The Chinese Journal of Nonferrous Metals
JOURNAL OF RAILWAY SCIENCE AND ENGINEERING
|第13卷 第1期 总第70期 2016年1月|
摘 要: 准确检测桥面裂缝，对于提高桥梁养护效率和保障车辆通行安全具有重要意义，但是，传统的检测方法往往效率低下而且人力成本过高。针对这种现状，提出一种基于双目视觉的桥面裂缝检测方法，通过CCD工业相机采集裂缝图像，再对图像进行数字化处理，最后由双目视觉系统计算裂缝尺寸，从而达到检测目的。工程实践表明，该方法检测效率要显著高于传统方法，且检测精度亦能满足相关技术标准的规定。
（Changsha Planning & Design Institute Co. ,Ltd, Changsha 410007, China）
Abstract:Detecting the bridge deck crack accurately is of great significance for improving bridge maintenance efficiency and ensuring traffic safety. However, the traditional detection methods are often inefficient and labor-consuming. In this regard, this paper proposed a bridge crack detection method based on binocular vision, to capture the crack images through CCD industrial camera, which was digitized with a computer to calculate the size of crack with binocular vision theory. The engineering practice shows that the proposed method was far more efficient than the traditional one, and the detection accuracy can also meet the relevant technical standards.
Key words: bridge crack detection; binocular vision; image processing