论文部分内容阅读
为了研究沥青混合料CT图像空隙率大小及其分布特征,基于Matlab图像处理软件对各层位扫描CT图像进行图像增强、图像锐化处理。采用二维最大熵阈值分割法对沥青混合料不同层位的CT图像进行二值化,利用Canny算子模板进行边缘检测,检测出图像边缘内部空隙点数与试件CT图像总点数做熵运算,计算空隙占整个试件的百分率。同时,对试件空隙率检测计算结果差异的显著性进行数理统计检验。结果表明:空隙率随层位呈两头大中间小的分布趋势,验证了沥青混合料内部微观结构的非均匀性;CT图像计算的空隙率均值可以作为试件的空隙率值,同时试验数据结果与图像处理结果具有一致性,表明图像处理具有一定的准确性和可信度。
In order to study the size and distribution characteristics of voids in CT images of asphalt mixture, image enhancement and image sharpening were performed on all layers of CT images based on Matlab image processing software. The two-dimensional maximum entropy threshold segmentation method was used to binarize the CT images of different layers of asphalt mixture. Edge detection was carried out by using Canny operator template. The number of internal voids in the edge of the image and the total number of CT images in the specimen were calculated. Calculate the percentage of voids in the entire specimen. At the same time, the significance of the difference between the calculated results of the voidage detection of the specimens was tested by mathematical statistics. The results show that the void ratio shows a small distribution trend between the middle and the middle of the horizon, which verifies the inhomogeneity of the microstructure of the asphalt mixture. The average void fraction calculated by CT images can be used as the porosity value of the specimen. Consistent with the image processing results, indicating that the image processing has a certain accuracy and credibility.