级配分析中堆叠颗粒完整形态预测方法  

Prediction Method of Complete Form of Overlapping Particle in Granular Gradation Analysis

在线阅读下载全文

作  者:蔡可天 王建[1] 夏万求 彭泽豹 CAI Ketian;WANG Jian;XIA Wanqiu;PENG Zebao(College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210024,Jiangsu,China;Zhejiang Ninghai Pumped Storage Co.,Ltd.,Ninghai 315600,Zhejiang,China)

机构地区:[1]河海大学水利水电学院,江苏南京210024 [2]浙江宁海抽水蓄能有限公司,浙江宁海315600

出  处:《水力发电》2023年第8期72-78,共7页Water Power

基  金:国家自然科学基金资助项目(51779084)。

摘  要:颗粒级配是影响堆石坝等工程质量的重要因素,采用数字图像处理技术分析级配时,因目标颗粒常处于堆叠状态,颗粒之间存在遮挡,以致影响级配计算精度。从像素级角度分析边缘方向一致性、最小预测角度与预测点位选择之间的关系,提出了多阶段颗粒边缘形态统计学快速预测方法,统计9003组颗粒边缘信息,对面积缺失程度10%、20%、30%、40%的不完整颗粒边缘进行形态预测和补全。结果显示,边缘预测高效,预测图像符合颗粒基本形状,补全后的级配准确率大于95%,说明预测结果与颗粒真实状态具有较高的匹配度。The granular gradation has crucial influence on the construction quality of rock-fill dams and other similar projects.When using the digital image processing technique to analyze the gradation,the stacked state of particle affects the accuracy of aggregate distribution on account of the overlap between particles.A rapid multi-staged particle edge prediction method is proposed based on statistics at pixel-level,which analyzes the correlation between the choice of prediction positions and the consistency of edge direction and minimum prediction angle.The edge information of 9003 groups of particles are analyzed,and the prediction and complement for the original incomplete particle edge with area missing degrees of 10%,20%,30%and 40%are successfully carried out,respectively.The results indicate that this prediction method of particle edge is efficient,the predicted edges meet the basic shape of particles,and the average accuracy rate of completed granular gradation is greater than 95%,indicating that the predicted results have a high matching degree with the real state of particles.

关 键 词:级配分析 颗粒边缘预测 堆叠颗粒 颗粒形态补全 数字图像处理 

分 类 号:TP391.41[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象