土木工程结构中远距离非接触式表面缺陷及裂缝检测系统研究

发布时间：2018-03-05 13:29

本文选题：非接触　切入点：裂缝　出处：《中冶集团建筑研究总院》2013年硕士论文　论文类型：学位论文

【摘要】：在建构筑物工程建设和运营过程中,由于设计、施工以及使用等方面的原因,混凝土结构往往会产生裂缝,并影响其使用性及耐久性。而且,当裂缝宽度扩展到一定程度时,还极有可能危及主体结构的安全。虽然大量的科研成果和实践已经证明,混凝土结构出现裂缝是难以避免的,是混凝土固有的材料特性之一,是人们可以接受的,但是问题的关键是要将其有害程度严格控制在允许的范围之内。裂缝的开展状况,往往是诊断、分析、评判已有建筑物病态的重要条件和依据之一。根据其位置和形态,可以初步判断建筑结构上出现裂缝的原因,评判裂缝的危害,从而采取相应的解决措施。因此,及时准确掌握混凝土裂缝的形态、走向、宽度、长度和深度等信息,对于正确判断混凝土构件的健康状况,提出科学合理的处理方案起着至关重要的作用。目前,裂缝宽度检测以传统的接触式检测方法为主,这种方法准确度较高,但效率较低,影响检测进程和结果,人员安全隐患大,经济成本和人力成本较高。当前工程检测实际应用中急需一种远距离的非接触式裂缝测量设备,使得检测成本降低,提高工作效率,并在此基础上,实现远程遥控、实时的、大批量的高精度测量,以满足现代社会对建筑测量智能化的要求。因此,本科研课题的设立对于建筑物智能化检测、鉴定、评估工作具有重大意义。随着数字化光学摄影技术的发展,涉及到光学摄像的硬件技术如光学镜头、电荷耦合传感器、相关电子元器件等产品性能指标大幅度提高,且成本逐渐走低,为实现光学测量技术的应用创造了条件。本课题正是利用基于光测法的数字图像处理技术,对混凝土表面裂缝宽度等特征进行测量和分析。具体技术路线为：1、利用图像采集系统对裂缝形态进行拍摄；2、将采集到的图像进行滤波、锐化、对比度增强等处理；3、对裂缝的图像边缘进行处理,以此来锐化裂缝边缘,使其更加清晰可辨；4、将裂缝边缘作亚像素处理,确保图像中的裂缝在放大处理后依然保持清晰可辨的轮廓边缘,以利于计算时拾取边缘位置。本课题打破常规裂缝图像处理方法仅能计算水平和竖直方向裂缝宽度的局限,提出了一种全新的能计算任意走向和在一定倾角、俯仰角及在圆弧形特殊平面内的裂缝宽度计算方法,并详细的对其进行了多种误差修正,经多组实验比较、分析和验证,该方法具有较高的精度,完全能够满足建构筑物表面裂缝检测的需要。最后,基于Java开发环境,采用面向对象的程序设计方法,编写了裂缝宽度检测软件,实现了裂缝宽度的直接测量,使裂缝检测更加数字化,解决了传统人工检测裂缝的效率低、不确定性、主观性等方面的弊端,发挥了数字图像处理技术的优势。采用MySQL数据库,实现了裂缝信息的智能化管理。
[Abstract]:In the process of construction and operation of building structures, concrete structures often produce cracks and affect their serviceability and durability due to the reasons of design, construction and use. Moreover, when the width of cracks extends to a certain extent, Although a great deal of scientific research and practice has proved that cracks in concrete structures are unavoidable, are one of the inherent material properties of concrete, and are acceptable to people. But the crux of the problem is to strictly control the extent of damage within the allowable range. The condition of crack development is often one of the important conditions and bases for diagnosing, analyzing, and judging the morbid state of existing buildings, according to its location and shape. The cause of cracks in building structure can be preliminarily judged, the harm of cracks can be judged, and corresponding measures can be taken. Therefore, the information of form, direction, width, length and depth of concrete cracks can be grasped in time and accurately. It is very important to put forward a scientific and reasonable treatment scheme for judging the health condition of concrete members correctly. At present, the traditional contact detection method is the main method of crack width detection, which has high accuracy but low efficiency. In order to reduce the detection cost and improve the working efficiency, it is urgent to use a kind of remote non-contact crack measuring equipment in the practical application of engineering testing, which affects the testing process and results, and has a high economic cost and high labor cost. And on this basis, the remote control, real-time, high-precision measurement in large quantities is realized to meet the requirement of intelligent building survey in modern society. Therefore, the establishment of this research topic is useful for the intelligent detection and appraisal of buildings. The evaluation is of great significance. With the development of digital optical photography technology, the performance indexes of the hardware related to optical camera, such as optical lens, charge coupling sensor, related electronic components, etc., have been greatly improved. And the cost is decreasing gradually, which creates the conditions for the application of optical measurement technology. The concrete surface crack width and other characteristics are measured and analyzed. The concrete technical route is: 1. Using the image acquisition system to shoot the crack shape, the collected image is filtered and sharpened. Contrast enhancement and other processing, such as processing the image edge of the crack, so as to sharpen the edge of the crack, make it more clearly discernible and make the edge of the crack subpixel processing, To ensure that the cracks in the image remain clear and discernible contour edges after magnification, so as to pick up the edge position in the calculation. This subject breaks the limitation that the conventional crack image processing method can only calculate the horizontal and vertical crack width. A new method for calculating arbitrary strike and crack width in a certain inclination angle, pitch angle and in a special plane of circular arc is proposed, and various error corrections are carried out in detail, which are compared, analyzed and verified by many experiments. This method has high precision and can completely meet the need of crack detection on the surface of building structure. Finally, based on the Java development environment, an object-oriented programming method is used to compile the software for crack width detection. It realizes the direct measurement of crack width, makes the crack detection more digital, solves the disadvantages of the traditional manual detection of cracks, such as low efficiency, uncertainty, subjectivity and so on, and gives full play to the advantages of digital image processing technology. MySQL database is adopted. The intelligent management of crack information is realized.
【学位授予单位】：中冶集团建筑研究总院
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU317;TP391.41

【参考文献】