基于多特征融合的路面破损图像自动识别技术研究
|
摘要
近年来，随着我国公路网的不断扩大，公路养护和管理问题日渐突出。为适应大规模、高效率和高质量的公路养护管理要求，路面管理系统（Pavement Management System，PMS）得到了广泛的推广应用。它改变了传统落后的公路管理模式，使公路管理决策更加客观化、信息化和科学化。路面管理系统的有效性依赖于各种数据的准确、实时获取，而路面破损数据是评价路面质量状况的关键指标之一。目前，国内外已研发了多种路面图像自动采集与处理系统，可在线获取路面图像，离线进行路面破损图像自动识别。但是这些系统普遍存在“成像照度不均匀”、“识别算法通用性差、运行效率不高”等问题，成为影响该系统推广应用的“瓶颈”。
近40年来，路面破损图像自动识别技术一直是交通信息工程与模式识别领域的经典难题，受到相关研究者的广泛关注。由于路面图像的多纹理性、多目标性、目标的弱信号性和图像光强的多变性，使得路面破损目标的识别难度相对较大。现有算法大多是建立在路面图像质量好、裂缝目标清晰的基础上而开发的，缺乏对复杂环境的适应性，难以满足工程应用的实际需求。
针对以上问题，本文在国内外的研究基础上，提出了一种基于多特征融合的路面破损图像自动识别方法。该方法对路面破损图像的灰度、纹理、边缘、形状等多种特征进行了定性和定量分析，选择出能够准确描述路面破损目标特性的关键特征，并通过多种融合规则和融合方法对这些特征进行融合，最终实现路面图像的准确分类和破损目标的精确分割。该研究成果将拓展多特征融合理论在交通领域的应用，同时为公路养护管理部门提供更为准确的路面破损数据。
论文主要在以下几个方面展开了研究工作:
（1）提出了一种“先分类，后识别”的路面破损图像处理流程，不同于传统的“先识别，后分类”的处理模式，该处理流程采用较低复杂度的算法对图像进行定性分析，将图像分为完好路面、松散类破损、裂缝类破损、补丁类破损四大类，然后再采用复杂度较高的算法对分类后的破损图像进行定量分析，该处理流程可以大大提高路面破损图像处理的效率和精度。
（2）提出了一种融合边缘和灰度特征的道路标线提取算法，该算法采用改进的beamlet算法提取路面二值图像中的直线边缘，实现了道路标线图像的快速筛选，该直线检测算法的耗时仅为传统Hough变换算法的1/10。然后根据检测出的直线边缘对图像进行初步分割，最后融合分割区域的灰度特征，采用分裂-合并算法实现道路标线区域的精确分割。实验结果表明：论文提出的道路标线分割算法准确率高于基于动态阈值的分割算法。
（3）提出了一种新的形状描述子——形状有效长度。该描述子基于傅立叶描述子重构方法得到目标的抽象轮廓，然后采用中轴变换得到形状的中轴，最后通过目标圆形度对该中轴长度进行校正。该描述子非常适合于区分裂缝类线状目标与非裂缝凸多边形目标。
（4）提出了一种基于纹理和形状特征融合的路面破损图像初始分类算法。该算法将局部对比度增强图像和整体灰度校正图像进行像素级融合，得到增强图像，并对增强图像进行三层小波分解以获得图像的纹理特征。然后采用改进的P分位法对增强图像进行二值化，并提取二值化图像中多个形状特征。最后采用BP神经网络对以上特征进行串行融合，实现了路面图像的有效初始分类，准确率高于采用单一特征的分类结果。
（5）提出了一种基于灰度、脊边缘及形状等多特征融合的裂缝类目标检测算法。该算法首先将路面图像分成多个互不重叠的图像子块，然后采用直方图估计算法获取每一个图像子块的最优阈值，并对其进行二值化得到灰度特征图像。然后采用多尺度脊边缘融合算法获得裂缝图像的边缘特征图像，并采用或运算方式对二者进行像素级融合。在此基础上，采用D-S证据理论和形状特征对融合后图像的斑点噪声进行去除。最后采用多种后续处理操作实现了裂缝目标的精确分割。实验结果表明：该算法的分割效果要优于文献中提出的三种常用的经典算法。
以上所有成果均在本文开发的原型系统上实现完成并得到应用。
In recent years, with the expanding of China's road network, road maintenance andmanagement issues have become increasingly prominent. In order to meet the requirements oflarge-scale, high efficiency and quality of road maintenance management, PavementManagement System (PMS) has been widelypromoted,which changed the traditional roadmanagement mode and made highway management decision-making more objective andscientific. The effectiveness of the pavement management system is dependent on accuratedata, and the pavement distress data is one of the key indicators to evaluate the status ofroad quality. At present, domestic and foreign research and developed a variety of pavementimage acquisition and processing systems, these systems can capture the pavement imageson the line, and process the images off the line.There problems such as uneven illumination,poor versatility and efficiency of the identification algorithm have become the" bottleneck"affecting the promotion and application of the system.
During the past40years, the automatic pavement distress image identificationtechnology has been a classic problem on the field of traffic information engineering andpattern recognition. Because the pavement images have complex textures, object types, objectsignal and illumination, it is harder to identify them than other images.Existing algorithms aredeveloped on the pavement image with good qualityand clear cracks, which is not suitable tocomplex environments, it is difficult to meet the needs of engineering applications.
Aiming at the problems above, on the basis of both domestic and international research,this paper proposed a new automatic pavement distress identification algorithm based onmulti-feature fusion technology. This paper selected the pavement distress as the researchobjects such as road marks, road cracks, loosen broken and surface patch. Through analyzingof the grayscale, texture, edges and geometry shape of the distress objects, severalfeature-fusion models are established in this paper. With these models, good segmentationresults have been achieved.This research has fruits and innovation as follows:
(1) This desertation proposed a new pavement distress image processing flow of“classification before segmentation", its processing order is contrary to that from thetraditional methods. The proposed flow employs low complexity algorithm for qualitativeanalyses of the images. After the analyses, the images are classified into4categories: theintact pavement, loosen broken, cracks and patches. Then the high complexity algorithms areemployed for quantitative analyses of the different images. Finally, the post processes such asrecognition, segmentation, measurement and evaluation of the pavement distresses are done.This process flow can greatly improve the efficiency and accuracy of the algorithm for theidentification of the pavement distress objects, which makes various algorithm modules havebetter pertinence.
(2) A new extraction algorithm fused the edge and grayscale characteristics of the roadmark images is proposed, the algorithm uses beamlet transform to extract the straight edge ofa binary image, which realizes the rapid screening of the pavement images including road marks. With the extracted straight edges, a pavement image can be segmented for manyregions. Finally, according to the grayscale characteristics of these regions, the split-mergealgorithm is used to segment the road mark area accurately. In experiments, the results showthat the efficiency and detection accuracy of the proposed algorithm are much higher than thatof the dynamic threshold segmentation algorithm.
(3) A new shape descriptor–the Equivalent Length of the Connected Component isproposed. This descriptor gets the abstract outline of shape objects based on the Fourierdescriptor reconstruction method, and then the medial axis transform is used to gain themedial axis of the shape. Finally, the length of the medial axis is corrected by the object’sroundness. Experimental results show that this descriptor is very suited to distinguish thelinear target from the convex polygon.
(4) A pavement distress initial classification algorithm is proposed based on the fusion oftexture and shape features. First, the proposed algorithm fuses the local contrast enhancedimagePt with the global grayscale corrected imageI '
tto get an enhanced pavement distressimage. Secondly, the enhanced image is decomposed with three-layer wavelet transform,which obtains the texture features of the whole image. Thirdly, an improved P-tile method isused to obtain the binarization image. From the binarization image, three shape features areextracted such as AA (the Average Area of all Connected Components), AM (the Area of theMaximum Connected Component), EL (the Equivalent Length of the longest ConnectedComponent). Eventually the neural network is used to fuse the texture and shape featuresserially. This algorithm achieves an effective classification of the pavement distress image,and its accuracy is higher than the algorithm with single type features.
(5) A crack target detection algorithm fused grayscale, ridge edge and shapecharacteristic is proposed. First, a pavement crack image is divided into manynon-overlapping sub-blocks, and then a histogram estimation method is used to take theoptimal threshold of each sub-block. Through these thresholds,the grayscale feature image ofthe pavement cracks is obtained. Secondly, the edge feature image of the pavement cracks isgot based on multi-scale ridge edges fusion algorithm, and then pixel-level fusion is operatedon the feature images with OR operation. Thirdly, the noise of the pixel-level fused image isremoved through the DS evidence theory and shape analysis method. At last, the improvedconnection algorithm is adopted to achieve the accurate segmentation of the crack target. Theexperimental results show that the segmentation results of the proposed algorithm are superiorto that from the other three classic segmentation algorithms.
All the algorithms and methods proposed in this desertation have been realized in thereal developed system.
近40年来，路面破损图像自动识别技术一直是交通信息工程与模式识别领域的经典难题，受到相关研究者的广泛关注。由于路面图像的多纹理性、多目标性、目标的弱信号性和图像光强的多变性，使得路面破损目标的识别难度相对较大。现有算法大多是建立在路面图像质量好、裂缝目标清晰的基础上而开发的，缺乏对复杂环境的适应性，难以满足工程应用的实际需求。
针对以上问题，本文在国内外的研究基础上，提出了一种基于多特征融合的路面破损图像自动识别方法。该方法对路面破损图像的灰度、纹理、边缘、形状等多种特征进行了定性和定量分析，选择出能够准确描述路面破损目标特性的关键特征，并通过多种融合规则和融合方法对这些特征进行融合，最终实现路面图像的准确分类和破损目标的精确分割。该研究成果将拓展多特征融合理论在交通领域的应用，同时为公路养护管理部门提供更为准确的路面破损数据。
论文主要在以下几个方面展开了研究工作:
（1）提出了一种“先分类，后识别”的路面破损图像处理流程，不同于传统的“先识别，后分类”的处理模式，该处理流程采用较低复杂度的算法对图像进行定性分析，将图像分为完好路面、松散类破损、裂缝类破损、补丁类破损四大类，然后再采用复杂度较高的算法对分类后的破损图像进行定量分析，该处理流程可以大大提高路面破损图像处理的效率和精度。
（2）提出了一种融合边缘和灰度特征的道路标线提取算法，该算法采用改进的beamlet算法提取路面二值图像中的直线边缘，实现了道路标线图像的快速筛选，该直线检测算法的耗时仅为传统Hough变换算法的1/10。然后根据检测出的直线边缘对图像进行初步分割，最后融合分割区域的灰度特征，采用分裂-合并算法实现道路标线区域的精确分割。实验结果表明：论文提出的道路标线分割算法准确率高于基于动态阈值的分割算法。
（3）提出了一种新的形状描述子——形状有效长度。该描述子基于傅立叶描述子重构方法得到目标的抽象轮廓，然后采用中轴变换得到形状的中轴，最后通过目标圆形度对该中轴长度进行校正。该描述子非常适合于区分裂缝类线状目标与非裂缝凸多边形目标。
（4）提出了一种基于纹理和形状特征融合的路面破损图像初始分类算法。该算法将局部对比度增强图像和整体灰度校正图像进行像素级融合，得到增强图像，并对增强图像进行三层小波分解以获得图像的纹理特征。然后采用改进的P分位法对增强图像进行二值化，并提取二值化图像中多个形状特征。最后采用BP神经网络对以上特征进行串行融合，实现了路面图像的有效初始分类，准确率高于采用单一特征的分类结果。
（5）提出了一种基于灰度、脊边缘及形状等多特征融合的裂缝类目标检测算法。该算法首先将路面图像分成多个互不重叠的图像子块，然后采用直方图估计算法获取每一个图像子块的最优阈值，并对其进行二值化得到灰度特征图像。然后采用多尺度脊边缘融合算法获得裂缝图像的边缘特征图像，并采用或运算方式对二者进行像素级融合。在此基础上，采用D-S证据理论和形状特征对融合后图像的斑点噪声进行去除。最后采用多种后续处理操作实现了裂缝目标的精确分割。实验结果表明：该算法的分割效果要优于文献中提出的三种常用的经典算法。
以上所有成果均在本文开发的原型系统上实现完成并得到应用。
In recent years, with the expanding of China's road network, road maintenance andmanagement issues have become increasingly prominent. In order to meet the requirements oflarge-scale, high efficiency and quality of road maintenance management, PavementManagement System (PMS) has been widelypromoted,which changed the traditional roadmanagement mode and made highway management decision-making more objective andscientific. The effectiveness of the pavement management system is dependent on accuratedata, and the pavement distress data is one of the key indicators to evaluate the status ofroad quality. At present, domestic and foreign research and developed a variety of pavementimage acquisition and processing systems, these systems can capture the pavement imageson the line, and process the images off the line.There problems such as uneven illumination,poor versatility and efficiency of the identification algorithm have become the" bottleneck"affecting the promotion and application of the system.
During the past40years, the automatic pavement distress image identificationtechnology has been a classic problem on the field of traffic information engineering andpattern recognition. Because the pavement images have complex textures, object types, objectsignal and illumination, it is harder to identify them than other images.Existing algorithms aredeveloped on the pavement image with good qualityand clear cracks, which is not suitable tocomplex environments, it is difficult to meet the needs of engineering applications.
Aiming at the problems above, on the basis of both domestic and international research,this paper proposed a new automatic pavement distress identification algorithm based onmulti-feature fusion technology. This paper selected the pavement distress as the researchobjects such as road marks, road cracks, loosen broken and surface patch. Through analyzingof the grayscale, texture, edges and geometry shape of the distress objects, severalfeature-fusion models are established in this paper. With these models, good segmentationresults have been achieved.This research has fruits and innovation as follows:
(1) This desertation proposed a new pavement distress image processing flow of“classification before segmentation", its processing order is contrary to that from thetraditional methods. The proposed flow employs low complexity algorithm for qualitativeanalyses of the images. After the analyses, the images are classified into4categories: theintact pavement, loosen broken, cracks and patches. Then the high complexity algorithms areemployed for quantitative analyses of the different images. Finally, the post processes such asrecognition, segmentation, measurement and evaluation of the pavement distresses are done.This process flow can greatly improve the efficiency and accuracy of the algorithm for theidentification of the pavement distress objects, which makes various algorithm modules havebetter pertinence.
(2) A new extraction algorithm fused the edge and grayscale characteristics of the roadmark images is proposed, the algorithm uses beamlet transform to extract the straight edge ofa binary image, which realizes the rapid screening of the pavement images including road marks. With the extracted straight edges, a pavement image can be segmented for manyregions. Finally, according to the grayscale characteristics of these regions, the split-mergealgorithm is used to segment the road mark area accurately. In experiments, the results showthat the efficiency and detection accuracy of the proposed algorithm are much higher than thatof the dynamic threshold segmentation algorithm.
(3) A new shape descriptor–the Equivalent Length of the Connected Component isproposed. This descriptor gets the abstract outline of shape objects based on the Fourierdescriptor reconstruction method, and then the medial axis transform is used to gain themedial axis of the shape. Finally, the length of the medial axis is corrected by the object’sroundness. Experimental results show that this descriptor is very suited to distinguish thelinear target from the convex polygon.
(4) A pavement distress initial classification algorithm is proposed based on the fusion oftexture and shape features. First, the proposed algorithm fuses the local contrast enhancedimagePt with the global grayscale corrected imageI '
tto get an enhanced pavement distressimage. Secondly, the enhanced image is decomposed with three-layer wavelet transform,which obtains the texture features of the whole image. Thirdly, an improved P-tile method isused to obtain the binarization image. From the binarization image, three shape features areextracted such as AA (the Average Area of all Connected Components), AM (the Area of theMaximum Connected Component), EL (the Equivalent Length of the longest ConnectedComponent). Eventually the neural network is used to fuse the texture and shape featuresserially. This algorithm achieves an effective classification of the pavement distress image,and its accuracy is higher than the algorithm with single type features.
(5) A crack target detection algorithm fused grayscale, ridge edge and shapecharacteristic is proposed. First, a pavement crack image is divided into manynon-overlapping sub-blocks, and then a histogram estimation method is used to take theoptimal threshold of each sub-block. Through these thresholds,the grayscale feature image ofthe pavement cracks is obtained. Secondly, the edge feature image of the pavement cracks isgot based on multi-scale ridge edges fusion algorithm, and then pixel-level fusion is operatedon the feature images with OR operation. Thirdly, the noise of the pixel-level fused image isremoved through the DS evidence theory and shape analysis method. At last, the improvedconnection algorithm is adopted to achieve the accurate segmentation of the crack target. Theexperimental results show that the segmentation results of the proposed algorithm are superiorto that from the other three classic segmentation algorithms.
All the algorithms and methods proposed in this desertation have been realized in thereal developed system.
引文
[1] http://mobile.dili360.com/tbch/2009/0915934.shtml中国国家地理网：中国古代国道.
[2]之超.中国道路史话,道路交通管理,1998(4):20-21.
[3] http://finance.eastday.com/economic/m1/20111230/u1a6286393.html2010年公路水路交通运输行业发展统计公报.
[4]啜二勇.国外路面自动检测系统发展综述,交通标准化,2009(17):30-33.
[5]交通运输部综合规划司2010年一季度交通运输经济运行情况公报,2010年04月14日.
[6] Gang Wang,Xiu-wei xu, Liang Xiao, Algorithm Based on the finite ridgelet transform for enhancingfaint pavement cracks, Optical Engineering,47(1):017004,2008/01.
[7]交通行业标准《JTJ073.2-2001公路沥青路面养护技术规范》.
[8] J. Baker, B. Dahlstrom, K. Longenecker, T. Buu, Video image distress analysis technique forIdahotransportation department pavement management system, Transportation Research Record1117(1987)159-163.
[9] H.D. Cheng, M. Miyojim, Automatic pavement distress detection system, Journal of InformationSciences108(1998)219-240.
[10] Kelvin C P W,Highway data collection and information management [C].The3rd InternationalConference on Road&Airfield Pavement Technology,1998.
[11] Yaxiong Huang,Bugao Xu, Automatic inspection of pavement cracking distress[J], Journal ofElectronic Imaging15(1),013017(Jan–Mar2006).
[12] Salvatore Cafiso, Alessandro Di Graziano. Evaluation Of Pavement Surface Distress Using DigitalImage Collection And Analysis, Seventh International Congress on Advances in Civil Engineering,October11-13,2006,Yildiz Technical University, Istanbul, Turkey.
[13]张宏,英红.沥青路面裂缝图像识别技术研究进展,华东公路2009(4)：81-84.
[14]张娟,沙爱民,孙朝云,高怀钢.基于相位编组法的路面裂缝自动识别[J],中国公路学报,2008,21(2):39-42.
[15] Kim, Jung Yong, Development of new automated crack measurement algorithm using laser imagesof pavement surface, Ph.D. Paper, THE UNIVERSITY OF IOWA,2008.
[16] http://www.pavemetrics.com.
[17] http://www.internationalcybernetics.com.
[18] Tarek A. Monem, Amr A. Oloufa, and Hesham Mahgoub,Asphalt Crack Detection UsingThermography, InfraMation2005Proceedings,2005:1-12.
[19] Kelvin C.P. WANG, Elements of automated survey of pavements and a3D methodology, Journal ofModern Transportation,2011,19(1):51-57.
[20] Kelvin C. P. Wang, Weiguo Gong. Real-Time Automated Survey System of Pavement Cracking inParallel Environment[J].Journal of infrastructure systems. September2005:154-164.
[21] H.D. Cheng, M. Miyojim. Automatic pavement distress detection system[J]. Journal of InformationSciences,1998(108):219-240.
[22]王刚．路面病害光学无损检测技术[D]．南京：南京理工大学,2007．
[23]王建锋,激光路面三维检测专用车技术与理论研究,西安：长安大学,2010.
[24] H. D.Cheng, Chen, J. R., Glazier, C., and Hu, Y. G.1999. Novel approach to pavement distressdetection based on fuzzy set theory[J]. Journal of computing in civil engineering,13(4):270–280.
[25]高建贞,任明武,唐振民等.路面裂缝的自动检测与识别[J].计算机工程,2003,Vol.29(2):149-150.
[26]唐磊,赵春霞,王鸿南等.路面图像增强的多偏微分方程融合法[J].中国图象图形学报,2008,Vol.13(9)：1661-1666.
[27]王兴建,秦国锋,赵慧丽.基于多级去噪模型的路面裂缝检测方法[J].计算机应用,2010, Vol.30(6):1606-1612.
[28]闫茂德,伯绍波,李雪,一种自适应模糊的局部区域图像增强算法[C],第26届中国控制学会会议论文集,2007：308-311.
[29]刘玉臣,王国强,林建荣.基于模糊理论的路面裂缝图像增强算法[J].养护机械与施工技术,2006（2）：35-37.
[30] Yongxia Zuo; Guoqiang Wang; Chuncheng Zuo. Wavelet Packet Denoising for Pavement SurfaceCracks Detection,International Conference on Computational Intelligence and Security,2008. CIS'08. Vol.2:481–484.
[31] Daqi Zhang,Shiru Qu, Li He, Shuang Shi. Automatic ridgelet image enhancement algorithm forroad crack image based on fuzzy entropy and fuzzy divergence[J]. Optics and Lasers in Engineering,2009(47):1216–1225.
[32]李清泉,胡庆武.基于图像自动匀光的路面裂缝图像分析方法[J].公路交通科技,2010, Vol.27(4):1-5.
[33] K. R. Kirschke, S. A. Velinsky. Histogram-based approach for automated pavement-crack sensing[J]. Journal of Transportation Engineering,1992, Vol.118(4):700-710.
[34] Siriphan Jitprasithsiri. Development of a new digital pavement image processing algorithm forunified crack index computation[D]. A Dissertation Submitted to the Faculty of the University ofUtah,1997.
[35] Oh, H., Garrick, N. W., and Achenie, L. E. K.1997.“Segmentation algorithm using iteratedclipping for processing noisy pavement images.” Proc.,2nd Int. Conf. of Imaging Technologies:Techniques and Applications in Civil Engineering, ASCE, Reston, Va.,138–147.
[36] H. D. Cheng, Shi, X. J., and Glazier, C.2003.“Real-time image thresh-olding based on samplespace reduction and interpolation approach. J. Comput. Civ. Eng.,174,264–272.
[37]孙波成,邱延峻.路面裂缝图像处理算法研究[J].公路交通科技,2008,25（2）：64-68.
[38]李清泉,刘向龙.路面裂缝影像几何特征提取算法[J].中国科技论文在线,2007年7月, Vol2,No.7:517-522.
[39]伯绍波,闫茂德,孙国军等.沥青路面裂缝检测图像处理算法研究[J].微计算机信息(管控一体化).2007年第23卷第4-3期：280-282.
[40]冯永安,刘万军.边缘检测改进算法在路面破损检测中的应用[J].辽宁工程技术大学学报,2007年11月,第26卷增刊.176-178.
[41]李晋惠.公路路面裂缝类病害图像处理算法研究[J].计算机工程与应用2003.35.：212-213.
[42] Siwaporn Sorncharean, Suebskul Phiphobmongkol. Crack Detection on Asphalt Surface ImageUsing Enhanced Grid Cell Analysis[J].4th IEEE International Symposium on Electronic Design,Test&Application,2008：49-54.
[43]唐磊,赵春霞,王鸿南,邵文泽.基于图像三维地形模型的路面裂缝自动检测,计算机工程,vol.34(5):20-21.
[44] Peggy Subirats, Olivier Fabre, Jean Dumoulin etc. A combined wavelet-based image processingmethod for emergent crack detection on pavement surface images. European Signal ProcessingConference (EUSIPCO2004).2004,257-260.
[45] Zhou, J., Huang, P. S.,&Chiang, F.-P. Wavelet-based pavement distress detection and evaluation.Optical Engineering,2006,45:027007-1~027007-10.
[46] Kelvin C.P. Wang, Li, Q. and Gong, Weiguo "Wavelet-Based Edge Detection with à TrousAlgorithm for Pavement Distress Survey", Journal of the Transportation Research Board,2007,2024.
[47] Zhang-Xia Ma, Chun-Xia Zhao, Ying-kun Hou. Pavement Distress Detection Based onNonsubsampled Contourlet Transform[C].2008International Conference on Computer Science andSoftware Engineering,2008:28-31.
[48]王刚,贺安之,肖亮.基于高速公路裂纹局部线性特征内容的脊波变换域算法研究[J].光学学报,2006,26(3):341-346.
[49]初秀民,王荣本.基于神经网络的沥青路面破损图像识别研究[J].武汉理工大学学报(交通科学与工程版),2004,28(3):373-376.
[50]储江伟,初秀民,王荣本等.沥青路面破损图像特征提取方法研究[J].中国图象图形学报A辑,2003,8A(10):1211-1217.
[51]初秀民,王荣本,储江伟等.沥青路面破损图像分割方法研究[J].中国公路学报,2003,16(3):11-14.
[52]储江伟,初秀民,王荣本.沥青路面破损图像自动检测系统设计[J].光学技术,2003,29(3):316-319.
[53]初秀民,严新平,陈先桥.路面破损图像二值化方法研究[J].计算机工程与应用,2008,44(28):161-165.
[54] Henrique Oliveira, Paulo Lobato Correia. Supervised strategies for cracks detection in images ofroad pavement flexible surfaces.16th European Signal Processing Conference (EUSIPCO2008),Lausanne, Switzerland, August25-29,2008.
[55]王华,朱宁,王祁.公路路面分形纹理特征分析和分类[J].哈尔滨工业大学学报,2005,37(6):816-818.
[56]胡勇,赵春霞,郭志波.基于多尺度布朗运动模型的路面破损检测[J].计算机工程与应用,2008,44(3):234-235.
[57]王华,朱宁,王祁.应用计盒维数方法的路面裂缝图像分割[J].哈尔滨工业大学学报,2007,39(1):142-144.
[58]章秀华,洪汉玉,侯佳等.路面破损图像实时检测方法研究[J].电子设计工程,2009,17(6):36-40.
[59] N. Tanaka, K. Uematsu. A crack detection method in road surface images usingmorphology,Workshop on Machine Vision Applications,(1998)154-157.
[60] Yan, M.D., Bo, S.B., Xu, K., He, Y.Y.,2007. Pavement crack detection and analysis for high-gradehighway. In: Proc. Internat. Conf. on Electronic Measurement and Instruments (ICEMI’07):548–552.
[61] Liu, Fanfan; Xu, Guoai; Yang, Yixian, et al. Novel approach to pavement cracking automaticdetection based on segment extending(C].//Proceedings一2008InternationalSymposium onKnowledge Acquisition and Modeling, Inst. of Elec. and Elec. Eng.Computer Society, Wuhan2008:610-614.
[62]刘凡凡,徐国爱,肖靖等.基于连通域相关及Hough变换的公路路面裂缝提取[J].北京邮电大学学报,2009年4月,第32卷,第2期.
[63] Sylvie Chambona,Peggy Subiratsb, Jean Dumoulina. Introduction of a wavelet transform based on2D matched filter in a Markov Random Field for fine structure extraction: Application on road crackdetection[J]. Proc. of SPIE-IS&T Electronic Imaging,2009, SPIE Vol.7251:72510A.1-72510A.12.
[64]张洪光,王祁,魏玮.基于人工种群的路面裂纹检测[J].南京理工大学学报,2005,29(4):128-132.
[65] Alekseychuk, O.2006. Detection of crack-like indications in digital radiography by globaloptimisation of a probabilistic estimation function.Ph.D. thesis, BAM-Dissertationsreihe, Band18,Berlin.
[66]李清泉,邹勤,毛庆洲.基于最小代价路径搜索的路面裂缝检测[J].中国公路学报,2010,23(6):28-33.
[67] Vivek Kaul, Yichang Tsai, and Anthony Yezzi. Detection of Curves with Unknown Endpoints usingMinimal Path Techniques. Proceedings of the British Machine Vision Conference, pages62.1-62.12.BMVA Press, September2010.
[68]李刚.基于灰色系统理论的路面图像裂缝检测算法研究[D].博士学位论文,武汉,武汉理工大学,2010.
[69] Chua, K. M.,&Xu, L.(1994). Simple procedure for identifying pavement distresses from videoimages[J]. Journal of Transportation Engineering,120(3),412–431.
[70] Acosta, J. A., Figueroa, J. L.,&Mullen, R. L.(1995). Algorithm for pavement distress classificationby video image analysis. Transportation Research Record,1505,27–38.
[71] Miyojim M, H. D. Cheng. Novel system for automatic pavement distress detection[J]. ASCE,1998,12(3):145-152.
[72] H. D. Cheng, Jiang, X., Li, J.,&Glazier, C.(1999). Automated real time pavement distress analysis.Transportation Research Record,1655,55–64.
[73] Kelvin C.P. Wang, Watkins,&Kuchikulla.(2002). Digital distress survey of airport pavementsurface. Federal Aviation Administration Airport Technology Transfer Conference.
[74] Lee, David Lee.(2003). A robust position invariant artificial neural network for digital pavementcrack analysis. Technical Report TRB2003-000996.
[75] Zhou, J., Huang, P. S.,&Chiang, F.-P. Wavelet-based pavement distress detection and evaluation.Optical Engineering,2006,45：027007-1~027007-10.
[76] Lee,B.J and H.Lee,A Robust Position Invariant Neural Network for Digital Pavement CrackAnalysis.Computer-Aided Civil and Infrastructure Engineering,2004,19:105-118.
[77]丁爱玲,焦李成.基于支撑矢量机的路面破损识别[J].长安大学学报(自然科学版),2007,27(2):34-37.
[78] Fereidoon Moghadas Nejad, Hamzeh Zakeri. An optimum feature extraction method based onWavelet–Radon Transform and Dynamic Neural Network for pavement distress classification[J].Expert Systems with Applications38(2011)9442–9460.
[79]肖旺新,严新平,张雪.基于混合密度因子的路面破损自动识别研究[J].交通运输工程与信息学报,2005,3(2):19-26.
[80]肖旺新,张雪,黄卫.基于破损密度因子的路面破损识别新方法[J].交通运输工程与信息学报,2004,2(2):82-89.
[81]李清泉,刘向龙.路面影像破损加权评定方法[J].中国公路学报,2009,22(4):44-49.
[82]Xiao Mei,automatic detection of pavement suface crack depth on florida roadways, Phd paper, SouthFlorida,Department of Civil and Environmental Engineering College of The University of SouthFlorida,2001.
[83]王泽民.混凝土路面破损评价与维修对策研究[D],博士学位论文,西安,长安大学,2007
[84]王刚.路面病害光学无损检测技术[D],博士学位论文,南京理工大学,2007.
[85]唐磊.基于图像分析的路面病害自动检测[D],博士学位论文,南京,南京理工大学,2007
[86]肖旺新.路面破损图像自动识别关键技术研究[D],博士学位论文,南京,东南大学,2005
[87] http://baike.baidu.com/view/297217.htm百度百科
[88]蔡涛.面向目标识别的图像特征融合提取技术研究[D],博士学位论文,长沙,国防科学技术大学,2000,10:2-3
[89]刘帅师,田彦涛,万川.基于Gabor多方向特征融合与分块直方图的人脸表情识别方法,自动化学报,2011,37（12）：1455-1463.
[90]潘瑜,郑钰辉,孙权森,等.基于PCA和总变差模型的图像融合框架,计算机辅助设计与图形学学报,2011,23（7）：1200-1210.
[91]孙权森,曾生根,王平安,等.典型相关分析的理论及其在特征融合中的应用,计算机学报,2005,28（9）：1524-1533.
[92]潘瑜,孙权森,夏德深.基于PCA分解到图像融合框架,计算机工程,2011,37（13）：210-212
[93]李建科,赵保军,张辉,等.DCT和LBP特征融合的人脸识别,北京理工大学学报,2010,30（11）：1355-1358.
[94]张俊虎,郝晓剑,刑昊,人脸图像识别中的PCA算法实现,模式识别,2010,26（7-1）：186-187.
[95]韩智,刘昌平.基于多特征融合的指纹识别方法,计算机科学,2010,37（7）：255-259.
[96]谭利,李彬,田联房,等.基于多特征融合跟踪的微小肺结节识别算法,生物医学工程学杂志,2011,28（3）：437-441.
[97]王坤峰,李镇江,汤淑明.基于多特征融合的视频交通数据采集方法,自动化学报,2011,37（3）：322-330.
[98]顾鑫,王海涛,汪凌峰,等.基于不确定性度量的多特征融合跟踪,自动化学报,2011,37（5）：550-559.
[99]王莉莉,郝爱民,何兵,等.基于D-S证据理论的城市航拍道路提取方法,软件学报,2005,16（9）：1534-1541.
[100]张雷,肖梅,马建,等,基于信息融合的车道检测算法,郑州大学学报（工学版）,2010,31（1）：98-102.
[101]朱国康,王运锋,基于多特征融合的道路交通标志检测,信号处理,2011,27（10）：1616-1620.
[102]孙显,王宏琦,张道兵,等.基于多特征融合的城市遥感图像自动解译方法,光子学报,2010,39（1）：178-183.
[103]林国余,陈旭,张为公.基于多信息融合优化的鲁棒性车道检测算法,东南大学学报（自然科学版）,2010,40（4）：771-777.
[104]李远征,卢朝阳,高全学,等.基于多特征融合的均值迁移粒子滤波跟踪算法,电子与信息学报,2010,82（2）：411-416.
[105]周静,黄心汉,彭刚.基于多特征融合的飞机目标识别,华中科技大学学报（自然科学版）,2009,37（1）：38-41.
[106]宁慧君,李映,胡杰.基于多尺度特征融合的SAR图像分割,计算机工程与应用,2011,47（14）：196-199.
[107]李正周,马齐佑,郑微,等.基于多特征融合的微弱红外运动目标跟踪方法,强激光与粒子束,2011,23（1）：54-58.
[108]陈皓,马彩文,陈岳承,等.基于多特征融合的复杂背景下弱小多目标检测和跟踪算法,光子学报,2009,38(9):2444-2448.
[109]王莹,李文辉,基于多特征融合的高精度视频火焰检测算法,吉林大学学报（工学版）,2010,40（3）：769-775.
[110]李先锋,朱伟兴,花小朋,等.基于D-S证据理论的决策级多特征融合苹果分级方法,农业机械学报,2011,42（6）：188-192.
[111]李先锋,朱伟兴,孔令东,等.基于SVM和D-S证据理论的多特征融合杂草识别方法,农业机械学报,2011,42（11）：164-168.
[112]范振东,吝宁.基于并行特征融合的金融汉字特征提取方法,计算机应用,2007,27（6）：127-129.
[113] WaldL.,SomeTerms of Referenee in Data Fusion,IEEE Trans.on GRS,1999,37(3):l-3.
[114] David L.Hall, James Llinas,An Introduction to Multisensor DataFusion,Proc.IEEE,1997,85(l):6-28.
[115]赵宗贵等,多传感器数据融合,北京:电子部27所,1995.
[116]韩崇昭,朱洪艳,段战胜等.多源信息融合(第二版)[M].北京:清华大学出版社,2010,9.
[117] Yang J.,Yang J.Y.,Zhang D.,Lu J.F.Feature fusion Parallel strategy vs.serial strategy.PatternRecognition[J],2003,Vol.36(6):1369-1381.
[118] Xuguang Zhang,at el..Integrated Intensity,Orientation Code and Spatial Information for RobustTracking[C].ICIEA2007:1-4.
[119] Wang Dawei,Ge Wei,Wang Yan jie.Using BBPSO for Feature Select in Feature-Level FusionTarget Recognition[C].4th IEEE conference on industrial electronics and applications, Xi’an,China,May.2009:1-4.
[120] Fabien Scalzo, George Bebis, Mircea Nicolescu, Leandro Loss. Evolutionary Learning of FeatureFusion Hierarchies[C].IEEE ICPR2008, Dec.2008:1-4.
[121]覃征,鲍福民,李爱国.数字图像融合[M].西安:西安交通大学出版社,2004.
[122]王大伟.基于特征级图像融合的目标识别技术研究[D].中国科学院研究生院,博士学位论文,2010.
[123]王大伟,陈浩,王延杰.核典型相关分析的融合人脸识别算法[J].激光与红外,2009(11):1241-1245.
[124] Pavement Distress Identification Manual for the NPS Road Inventory Program Cycle4,2006-2009.
[125]毛成.沥青路面裂纹形成机理及扩展行为研究[D],博士学位论文,成都,西南交通大学,2004.
[126]吴国雄.水泥混凝土路面开裂机理及破坏过程研究[D],博士学位论文,成都,西南交通大学,2003.
[127](美)冈萨雷斯(Gonzalez,R.C)等著.数字图像处理(MATLAB版)[M].北京:电子工业出版社,2005,9.
[128]程宏煌,戴卫恒.图像分割方法综述[J].电信快报,2000(10)：39-41.
[129] Kittler J,Illingworth J. Minimum error threshol-ding[J]. Pattern Recognition,1986,19(1)：41-47.
[130] Chow C K,Kaneko T. Automatic boundary detec-tion of the left ventricle fromcineangiograms[J]. Computers and Biomedical Research,1972(5)：338-410.
[131]杨喜宁,段建民,高德芝,郑榜贵.基于改进Hough变换的车道线检测技术.计算机测量与控.2010,18(2)：292-298.
[132] D.L.Donoho,X.Huo. Beamlets pyramids：A new form of multi-resolution analysis,suited forextracting lines,curves,and objects from very noise image data[C].Proceedings ofSPIE,2000,4119(1)：434-444
[133] D.L.Donoho,X.Huo. Beamlets and Multi-scale Image Analysis[R],Multi-scale andMulti-resolution Methods[C].Springer Lecture Notes in Computational Science andEngineering,2002,20：149-196
[134] D.L.Donoho,O.Levi,J.L.Starch,et al. Multi-scale geometric analysis for3.d catalogues,Astronomical Data AnalysisⅡ[J].Proceeding of SPIE,2002,4847：101-111.
[135]史勤峰.Beamlet及其在SAR图像边缘检测中的应用[D].硕士学位论文,西安市：西北工业大学,2006.
[136]屈庆春.基于beamlet的多尺度图像分析及其在CT图像处理中的应用[D].硕士学位论文,济南市：山东大学,2005.
[137]邓双城,蒋力培,焦向东等.基于束波变换的焊缝图像处理技术[J].焊接学报,2009,30(1)：68-72.
[138]焦李成,侯彪,王爽等.图像多尺度几何分析理论与应用-后小波分析理论与应用[M].西安：西安电子科技大学出版社,2008：210-238.
[139]侯阿临,王明明,史东承等.基于离散Beamlet变换的线特征检测算法[J].计算机安全,2009,17(3)：8-11.
[140] Qin F S,Yan N Z. Adaptive Linear feature detection based on beamlet[J].IEEE,2009,3981-3984.
[141]潘广林,梁彦,金术玲等.基于子束变换的自适应线性特征检测[J].火力与指挥控制,2008,33(10)：20-23.
[142]沈丽琴,胡栋梁,戚飞虎.基于知识的线状目标的综合理解,自动化学报,1997.23C61：839-841.
[143] Frank R J, Grabowski T J, Damasi o H. Voxel vise. Percentage tissue segmentation of human brainMRI[C]. In:25th AnnualMeeting, Society for Neuroscience, Society of Neuroscience[C],Washington,1995:694.
[144]邹丹平,冯涛,李咸伟,刘其真,基于EM的直方图逼近及其应用[J],中国图象图形学报,2005,10(11):1458-1461.
[145]李亚平,杨华,陈霞,基于EM和BIC的直方图拟合方法应用于遥感变化检测阈值确定[J],遥感学报,2008,12(1):85-91.
[146]阎冬梅.基于特征融合的遥感影像典型现状目标提取技术研究[D],博士学位论文,北京,中国科学院研究生院,2003.
[147] Bob Zhang, LinZhangb, LeiZhangb, FakhriKarraya Retinal vessel extraction by matched filter withfirst-order derivative of Gaussian, Computers in Biology and Medicine2010(10):438–445.
[148]张卫刚,郑晶,徐琨,徐建民.信号与线性系统西安电子科技大学出版社2005.8
[149]彭绍武.基于形状与语义建模的物体识别[D],博士学位论文,武汉,华中科技大学,2009
[150]陆承恩.基于形状的图像轮廓赋形及目标检测技术[D],博士学位论文,武汉,华中科技大学,2009.
[151]陈运文.形状识别与图像分割方法研究[D],博士学位论文,上海,复旦大学,2008
[152]徐志刚,赵祥模,宋焕生,雷涛.基于直方图估计和形状分析的沥青路面裂缝识别算法,仪器仪表学报,20101(10):2260-2266.
[153] Zhigang Xu, Xiangmo Zhao,Licheng Zhang, Ge Zhang, Asphalt Pavement Crack RecognitionAlgorithm Using Shape Analysis, ICIC Express Letter. June2011,2(6):671-678.
[2]之超.中国道路史话,道路交通管理,1998(4):20-21.
[3] http://finance.eastday.com/economic/m1/20111230/u1a6286393.html2010年公路水路交通运输行业发展统计公报.
[4]啜二勇.国外路面自动检测系统发展综述,交通标准化,2009(17):30-33.
[5]交通运输部综合规划司2010年一季度交通运输经济运行情况公报,2010年04月14日.
[6] Gang Wang,Xiu-wei xu, Liang Xiao, Algorithm Based on the finite ridgelet transform for enhancingfaint pavement cracks, Optical Engineering,47(1):017004,2008/01.
[7]交通行业标准《JTJ073.2-2001公路沥青路面养护技术规范》.
[8] J. Baker, B. Dahlstrom, K. Longenecker, T. Buu, Video image distress analysis technique forIdahotransportation department pavement management system, Transportation Research Record1117(1987)159-163.
[9] H.D. Cheng, M. Miyojim, Automatic pavement distress detection system, Journal of InformationSciences108(1998)219-240.
[10] Kelvin C P W,Highway data collection and information management [C].The3rd InternationalConference on Road&Airfield Pavement Technology,1998.
[11] Yaxiong Huang,Bugao Xu, Automatic inspection of pavement cracking distress[J], Journal ofElectronic Imaging15(1),013017(Jan–Mar2006).
[12] Salvatore Cafiso, Alessandro Di Graziano. Evaluation Of Pavement Surface Distress Using DigitalImage Collection And Analysis, Seventh International Congress on Advances in Civil Engineering,October11-13,2006,Yildiz Technical University, Istanbul, Turkey.
[13]张宏,英红.沥青路面裂缝图像识别技术研究进展,华东公路2009(4)：81-84.
[14]张娟,沙爱民,孙朝云,高怀钢.基于相位编组法的路面裂缝自动识别[J],中国公路学报,2008,21(2):39-42.
[15] Kim, Jung Yong, Development of new automated crack measurement algorithm using laser imagesof pavement surface, Ph.D. Paper, THE UNIVERSITY OF IOWA,2008.
[16] http://www.pavemetrics.com.
[17] http://www.internationalcybernetics.com.
[18] Tarek A. Monem, Amr A. Oloufa, and Hesham Mahgoub,Asphalt Crack Detection UsingThermography, InfraMation2005Proceedings,2005:1-12.
[19] Kelvin C.P. WANG, Elements of automated survey of pavements and a3D methodology, Journal ofModern Transportation,2011,19(1):51-57.
[20] Kelvin C. P. Wang, Weiguo Gong. Real-Time Automated Survey System of Pavement Cracking inParallel Environment[J].Journal of infrastructure systems. September2005:154-164.
[21] H.D. Cheng, M. Miyojim. Automatic pavement distress detection system[J]. Journal of InformationSciences,1998(108):219-240.
[22]王刚．路面病害光学无损检测技术[D]．南京：南京理工大学,2007．
[23]王建锋,激光路面三维检测专用车技术与理论研究,西安：长安大学,2010.
[24] H. D.Cheng, Chen, J. R., Glazier, C., and Hu, Y. G.1999. Novel approach to pavement distressdetection based on fuzzy set theory[J]. Journal of computing in civil engineering,13(4):270–280.
[25]高建贞,任明武,唐振民等.路面裂缝的自动检测与识别[J].计算机工程,2003,Vol.29(2):149-150.
[26]唐磊,赵春霞,王鸿南等.路面图像增强的多偏微分方程融合法[J].中国图象图形学报,2008,Vol.13(9)：1661-1666.
[27]王兴建,秦国锋,赵慧丽.基于多级去噪模型的路面裂缝检测方法[J].计算机应用,2010, Vol.30(6):1606-1612.
[28]闫茂德,伯绍波,李雪,一种自适应模糊的局部区域图像增强算法[C],第26届中国控制学会会议论文集,2007：308-311.
[29]刘玉臣,王国强,林建荣.基于模糊理论的路面裂缝图像增强算法[J].养护机械与施工技术,2006（2）：35-37.
[30] Yongxia Zuo; Guoqiang Wang; Chuncheng Zuo. Wavelet Packet Denoising for Pavement SurfaceCracks Detection,International Conference on Computational Intelligence and Security,2008. CIS'08. Vol.2:481–484.
[31] Daqi Zhang,Shiru Qu, Li He, Shuang Shi. Automatic ridgelet image enhancement algorithm forroad crack image based on fuzzy entropy and fuzzy divergence[J]. Optics and Lasers in Engineering,2009(47):1216–1225.
[32]李清泉,胡庆武.基于图像自动匀光的路面裂缝图像分析方法[J].公路交通科技,2010, Vol.27(4):1-5.
[33] K. R. Kirschke, S. A. Velinsky. Histogram-based approach for automated pavement-crack sensing[J]. Journal of Transportation Engineering,1992, Vol.118(4):700-710.
[34] Siriphan Jitprasithsiri. Development of a new digital pavement image processing algorithm forunified crack index computation[D]. A Dissertation Submitted to the Faculty of the University ofUtah,1997.
[35] Oh, H., Garrick, N. W., and Achenie, L. E. K.1997.“Segmentation algorithm using iteratedclipping for processing noisy pavement images.” Proc.,2nd Int. Conf. of Imaging Technologies:Techniques and Applications in Civil Engineering, ASCE, Reston, Va.,138–147.
[36] H. D. Cheng, Shi, X. J., and Glazier, C.2003.“Real-time image thresh-olding based on samplespace reduction and interpolation approach. J. Comput. Civ. Eng.,174,264–272.
[37]孙波成,邱延峻.路面裂缝图像处理算法研究[J].公路交通科技,2008,25（2）：64-68.
[38]李清泉,刘向龙.路面裂缝影像几何特征提取算法[J].中国科技论文在线,2007年7月, Vol2,No.7:517-522.
[39]伯绍波,闫茂德,孙国军等.沥青路面裂缝检测图像处理算法研究[J].微计算机信息(管控一体化).2007年第23卷第4-3期：280-282.
[40]冯永安,刘万军.边缘检测改进算法在路面破损检测中的应用[J].辽宁工程技术大学学报,2007年11月,第26卷增刊.176-178.
[41]李晋惠.公路路面裂缝类病害图像处理算法研究[J].计算机工程与应用2003.35.：212-213.
[42] Siwaporn Sorncharean, Suebskul Phiphobmongkol. Crack Detection on Asphalt Surface ImageUsing Enhanced Grid Cell Analysis[J].4th IEEE International Symposium on Electronic Design,Test&Application,2008：49-54.
[43]唐磊,赵春霞,王鸿南,邵文泽.基于图像三维地形模型的路面裂缝自动检测,计算机工程,vol.34(5):20-21.
[44] Peggy Subirats, Olivier Fabre, Jean Dumoulin etc. A combined wavelet-based image processingmethod for emergent crack detection on pavement surface images. European Signal ProcessingConference (EUSIPCO2004).2004,257-260.
[45] Zhou, J., Huang, P. S.,&Chiang, F.-P. Wavelet-based pavement distress detection and evaluation.Optical Engineering,2006,45:027007-1~027007-10.
[46] Kelvin C.P. Wang, Li, Q. and Gong, Weiguo "Wavelet-Based Edge Detection with à TrousAlgorithm for Pavement Distress Survey", Journal of the Transportation Research Board,2007,2024.
[47] Zhang-Xia Ma, Chun-Xia Zhao, Ying-kun Hou. Pavement Distress Detection Based onNonsubsampled Contourlet Transform[C].2008International Conference on Computer Science andSoftware Engineering,2008:28-31.
[48]王刚,贺安之,肖亮.基于高速公路裂纹局部线性特征内容的脊波变换域算法研究[J].光学学报,2006,26(3):341-346.
[49]初秀民,王荣本.基于神经网络的沥青路面破损图像识别研究[J].武汉理工大学学报(交通科学与工程版),2004,28(3):373-376.
[50]储江伟,初秀民,王荣本等.沥青路面破损图像特征提取方法研究[J].中国图象图形学报A辑,2003,8A(10):1211-1217.
[51]初秀民,王荣本,储江伟等.沥青路面破损图像分割方法研究[J].中国公路学报,2003,16(3):11-14.
[52]储江伟,初秀民,王荣本.沥青路面破损图像自动检测系统设计[J].光学技术,2003,29(3):316-319.
[53]初秀民,严新平,陈先桥.路面破损图像二值化方法研究[J].计算机工程与应用,2008,44(28):161-165.
[54] Henrique Oliveira, Paulo Lobato Correia. Supervised strategies for cracks detection in images ofroad pavement flexible surfaces.16th European Signal Processing Conference (EUSIPCO2008),Lausanne, Switzerland, August25-29,2008.
[55]王华,朱宁,王祁.公路路面分形纹理特征分析和分类[J].哈尔滨工业大学学报,2005,37(6):816-818.
[56]胡勇,赵春霞,郭志波.基于多尺度布朗运动模型的路面破损检测[J].计算机工程与应用,2008,44(3):234-235.
[57]王华,朱宁,王祁.应用计盒维数方法的路面裂缝图像分割[J].哈尔滨工业大学学报,2007,39(1):142-144.
[58]章秀华,洪汉玉,侯佳等.路面破损图像实时检测方法研究[J].电子设计工程,2009,17(6):36-40.
[59] N. Tanaka, K. Uematsu. A crack detection method in road surface images usingmorphology,Workshop on Machine Vision Applications,(1998)154-157.
[60] Yan, M.D., Bo, S.B., Xu, K., He, Y.Y.,2007. Pavement crack detection and analysis for high-gradehighway. In: Proc. Internat. Conf. on Electronic Measurement and Instruments (ICEMI’07):548–552.
[61] Liu, Fanfan; Xu, Guoai; Yang, Yixian, et al. Novel approach to pavement cracking automaticdetection based on segment extending(C].//Proceedings一2008InternationalSymposium onKnowledge Acquisition and Modeling, Inst. of Elec. and Elec. Eng.Computer Society, Wuhan2008:610-614.
[62]刘凡凡,徐国爱,肖靖等.基于连通域相关及Hough变换的公路路面裂缝提取[J].北京邮电大学学报,2009年4月,第32卷,第2期.
[63] Sylvie Chambona,Peggy Subiratsb, Jean Dumoulina. Introduction of a wavelet transform based on2D matched filter in a Markov Random Field for fine structure extraction: Application on road crackdetection[J]. Proc. of SPIE-IS&T Electronic Imaging,2009, SPIE Vol.7251:72510A.1-72510A.12.
[64]张洪光,王祁,魏玮.基于人工种群的路面裂纹检测[J].南京理工大学学报,2005,29(4):128-132.
[65] Alekseychuk, O.2006. Detection of crack-like indications in digital radiography by globaloptimisation of a probabilistic estimation function.Ph.D. thesis, BAM-Dissertationsreihe, Band18,Berlin.
[66]李清泉,邹勤,毛庆洲.基于最小代价路径搜索的路面裂缝检测[J].中国公路学报,2010,23(6):28-33.
[67] Vivek Kaul, Yichang Tsai, and Anthony Yezzi. Detection of Curves with Unknown Endpoints usingMinimal Path Techniques. Proceedings of the British Machine Vision Conference, pages62.1-62.12.BMVA Press, September2010.
[68]李刚.基于灰色系统理论的路面图像裂缝检测算法研究[D].博士学位论文,武汉,武汉理工大学,2010.
[69] Chua, K. M.,&Xu, L.(1994). Simple procedure for identifying pavement distresses from videoimages[J]. Journal of Transportation Engineering,120(3),412–431.
[70] Acosta, J. A., Figueroa, J. L.,&Mullen, R. L.(1995). Algorithm for pavement distress classificationby video image analysis. Transportation Research Record,1505,27–38.
[71] Miyojim M, H. D. Cheng. Novel system for automatic pavement distress detection[J]. ASCE,1998,12(3):145-152.
[72] H. D. Cheng, Jiang, X., Li, J.,&Glazier, C.(1999). Automated real time pavement distress analysis.Transportation Research Record,1655,55–64.
[73] Kelvin C.P. Wang, Watkins,&Kuchikulla.(2002). Digital distress survey of airport pavementsurface. Federal Aviation Administration Airport Technology Transfer Conference.
[74] Lee, David Lee.(2003). A robust position invariant artificial neural network for digital pavementcrack analysis. Technical Report TRB2003-000996.
[75] Zhou, J., Huang, P. S.,&Chiang, F.-P. Wavelet-based pavement distress detection and evaluation.Optical Engineering,2006,45：027007-1~027007-10.
[76] Lee,B.J and H.Lee,A Robust Position Invariant Neural Network for Digital Pavement CrackAnalysis.Computer-Aided Civil and Infrastructure Engineering,2004,19:105-118.
[77]丁爱玲,焦李成.基于支撑矢量机的路面破损识别[J].长安大学学报(自然科学版),2007,27(2):34-37.
[78] Fereidoon Moghadas Nejad, Hamzeh Zakeri. An optimum feature extraction method based onWavelet–Radon Transform and Dynamic Neural Network for pavement distress classification[J].Expert Systems with Applications38(2011)9442–9460.
[79]肖旺新,严新平,张雪.基于混合密度因子的路面破损自动识别研究[J].交通运输工程与信息学报,2005,3(2):19-26.
[80]肖旺新,张雪,黄卫.基于破损密度因子的路面破损识别新方法[J].交通运输工程与信息学报,2004,2(2):82-89.
[81]李清泉,刘向龙.路面影像破损加权评定方法[J].中国公路学报,2009,22(4):44-49.
[82]Xiao Mei,automatic detection of pavement suface crack depth on florida roadways, Phd paper, SouthFlorida,Department of Civil and Environmental Engineering College of The University of SouthFlorida,2001.
[83]王泽民.混凝土路面破损评价与维修对策研究[D],博士学位论文,西安,长安大学,2007
[84]王刚.路面病害光学无损检测技术[D],博士学位论文,南京理工大学,2007.
[85]唐磊.基于图像分析的路面病害自动检测[D],博士学位论文,南京,南京理工大学,2007
[86]肖旺新.路面破损图像自动识别关键技术研究[D],博士学位论文,南京,东南大学,2005
[87] http://baike.baidu.com/view/297217.htm百度百科
[88]蔡涛.面向目标识别的图像特征融合提取技术研究[D],博士学位论文,长沙,国防科学技术大学,2000,10:2-3
[89]刘帅师,田彦涛,万川.基于Gabor多方向特征融合与分块直方图的人脸表情识别方法,自动化学报,2011,37（12）：1455-1463.
[90]潘瑜,郑钰辉,孙权森,等.基于PCA和总变差模型的图像融合框架,计算机辅助设计与图形学学报,2011,23（7）：1200-1210.
[91]孙权森,曾生根,王平安,等.典型相关分析的理论及其在特征融合中的应用,计算机学报,2005,28（9）：1524-1533.
[92]潘瑜,孙权森,夏德深.基于PCA分解到图像融合框架,计算机工程,2011,37（13）：210-212
[93]李建科,赵保军,张辉,等.DCT和LBP特征融合的人脸识别,北京理工大学学报,2010,30（11）：1355-1358.
[94]张俊虎,郝晓剑,刑昊,人脸图像识别中的PCA算法实现,模式识别,2010,26（7-1）：186-187.
[95]韩智,刘昌平.基于多特征融合的指纹识别方法,计算机科学,2010,37（7）：255-259.
[96]谭利,李彬,田联房,等.基于多特征融合跟踪的微小肺结节识别算法,生物医学工程学杂志,2011,28（3）：437-441.
[97]王坤峰,李镇江,汤淑明.基于多特征融合的视频交通数据采集方法,自动化学报,2011,37（3）：322-330.
[98]顾鑫,王海涛,汪凌峰,等.基于不确定性度量的多特征融合跟踪,自动化学报,2011,37（5）：550-559.
[99]王莉莉,郝爱民,何兵,等.基于D-S证据理论的城市航拍道路提取方法,软件学报,2005,16（9）：1534-1541.
[100]张雷,肖梅,马建,等,基于信息融合的车道检测算法,郑州大学学报（工学版）,2010,31（1）：98-102.
[101]朱国康,王运锋,基于多特征融合的道路交通标志检测,信号处理,2011,27（10）：1616-1620.
[102]孙显,王宏琦,张道兵,等.基于多特征融合的城市遥感图像自动解译方法,光子学报,2010,39（1）：178-183.
[103]林国余,陈旭,张为公.基于多信息融合优化的鲁棒性车道检测算法,东南大学学报（自然科学版）,2010,40（4）：771-777.
[104]李远征,卢朝阳,高全学,等.基于多特征融合的均值迁移粒子滤波跟踪算法,电子与信息学报,2010,82（2）：411-416.
[105]周静,黄心汉,彭刚.基于多特征融合的飞机目标识别,华中科技大学学报（自然科学版）,2009,37（1）：38-41.
[106]宁慧君,李映,胡杰.基于多尺度特征融合的SAR图像分割,计算机工程与应用,2011,47（14）：196-199.
[107]李正周,马齐佑,郑微,等.基于多特征融合的微弱红外运动目标跟踪方法,强激光与粒子束,2011,23（1）：54-58.
[108]陈皓,马彩文,陈岳承,等.基于多特征融合的复杂背景下弱小多目标检测和跟踪算法,光子学报,2009,38(9):2444-2448.
[109]王莹,李文辉,基于多特征融合的高精度视频火焰检测算法,吉林大学学报（工学版）,2010,40（3）：769-775.
[110]李先锋,朱伟兴,花小朋,等.基于D-S证据理论的决策级多特征融合苹果分级方法,农业机械学报,2011,42（6）：188-192.
[111]李先锋,朱伟兴,孔令东,等.基于SVM和D-S证据理论的多特征融合杂草识别方法,农业机械学报,2011,42（11）：164-168.
[112]范振东,吝宁.基于并行特征融合的金融汉字特征提取方法,计算机应用,2007,27（6）：127-129.
[113] WaldL.,SomeTerms of Referenee in Data Fusion,IEEE Trans.on GRS,1999,37(3):l-3.
[114] David L.Hall, James Llinas,An Introduction to Multisensor DataFusion,Proc.IEEE,1997,85(l):6-28.
[115]赵宗贵等,多传感器数据融合,北京:电子部27所,1995.
[116]韩崇昭,朱洪艳,段战胜等.多源信息融合(第二版)[M].北京:清华大学出版社,2010,9.
[117] Yang J.,Yang J.Y.,Zhang D.,Lu J.F.Feature fusion Parallel strategy vs.serial strategy.PatternRecognition[J],2003,Vol.36(6):1369-1381.
[118] Xuguang Zhang,at el..Integrated Intensity,Orientation Code and Spatial Information for RobustTracking[C].ICIEA2007:1-4.
[119] Wang Dawei,Ge Wei,Wang Yan jie.Using BBPSO for Feature Select in Feature-Level FusionTarget Recognition[C].4th IEEE conference on industrial electronics and applications, Xi’an,China,May.2009:1-4.
[120] Fabien Scalzo, George Bebis, Mircea Nicolescu, Leandro Loss. Evolutionary Learning of FeatureFusion Hierarchies[C].IEEE ICPR2008, Dec.2008:1-4.
[121]覃征,鲍福民,李爱国.数字图像融合[M].西安:西安交通大学出版社,2004.
[122]王大伟.基于特征级图像融合的目标识别技术研究[D].中国科学院研究生院,博士学位论文,2010.
[123]王大伟,陈浩,王延杰.核典型相关分析的融合人脸识别算法[J].激光与红外,2009(11):1241-1245.
[124] Pavement Distress Identification Manual for the NPS Road Inventory Program Cycle4,2006-2009.
[125]毛成.沥青路面裂纹形成机理及扩展行为研究[D],博士学位论文,成都,西南交通大学,2004.
[126]吴国雄.水泥混凝土路面开裂机理及破坏过程研究[D],博士学位论文,成都,西南交通大学,2003.
[127](美)冈萨雷斯(Gonzalez,R.C)等著.数字图像处理(MATLAB版)[M].北京:电子工业出版社,2005,9.
[128]程宏煌,戴卫恒.图像分割方法综述[J].电信快报,2000(10)：39-41.
[129] Kittler J,Illingworth J. Minimum error threshol-ding[J]. Pattern Recognition,1986,19(1)：41-47.
[130] Chow C K,Kaneko T. Automatic boundary detec-tion of the left ventricle fromcineangiograms[J]. Computers and Biomedical Research,1972(5)：338-410.
[131]杨喜宁,段建民,高德芝,郑榜贵.基于改进Hough变换的车道线检测技术.计算机测量与控.2010,18(2)：292-298.
[132] D.L.Donoho,X.Huo. Beamlets pyramids：A new form of multi-resolution analysis,suited forextracting lines,curves,and objects from very noise image data[C].Proceedings ofSPIE,2000,4119(1)：434-444
[133] D.L.Donoho,X.Huo. Beamlets and Multi-scale Image Analysis[R],Multi-scale andMulti-resolution Methods[C].Springer Lecture Notes in Computational Science andEngineering,2002,20：149-196
[134] D.L.Donoho,O.Levi,J.L.Starch,et al. Multi-scale geometric analysis for3.d catalogues,Astronomical Data AnalysisⅡ[J].Proceeding of SPIE,2002,4847：101-111.
[135]史勤峰.Beamlet及其在SAR图像边缘检测中的应用[D].硕士学位论文,西安市：西北工业大学,2006.
[136]屈庆春.基于beamlet的多尺度图像分析及其在CT图像处理中的应用[D].硕士学位论文,济南市：山东大学,2005.
[137]邓双城,蒋力培,焦向东等.基于束波变换的焊缝图像处理技术[J].焊接学报,2009,30(1)：68-72.
[138]焦李成,侯彪,王爽等.图像多尺度几何分析理论与应用-后小波分析理论与应用[M].西安：西安电子科技大学出版社,2008：210-238.
[139]侯阿临,王明明,史东承等.基于离散Beamlet变换的线特征检测算法[J].计算机安全,2009,17(3)：8-11.
[140] Qin F S,Yan N Z. Adaptive Linear feature detection based on beamlet[J].IEEE,2009,3981-3984.
[141]潘广林,梁彦,金术玲等.基于子束变换的自适应线性特征检测[J].火力与指挥控制,2008,33(10)：20-23.
[142]沈丽琴,胡栋梁,戚飞虎.基于知识的线状目标的综合理解,自动化学报,1997.23C61：839-841.
[143] Frank R J, Grabowski T J, Damasi o H. Voxel vise. Percentage tissue segmentation of human brainMRI[C]. In:25th AnnualMeeting, Society for Neuroscience, Society of Neuroscience[C],Washington,1995:694.
[144]邹丹平,冯涛,李咸伟,刘其真,基于EM的直方图逼近及其应用[J],中国图象图形学报,2005,10(11):1458-1461.
[145]李亚平,杨华,陈霞,基于EM和BIC的直方图拟合方法应用于遥感变化检测阈值确定[J],遥感学报,2008,12(1):85-91.
[146]阎冬梅.基于特征融合的遥感影像典型现状目标提取技术研究[D],博士学位论文,北京,中国科学院研究生院,2003.
[147] Bob Zhang, LinZhangb, LeiZhangb, FakhriKarraya Retinal vessel extraction by matched filter withfirst-order derivative of Gaussian, Computers in Biology and Medicine2010(10):438–445.
[148]张卫刚,郑晶,徐琨,徐建民.信号与线性系统西安电子科技大学出版社2005.8
[149]彭绍武.基于形状与语义建模的物体识别[D],博士学位论文,武汉,华中科技大学,2009
[150]陆承恩.基于形状的图像轮廓赋形及目标检测技术[D],博士学位论文,武汉,华中科技大学,2009.
[151]陈运文.形状识别与图像分割方法研究[D],博士学位论文,上海,复旦大学,2008
[152]徐志刚,赵祥模,宋焕生,雷涛.基于直方图估计和形状分析的沥青路面裂缝识别算法,仪器仪表学报,20101(10):2260-2266.
[153] Zhigang Xu, Xiangmo Zhao,Licheng Zhang, Ge Zhang, Asphalt Pavement Crack RecognitionAlgorithm Using Shape Analysis, ICIC Express Letter. June2011,2(6):671-678.