基于图像处理的集料形态特征定量评价  

Quantitative Evaluation of Morphological Feature of Aggregate Based on Image Processing

作　　者：耿超[1,2] GENG Chao(School of Highway,Chang'an University,Xi'an Shaanxi 710364,China;Anhui Expressway Co.,Ltd.,Hefei Anhui 230088,China)

机构地区：[1]长安大学公路学院,陕西西安710064 [2]安徽皖通高速公路股份有限公司,安徽合肥230088

出　　处：《公路交通科技》2018年第12期42-47,共6页Journal of Highway and Transportation Research and Development

摘　　要：数字图像处理技术的引入,将会实现矿质集料颗粒形态特征的直接、客观的定量评价。通过对矿质集料数字图像进行噪声消除、图像分割、二值化形态学处理、边缘跟踪以及颗粒分析,得到了各个集料颗粒的面积、周长、拟合椭圆等形态特征参数。进一步对各参数进行统计分析,得到了针对集料的圆度、棱角性、表面纹理以及针片状颗粒含量等形态特征指标的分析结果。基于成像相关知识,采用基于LED光源的背光图像采集平台,对不同档位的集料颗粒形态特征进行定量评价和分析。试验结果表明:通过数字图像处理技术对不同档位集料颗粒形态特征参数进行图像描述,可以遴选出高效可行的有利于定量分析形态特征的参数;背光源采集方法可以更好提取集料颗粒图像边缘特征,使图像更有利于后续处理;不同档位下集料的圆度趋势比较明显,随粒径的增大而减小;棱角性变化与圆度截然不同,其中粗集料棱角性基本不变,而细集料随粒径的增大变化明显;粗集料的纹理指标值变化不大;粗集料的针片状颗粒含量为0,细集料的针片状颗粒含量大于1%。研究可以为矿质集料进场质量的过程控制提供快速高效的定量依据。The introduction of digital image processing technology will realize the direct and objective quantitative evaluation of the morphological characteristics of mineral aggregate particles.The morphological feature parameters of each kind of particles such as area,perimeter and fitting ellipse are obtained though the noise elimination,image segmentation,morphological binarization,edge tracking and the particle analysis to the digital images of the mineral aggregates.By statistical analysis on these parameters,the result of the morphological feature of aggregates such as roundness,angularity,surface texture,fiat and elongated particle content are obtained.Based on imaging knowledge,a backlight image acquisition platform with LED light source is used to quantitatiG,ely evaluate and analyze the morphological feature of aggregate particles in differeni grades..The test result shows that (1)by using digital image processing technology to describe the morphological feature parameters of aggregate particles of different grades,the efficient and feasible parameters which are beneficial to quantitative analysis of morphological feature can be'selected;(2)the image edge features of aggregate particles can be extracted by baeklight acquisition method to make it more suitable for following processing;(3)the roundness trends of aggregates at different grades are obvious, which decrease with the increase of particle size;(4)the change of angularity is quite different from that of roundness,in which the angularity of coarse aggregate did not changed basically,while that of fine aggregate changed obviously with the increase of particle size ;(5)the texture index values of coarse aggregate has notchanged much;(6)the flat and elongated particle content of coarse aggregates is O,and that of fine aggregate is more than 1%.The research can provide a fast and lefficient quantitative basis for process control of the incoming quality of mineral aggregates.

关 键 词：道路工程 矿质集料 定量评价 图像分析 集料形态特征 背光图像采集平台 

分 类 号：U416.217[交通运输工程—道路与铁道工程]