基于卷积神经网络的嵌入式排水管道缺陷检测系统

Embedded Drainage Pipeline Defect Detection System Basedon Convolutional Neural Network

随着城市规模扩大、人口日益增多,城市排水系统压力越来越大,由于初期排水管道设计得不合理,技术落后等原因,出现了破裂、错口、异物插入等缺陷,严重影响社会和财产安全,更甚者影响社会的发展。CCTV视频探损法是近年来最主流的检测方法,但传统的采集数据,人工判读的效率低,主观影响大,已严重影响工程进度;本文提出一种基于卷积神经网络的嵌入式排水管道缺陷检测系统,具有实时检测,功能齐备等特点,有效提高了排水管道缺陷检测效率、提升了成果的客观性以及降低了项目人工成本。

With the expansion of the scale of the city and the increasing population,the pressure on the urban drainage system is increasing.Due to the unreasonable design of the initial drainage pipes and the backward technology,defects such as cracks,misalignments,and insertion of foreign objects have appeared,which have seriously affected the society and property.Security,even worse,affects the development of society.The CCTV video detection method is the most mainstream detection method in recent years,but the traditional collection of data,the efficiency of manual interpretation is low,and the subjective impact has seriously affected the project progress;this paper proposes an embedded drainage pipeline defect based on convolutional neural network the detection system has the characteristics of real-time detection and complete functions,which effectively improves the efficiency of drainage pipeline defect detection,improves the objectivity of the results,and reduces the labor cost of the project.