基于循环频谱特征提取的路面裂缝分类方法

详细信息    查看全文 | 推荐本文 |

英文篇名：Pavement cracks classification method based on cyclic spectrum feature extraction 作者：李鹏 ; 袁文婷 ; 蒋威 ; 马味敏 英文作者：LI Peng;YUANG Wenting;JIANG Wei;MA Weimin;School of Electronic and Information Engineering,Nanjing University of Information Science & Technology; 关键词：路面裂缝 ; 纹理特征 ; 频谱相关函数 ; 循环谱 英文关键词：pavement crack;;texture feature;;spectral correlation function;;cyclic spectrum 中文刊名：安徽大学学报(自然科学版) 英文刊名：Journal of Anhui University(Natural Science Edition) 机构：南京信息工程大学电子与信息工程学院; 出版日期：2019-07-15 出版单位：安徽大学学报(自然科学版) 年：2019 期：04 基金：国家自然科学基金资助项目(41075115);; 江苏省第11批六大高峰人才项目(2014-XXRJ-006);; 江苏省重点研发计划社会发展项目(BE201569);; 江苏省高校优势学科Ⅱ期建设工程项目 语种：中文; 页：54-57 页数：4 CN：34-1063/N ISSN：1000-2162 分类号：TP391.41;U418.6

摘要

提出基于循环频谱特征提取的路面裂缝分类方法.首先从图像像素获得2个1维信号,然后通过累积傅里叶变换计算每个信号的频谱相关函数,最后根据不同区域频谱相关函数的能量和标准差对路面图像进行分类.实验结果表明:与基于Gabor小波、离散小波、双树复小波及双树旋转复小波变换的特征提取分类方法相比,所提方法得到了较高的分类正确率.
        Pavement cracks classification method based on cyclic spectrum feature extraction was proposed.Firstly,two one-dimensional signals were obtained by pixels from each image.Then,the spectral correlation function of each signal was calculated by the accumulated Fourier transform.Finally,the pavement crack images were classified by the energy and standard deviation of the spectral characteristics of different regions of the correlation function.Compared with the classification method of feature extraction based on Gabor wavelet,discrete wavelet,double tree complex wavelet and double tree rotating complex wavelet transform,the experimental results showed that the proposed method could get high classification accuracy.

引文

[1] SUN L,KAMALIARDAKANI M,ZHANG Y.Weighted neighborhood pixels segmentation method for automated detection of cracks on pavement surface images[J].Joural of Computing on Civil Engineering,2015,30(2):04015021.
    [2] FANG Y,WU J J,HUANG B M.2D sparse signal recovery via 2D orthogonal matching pursuit[J].Science China(Information Sciences),2012(4):889-897.
    [3]章秀华,陈艳君,洪汉玉,等.基于加权融合纹理的路面裂缝检测[J].计算机与数字工程,2011,39(10):153-156.
    [4]高璐.路面裂缝图像检测和分类研究[D].洛阳:河南科技大学电子信息工程学院,2013.
    [5]马蓉贵,徐琨,刘飞飞.高速公路路面裂缝识别算法研究[J].交通信息与安全,2014(2):90-94.
    [6] KHACHANE M Y.Organ-based medical image classification using support vector machine[J].International Journal of Synthetic Emotions,2017,8(1):18-30.
    [7]邬志强,李大辉.肋骨皮质纹理特征提取分离算法设计[J].高师理科学刊.2016,36(6):35-37.
    [8]宋歌声,刘文慧,李进叶,等.利用超声图像纹理特征鉴别甲状腺实性节良性的研究[J].医学影像学杂志,2015,2015(4):612-616.
    [9]王红玉,冯筠,崔磊,等.应用显著纹理特征的医学图像配准[J].光学精密工程,2015,23(9):2657-2665.
    [10] AMIRANI M C,BEHESHTI A A.Texture classification using cyclic spectral function[C]//Proceedings of the First International Congress on Image and Signal Processing,2008:834-838.
    [11]戴幻尧,蒋鸿宇,李群.基于循环频谱分析的LPI信号特征检测新方法研究[J].信号处理学报,2009,25(5):781-786.
    [12] YU Z,CHEN H,LIU J,et al.Hybrid k-nearest neighbor chassifier[J].IEEE Transactions on Cybernetics,2016,46(6):1263-1275.
    [13] WANG K,HUANG Z,ZHONG Z.Multi-featrue fusion based lane under-standing algorithm[J].China Journal of Highway and Transport,2013,26(4):176-183.