表面光洁度
立体视
背景(考古学)
纹理(宇宙学)
分形维数
沥青
计算机科学
航程(航空)
沥青路面
判断
人工智能
计算机视觉
环境科学
分形
工程类
地质学
数学
图像(数学)
机械工程
地理
地图学
古生物学
法学
航空航天工程
数学分析
政治学
作者
Glenn R. Matlack,Andrea Horn,Aldo Aldo,Lubinda F. Walubita,Bhaven Naik,Issam Khoury
标识
DOI:10.1080/14680629.2021.2009902
摘要
Road managers require convenient, inexpensive methods to quickly assess pavement conditions over large areas. New imaging technologies potentially offer a substantial improvement over current methods. In this study, two hand-portable methods based on widely available consumer-grade technology were tested across a range of pavement conditions. To place the proposed methods in a familiar context, data were compared with the mean texture depth (MTD) calculated by the widely used volumetric method. MTD provided the clearest distinction between experimental roughness classes. Spatial pattern analysis (fractal dimension) of digital images discriminated between light, moderate and severe deterioration but failed to make fine distinctions between degrees of moderate wear. Three-dimensional models constructed by stereoscopic infrared scanning distinguished between all but the smoothest texture classes. It appears that both the proposed methods have potential for accurate, quick and convenient assessment of pavement macrotexture at little expense, although some judgement must be used in their application.
科研通智能强力驱动
Strongly Powered by AbleSci AI