Research on an Interactive Intelligent Defect Detection System for Pipeline Networks
Defect identification in urban drainage pipelines based on closed-circuit television(CCTV)data often requires laborious manual efforts or deep learning-based methods.How-ever,manual identification is characterized by an enormous workload and is time-consuming,while traditional deep learning approaches suffer from complex defect recognition with low accuracy and the need for a significant amount of labeled data,which are inadequate for anom-aly classification and precise defect localization in drainage pipelines.To address these chal-lenges,this paper proposes an interactive deep learning-based method for intelligent identifi-cation of defects in drainage pipelines,enabling continuous improvement of recognition per-formance even with limited training samples.The approach leverages appropriate human in-terventions to correct identification results and automatically incorporates them into the sam-ple repository,accumulating and enriching a diverse set of defect samples.Through continu-ous learning from updated samples,the accuracy of defect identification can be rapidly en-hanced.The effectiveness of the proposed method is validated through an experiment with only 1 627 defect samples from video data of the drainage pipeline from Wuhan,China,and its identification accuracy of overall defects could reach 68％,demonstrating promising ap-plication prospects.
deep learninginteractiveintelligentdrainage pipelinespipeline defects
NETL
NSTL
万方数据
维普
