摘要
为了研究粗集料表面的三维纹理特征,基于粗集料表面三维点云数据提出了改进的拉普拉斯平滑算法。为了限制拉普拉斯平滑前后点云的过度变形,通过引入偏差限制参数控制粗集料颗粒表面的点允许移动的最大距离,将平滑尺度限制在纹理尺度范围内。根据改进的拉普拉斯平滑算法原理,基于粗集料在平滑前后表面积的变化幅度提出了三维纹理评价指标。采用商用蓝光三维扫描仪采集了不同粒径石灰岩集料和玄武岩集料的三维点云图像。采用改进的拉普拉斯平滑算法,研究了不同粒径石灰岩集料和玄武岩集料的纹理特征。结果表明:(1)2.36,4.75,9.5 mm粒径的石灰岩集料的纹理指标都服从对数正态分布,13.2 mm和16 mm粒径的石灰岩集料纹理指标都服从正态分布,2.36,4.75,9.5 mm粒径的玄武岩集料的纹理指标都服从正态分布;(2)在石灰岩集料中,4.75 mm集料的纹理指标最大,当粒径大于4.75 mm时,随着集料粒径的增大,纹理指标的均值逐渐减小;(3)在玄武岩集料中,随着集料粒径的增大,纹理指标的均值逐渐减小,2.36,4.75,9.5 mm粒径的玄武岩集料的纹理指标均值都低于石灰岩集料;(4)在石灰岩和玄武岩集料中,随着粒径的增大,纹理指标的变异系数都是先增加后减小,4.75 mm集料纹理指标的离散性最显著。
In order to study the 3D texture features of coarse aggregate surface,an improved Laplacian smoothing algorithm is proposed based on the 3D point cloud data of coarse aggregate surface.To prevent the point cloud over-deformation before and after the Laplacian smoothing operation,the maximum distance allowed for the movement of the points on the surface of coarse aggregate particles is controlled by introducing a deviation limitation parameter,and the smoothing scale is limited within the texture scale range.According to the principle of the improved Laplacian smoothing algorithm,the 3D texture evaluation indicators are proposed based on the change amplitude of surface area of coarse aggregate before and after smoothing.The 3D point cloud images of different particle sized limestone aggregate and basalt aggregate are collected by commercial blue light 3D scanner.The texture features of different particle sized limestone aggregate and basalt aggregate are studied by using the improved Laplacian smoothing algorithm.The result shows that(1)The texture indicators of limestone aggregates with particle sizes of 2.36,4.75,9.5 mm all obey lognormal distribution.The texture indicators of limestone aggregates with particle sizes of 13.2 mm and 16 mm all obey normal distribution.The texture indicators of basalt aggregates with particle sizes of 2.36,4.75,9.5 mm obey normal distribution.(2)In the limestone aggregates,the texture indicators of the 4.75 mm aggregate is the largest,and the mean texture indicators gradually decrease with the increase of aggregate particle size when the particle size is larger than 4.75 mm.(3)In the basalt aggregates,the mean texture indicators gradually decrease with the increase of particle size.The average texture indicators of basalt aggregates with particle sizes of 2.36,4.75,9.5 mm are all smallerr than those of limestone aggregates.(4)In the limestone and basalt aggregates,the variation coefficients of the texture indicators both increase first and then decrease with the increase of the agg
作者
刘健宇
张海
王涛
张东
LIU Jian-yu;ZHANG Hai;WANG Tao;ZHANG Dong(Guizhou Highway Construction and Maintenance Group Co.,Ltd.,Guiyang Guizhou 550000,China;Nanjing Tech University,Nanjing Jiangsu 211816,China)
出处
《公路交通科技》
CSCD
北大核心
2023年第8期57-62,共6页
Journal of Highway and Transportation Research and Development
基金
贵州省交通运输厅科技项目(2021-122-006)
江苏省碳达峰碳中和科技创新专项资金项目(BE2022605)。
关键词
道路工程
粗集料
拉普拉斯平滑算法
纹理
统计分析
road engineering
coarse aggregate
Laplacian smoothing algorithm
texture
statistical analysis
