摘要
干滩长度是整个尾矿坝安全稳定性的一个重要参数。为了实时准确地测得干滩长度值,提出了一种高效、智能、准确的在线监测新方法——基于Mask R-CNN实例分割算法的干滩长度测量方法。此方法共分为4部分:(1)在尾矿坝两岸安装监控摄像头;(2)基于Mask R-CNN算法,训练出识别水线并输出水线坐标的网络模型;(3)将水线轮廓坐标与实际干滩长度进行回归分析,拟合出测量算法关系式;(4)将水线坐标输入上述关系式,即可通过视频画面实时测得干滩长度。研究结果表明,此模型能够准确地进行干滩长度的测量,且适用于光照不足、图像模糊、雨雪天气等情况。
The length of dry beach is an important parameter for the safety and stability of the whole tailings dam. In order to accurately measure the length of dry beach in real time,an efficient,intelligent and accurate online monitoring method based on Mask R-CNN algorithm( instance segmentation algorithm) was proposed. The method is divided into four parts:( 1) Installing monitoring cameras on both sides of the tailings dam;( 2) Training a network model to recognize the water line and output the coordinates of the water line based on the Mask R-CNN algorithm;( 3) Regression analysis was carried out between the contour coordinates of water line and the actual length of dry beach,and the relationship formula of measurement algorithm was fitted;( 4) The length of dry beach can be measured in real time by inputting the coordinates of water line into the above formula. The results show that the model can accurately measure the length of dry beach,and is suitable for the conditions of insufficient light,blurred images,rain and snow.
作者
杨俊
孙叶青
申屠南瑛
李青
YANG Jun;SUN Ye-qing;SHENTU Nan-ying;LI Qing(National and Local Joint Engineering Laboratory of Disaster Monitoring Technology and Instruments,China Jiliang University,Hangzhou,Zhejiang 310018,China)
出处
《计量学报》
CSCD
北大核心
2020年第12期1468-1474,共7页
Acta Metrologica Sinica
基金
国家重点研发计划(2017YFC0804604)
浙江省重点研发计划(2018C03040)
国家自然科学基金青年项目(61701467)
