水泥混凝土路面错台及裂缝检测算法研究

发布时间：2018-07-08 12:41

本文选题：错台 + 裂缝　；参考：《长安大学》2014年硕士论文

【摘要】：随着水泥混凝土路面的广泛应用，由于其在设计、施工、养护及运营等方面仍存在着一系列技术问题，使得该种路面早期病害屡见不鲜，因此研究有效的水泥混凝土路面病害检测算法来高效、准确的获取路面病害特征数据有着重要的价值和现实意义。分析已有的国内外错台、裂缝检测算法及较新的研究成果存在的缺陷，，确立了本文的具体研究内容以及需要重点解决的关键技术。本文针对激光三维成像系统的特点以及水泥混凝土路面的特性，对基于三维检测数据的错台、裂缝检测算法进行了深入研究，完成了错台量的计算和裂缝的识别与提取。首先提出双相标准差和基于组合结构元素的级联形态学相结合的图像预处理滤波算法；然后根据错台的形状特性，确定基于拐点的平均错台量和最大错台量计算方法。本文还实现将裂缝三维图像数据归一化为二维灰度图像数据，并采用目标减背景法进行裂缝图像去噪处理；然后研究基于最大类间方差局部阈值分割法的裂缝目标识别，并实现裂缝后处理与细化后的提取。针对本文提出的错台算法，设计进行了室内室外实验，可知错台三维检测结果与直尺测量结果基本吻合，两者的绝对误差在3mm以内，91%的错台在2mm内。并且本文错台算法还能够给出最大错台量的值，为研究水泥混凝土路面错台与水泥混凝土路面平整度、舒适性等的关系，以及错台分级、错台处置提供了辅助信息支持。同时裂缝的识别与提取方法研究为后续裂缝特征参数计算打下了基础。
[Abstract]:With the wide application of cement concrete pavement, there are still a series of technical problems in the design, construction, maintenance and operation of the pavement. Therefore, it is of great value and practical significance to study the effective algorithm of cement concrete pavement disease detection to obtain the pavement disease characteristic data efficiently and accurately. This paper analyzes the defects of the existing fault stations, crack detection algorithms and new research results at home and abroad, and establishes the specific research contents and key technologies to be solved in this paper. Aiming at the characteristics of laser 3D imaging system and the characteristics of cement concrete pavement, this paper makes a deep research on the algorithm of fault platform and crack detection based on 3D detection data, and completes the calculation of the number of staggered platforms and the identification and extraction of cracks. An image preprocessing filtering algorithm combining biphase standard deviation and cascaded morphology based on combined structural elements is proposed, and then the calculation method of the average and maximum staggered station quantity based on inflection point is determined according to the shape characteristics of the staggered station. In this paper, the fracture 3D image data is normalized to two-dimensional gray image data, and the target subtraction background method is used to de-noising the crack image, and then the crack target recognition based on the maximum inter-class variance local threshold segmentation method is studied. The post-treatment of cracks and the extraction of thinning are realized. According to the algorithm proposed in this paper, indoor and outdoor experiments are carried out. It can be seen that the results of three dimensional measurement are basically consistent with the measured results, and the absolute error between them is 91% within 3mm and 91% in 2mm. Moreover, this algorithm can also give the maximum amount of stagger, which provides auxiliary information support for the study of the relationship between the misalignment of cement concrete pavement and the smoothness and comfort of cement concrete pavement, as well as the classification and disposal of staggered platforms. At the same time, the research on the identification and extraction of fractures lays a foundation for the calculation of fracture characteristic parameters.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.41;U416.216;U418.6

【参考文献】