.57%。算法可有效达到对裂缝进行分类的目的,小样本条件下,线性裂缝的分类正确率较高。
Aiming at the difficulty of identifying and classifying common cracks in pavement,a method of identifying pavement cracks based on two-dimensional empirical mode decomposition and support vector machine is proposed.Through two-dimensional empirical mode decomposition of fracture image,the eigenmode function is extracted and the instantaneous mean value of each component is obtained.The eigenvector is put into the eigenvector,and the support vector machine is used to classify the fracture image.Experiments show that the accuracy of the algorithm is 97.14%for transverse fracture,94.28%for longitudinal fracture,94.28%for tortoise fracture and 88.57%for block fracture.The algorithm can effectively achieve the purpose of crack classification.Under the condition of small samples,the classification accuracy of linear cracks is higher.
作者
夏翔
宋逸君
俞传富
李鹏
XIA Xiang;SONG Yijun;YU Chuanfu;LI Peng(Nanjing University of Information Science and Technology,School of Electronic&Information Engineering,Nanjing 210044,China)
出处
《电子器件》
CAS
北大核心
2020年第1期224-228,共5页
Chinese Journal of Electron Devices
基金
江苏省第11批六大高峰人才项目(2014-XXRJ-006)
国家自然科学基金项目(41075115)。
关键词
图像处理
裂缝分类
二维经验模态分解
支持向量机
特征提取
image processing
crack classification
bidimensional empirical modal decomposition
support vector machine
feature extraction
