基于显微视觉的沥青路面微观纹理三维重构

Three-dimensional reconstruction of asphalt pavement microtexture based on micro-vision

针对沥青路面微观形貌常用检测方法的不足,融合显微视觉的局部放大特性和数字图像处理技术,提出了基于显微视觉的沥青路面微观纹理三维重构方法。根据激光三角测量原理建立显微视觉三维重构数学模型;利用立体显微镜采集序列图像,对图像进行插值处理和阈值分割,并提取光条中心线;采用Delaunay三角剖分算法实现表面三维重构。与基于聚焦深度的沥青路面微观纹理三维重构方法进行试验对比分析,结果表明,本文方法对光照均匀性无特殊要求,具有精度高、操作简便、三维直观性强等特点。

Aiming at the weakness of common detecting methods for asphalt pavement microtexture,this paper proposes a three-dimensional image reconstruction method of asphalt pavement microtexture based on micro-vision which combines the local amplifying characteristics of micro-vision and the digital image processing technique.A three-dimensional reconstruction mathematic model of micro-vision is established based on the laser triangulation principle.The sequential images are captured by stereo microscope,then processed by interpolating,threshold segmentation and extraction of the light stripe centerline.Finally,the visualization of asphalt pavement microtexture is realized by Delaunay triangulation algorithm.Compared with the 3Dimage reconstruction method of asphalt pavement microtexture based on focus depth,this method has no special requirement for illumination uniformity,and has the features of high accuracy,simple operation and strong 3Dvisualization ability.