No-Dig Inspection Technologies for Underground Pipelines

Inspection through diagnosis of pipe defects is the essential work of underground pipeline maintenance and rehabilitation. To avoid excavation, a No-dig inspection is commonly executed by manually interpretation on closed circuit television (CCTV) images, but this approach seems inefficient because of human's fatigue, subjectivity, and time-consumption. To solve this problem, several automated diagnosis systems have been developed to attempt to assist general staffs in diagnosin g pipe defects. In addition, many researchers pointed out that CCTV image quality would also influence accuracies of manually interpretation or automated diagnosis systems. Therefore, a procedure of inspection through pipe defect diagnosis is proposed and applied to sewer inspection. This paper briefly introduces a series of inspection technologies involving image quality assessment, morphological feature extraction of pipe defects, and radial basis network (RBN)-based automated diagnosis system. The experimental results show that the proposed image quality index is useful in assessing CCTV image quality and the overall diagnosis accuracies of above 90% can be obtained by the automated diagnosis system.

並列關鍵字

No-dig inspection ； diagnosis of sewer pipe defects ； CCTV image quality assessment ； radial basis network (RBN)

國際替代計量

全文下載

主題瀏覽