水利学报2025，Vol.56Issue(4)：445-454,465,11.DOI:10.13243/j.cnki.slxb.20240185
碾压混凝土坝仓面碾压及摊铺状态视觉监控方法
Visual monitoring method forthe rolling and paving status of the roller compacted concrete dam
摘要
Abstract
The control of construction time for roller compacted concrete(RCC)dam significantly impacts the quality of pouring.However,the environment of RCC dams is complex,and the traditional method of assigning dedicated personnel to manually record at fixed times and locations faces challenges such as high personnel costs,low efficiency,and difficulties in safety management.To address the above issues,this article proposes a visual-based intelligent monitoring and analysis method for the rolling and paving status of RCC dams.Using a video sur-veillance system arranged on the construction site to obtain image data of construction,the QueryInst deep learn-ing algorithm automatically identifies the construction status of various areas on the dam surface.Considering the relationship of the image data,statistical analysis is conducted on the construction stage and construction time of each stage on site.This method has been applied to the monitoring process of the construction status of a RCC dam in Southwest China.The results show that the proposed method can accurately and effectively identify the con-struction status of various areas on the warehouse surface and calculate the duration of each construction stage,providing rich data support and reference basis for better management level at construction sites.关键词
碾压混凝土坝/水利工程施工/视频监控/深度学习/计算机视觉/碾压施工Key words
roller compacted concrete dams/construction of hydraulic engineering/video monitoring/deep learning/computer vision/rolling construction分类
水利科学引用本文复制引用
宋本扬,崔博,张天伟,吴斌平,任炳昱,佟大威..碾压混凝土坝仓面碾压及摊铺状态视觉监控方法[J].水利学报,2025,56(4):445-454,465,11.基金项目
国家自然科学基金项目(U23B20148) （U23B20148）
华能澜沧江水电股份有限公司基金项目(HNKJ20-H21TB) （HNKJ20-H21TB）
