Research on Defect Identification System of Drainage Pipe Network Based on Mask R-CNN
In the CCTV-based drainage pipe network inspection project,the traditional identification method is to identify and record the internal conditions of the pipeline through manual video recording,which is not only time-consuming and labor-intensive,but also heavily reliant on the experience of the staff,making it difficult to meet the requirements of quality and speed for large-scale projects.To address the problem of low efficiency in manual identification,a deep learning method for identifying defects in drainage pipe networks based on Mask R-CNN was proposed,and a defect identification platform was built accordingly.Through labeling and identification analysis of the defect image library and videos of the drainage pipe network in a trunk road area of a district in Chongqing,the results showed that the classification prediction accuracy of this method reached 88.16%,which was better than similar algorithms,indicating that it has great practical value and application prospects in the field of defect identification for drainage pipe networks.
万方数据
