基于改进分水岭算法的堆积态再生混合粗骨料图像分析  

Practical Method for Identifying Morphological Information of Recycled Coarse Aggregate Bulk Based on Watershed Algorithm

作　　者：杨子晴 杨健[1,2,3] 熊吴越 YANG Ziqingu;YANG Jian;XIONG Wuyue(School of Naval Architecture,Ocean and Civil Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;State Key Laboratory of Ocean Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Shanghai Key Laboratory for Digital Maintenance of Buildings and Infrastructure,Shanghai 200240,China)

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200240 [2]上海交通大学海洋工程国家重点实验室,上海200240 [3]上海市公共建筑和基础设施数字化运维重点实验室,上海200240

出　　处：《硅酸盐学报》2021年第8期1691-1698,共8页Journal of The Chinese Ceramic Society

基　　金：上海市科委“科技创新行动计划”(18DZ1205603,20DZ1201301)。

摘　　要：提出了一种基于改进分水岭算法分析堆积态混凝土–碎砖再生混合粗骨料图像的方法,并基于体视学原理建立了二维轮廓信息与三维粗骨料几何特性的数学关系。根据混凝土再生粗骨料和碎砖再生粗骨料的色相差异和粗骨料颗粒间的距离,引入阈值分割和距离变换分步对堆积态的再生混合粗骨料图像进行颗粒边缘的识别和分割。结果表明:提出的分析方法可以识别不同种类粗骨料和提取粗骨料颗粒轮廓信息,并快速计算出粗骨料的粒径、级配、碎砖粗骨料含量、长宽比、球度等几何特性。In order to achieve particle segmentation and extract morphological information from the images of brick-containing recycled coarse aggregate bulk in the stacked state, an image-based method was proposed. Based on the differences in hue and distance between particles of the recycled concrete and crushed brick aggregate, the model was incorporated into the threshold segmentation and distance transformation, thus optimizing the watershed segmentation algorithm. It is indicated that the optimized segmentation algorithm could be used to effectively extract the contours of individual aggregate particles and calculate their morphological information, such as aggregate gradation, proportion of each aggregate, flakiness ratio and sphericity.

关 键 词：混凝土 再生骨料 数字图像技术 图像分割 几何特性 

分 类 号：TU525[建筑科学—建筑技术科学]