摘要
利用MATLAB对沥青混合料数字图像进行图像增强和二值化处理,得到了沥青混合料梁型试件断面图像中颗粒的质心、周长、面积、最小外接矩形及等效椭圆等二维形态特征。对描述颗粒粒径的特征参数进行了统计,并使仿真筛分过程精确,得出了碎石级配曲线。考虑到二维参数仿真评价三维实物粒径存在一定的误差,通过回归方程建立了模型。并利用Minitab软件辅助分析,拟合出各粒径通过率的回归方程,依次修正各粒径的通过率,仿真后的级配与设计级配接近。这表明该MATLAB的沥青混合料碎石级配检测方法具有快捷、精确及环保的优点,能有效替代传统筛分方法。
Two-dimensional morphological characteristics of the particles of the specimen sectional images（such as centroid,perimeter,area,minimum boundary rectangle,and equivalent ellipse）are obtained by the image enhancement and binary processing sequentially to the digital image of the asphalt mixture by MATLAB.Gravel grading curve is calculated by statistics and simulation analysis to the characteristic parameters of particle statistics.Because using two-dimensional parameters evaluated three-dimensional particles may result in certain error,regression equation of each particle size passing rate was fitted out by regression equation modeling and using Minitab software assisted analysis,passing rate of each particle size was fixed sequentially,the simulation gradation and design gradation will be close.The results show that the gravel gradation measurement method using MATLAB image processing techniques has many advantages（such as quickly,accurately and environmentally friendly）and can be an effective alternative to the traditional sieving methods.
出处
《交通科学与工程》
2015年第2期23-28,共6页
Journal of Transport Science and Engineering
