中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

Journal of Chongqing Jiaotong University(Natural Science) ›› 2019, Vol. 38 ›› Issue (03): 21-26.DOI: 10.3969/j.issn.1674-0696.2019.03.04

• Bridge & Tunnel Engineering • Previous Articles     Next Articles

Apparent Tunnel Diseases Identification Based on Digital Images

HE Guohua1, LIU Xingen2, 3, CHEN Yingying2, 3, YANG Jun1, ZHONG Bei2, 3   

  1. (1. Guizhou Expressway Group Co., Ltd., Guiyang 550004, Guizhou, P. R. China; 2. Shanghai Tongyan Civil Engineering Technology Co., Ltd., Shanghai 200092, P. R. China; 3. Shanghai Engineering Technology Research Centre of Underground Infrastructure Safety Inspection and Maintenance Equipment, Shanghai 200092, P. R. China)
  • Received:2018-03-01 Revised:2018-12-07 Online:2019-03-20 Published:2019-03-20

基于数字图像的隧道表观病害识别方法研究

何国华1, 刘新根2,3,陈莹莹2,3,杨俊1,钟北2,3   

  1. (1. 贵州高速公路集团有限公司,贵州 贵阳 550004; 2. 上海同岩土木工程科技股份有限公司,上海 200092; 3. 上海地下基础设施安全检测与养护装备工程技术研究中心,上海 200092)
  • 作者简介:何国华(1976—),男,广东翁源人,高级工程师,主要从事隧道养护方面的研究。E-mail:350634098@qq.com。 通信作者:刘新根(1981—),男,江西新余人,高级工程师,主要从事公路隧道快速检测装备及数值计算方面的研究。E- mail:xuezhongfei2000@163.com。
  • 基金资助:
    贵州省科技计划项目(黔科合支撑[2016]2318);上海人才发展资金资助计划(2017055);上海市2018年技术标准专项项目 (18DZ2202300)

Abstract: The principle of image acquisition of fast detection vehicle for highway tunnel structure was expounded. Aiming at the apparent image of tunnel structure, the image characteristics of tunnel cracks and seepage diseases were systematically analyzed. Based on the characteristics of fracture image, cracks can be accurately identified by combining CTA measurement algorithm with edge detection. According to the characteristics of seepage image, the improved CTA algorithm and morphological processing method can be used to identify and locate the seepage of tunnel structure. The experiment results show that: apparent diseases can be identified effectively by using CTA measure and its improved algorithm, which can effectively reduce the influence of cable interference and illumination change.

Key words: highway tunnel, inspection, image identification, crack, seepage water

摘要: 阐述了公路隧道结构快速检测车图像采集的原理,以隧道结构表观图像为研究目标,系统分析了隧道裂缝和渗漏水病害的图像特征; 基于裂缝图像特征,通过CTA测度算法和边缘检测结合可准确地识别裂缝;根据渗漏水图像特征,采用改进的CTA算法,并结合形态学处理方 法,可实现隧道结构渗漏水的识别与定位。研究表明:采用CTA测度及其改进算法可较好地实现表观病害识别,可有效地减弱线缆干扰、光 照变化带来的影响。

关键词: 公路隧道, 检测, 图像识别, 裂缝, 渗漏水

CLC Number: