摘要
针对目前检测裂缝宽度时通常采用的接触式人工直接测量方法费时费力、且难以准确测量最大裂缝宽度的问题,基于图像分析技术的应用,提出采用在混凝土表面粘贴纯色标定块来标定、修正原始图像,并对其进行对比增强、平滑等一系列处理后直接在图像上确定裂缝宽度的无接触式测量方法.基于所提方法,分别以正拍和斜拍图像对一开裂混凝土梁的表面裂缝进行了检测,结果表明:基于正拍和斜拍图像确定的裂缝宽度的识别精度分别为93.4%和90.9%,表明该方法能有效检测混凝土结构表面的裂缝宽度.
The crack width on structural concrete surface is one of the most important parameters in the detection and assessment of concrete structures. At present, the contact-based method is commonly used by inspectors to detect the maximum width of crack. However, this method is a great consumption of time and energy and can not measure the maximum width of crack accurately. Based on the application of image analysis technology, a new non-contact method was proposed in this paper, which used the calibration block on the concrete surface and revised the original deformed images, thus enhancing the contrast of the revised image and smoothing them, etc. Finally, the width of crack in the image can be calculated. The method has been verified by the laboratory experiment of a cracked concrete girder using vertical and oblique photographic techniques to detect the cracks and the accuracy of 93.4% and 90.9% have been obtained, which have shown that this new method can effectively detect the width of crack.
出处
《湖南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2012年第1期7-12,共6页
Journal of Hunan University:Natural Sciences
基金
湖南省交通厅科技计划项目
关键词
混凝土
结构检测
裂缝宽度
图像分析
图像修正
concrete structure
structural detection
crack width
image analysis
image correction
