就地热再生沥青混合料均匀性的细观评价指标

就地热再生能直接在现场一次性完成路面修复，但因其材料组成和施工工艺复杂等极易出现混合料不均匀问题，为解决此问题，尝试从集料入手，对就地热再生沥青混合料均匀性的评价指标进行研究。首先，采用数码相机获取就地热再生沥青混合料试件截面的数字图像，基于数字图像处理技术识别截面中所有新旧集料的细观结构；然后，采用环扇分割法将截面分成36个等面积区域，基于区域集料颗粒面积比和新集料颗粒偏离度分别提出集料均匀性评价指标D和新集料均匀性评价指标H；最后，通过改变RAP加热温度、RAP拌和时间、新沥青混合料拌和温度与新旧料混合时间4个因素，进行正交试验，进一步分析均匀性指标的变化规律和可靠性。结果表明：环形分区结合OTSU阈值分割方法可准确识别沥青混合料截面图像中的集料信息，保留绿色通道的方法可有效识别不同灰度值的新旧集料；对试件截面均匀性的定性分析初步验证了这2个均匀性指标的有效性；RAP加热温度与新沥青混合料拌和温度对D影响显著，RAP加热温度和新旧料混合时间对H影响显著；而且，随着RAP加热温度、新沥青混合料温度和新旧料混合时间的增加，就地热再生混合料的均匀性变好，这与以往的研究结论相一致，进一步验证了这2个均匀性评价指标的可靠性。

Mesoscale Evaluation Index for the Homogeneity of Hot In-place Recycling Asphalt Mixture

The pavement can be rehabilitated directly on a construction site using hot in-place recycling (HIR). However, the homogeneity of HIR asphalt mixture is difficult to control owing to the complex material and complicated construction process. In this study, the evaluation index for homogeneity of HIR asphalt mixture was investigated by analyzing the aggregates of the mixture. First, a digital camera was used to capture images of the cross-sections of the HIR asphalt mixture specimen, and the microstructure of virgin and RAP aggregates in the cross-section was extracted using digital image processing technique. Then, each cross-section of the specimen was divided into 36 regions of equal area using an annular segmentation with sector segmentation. Based on the area ratio of the aggregate and deviation degree of the virgin aggregate, the homogeneity coefficient of aggregates, D, and homogeneity coefficient of virgin aggregates, H, were proposed, respectively. Finally, an orthogonal experiment with four factors was carried out to analyze the reliability and trend of variation of these two homogeneity coefficients. These factors included the heating temperature and mixing time of RAP, heating temperature of virgin asphalt mixture, and mixing time of virgin and RAP mixture. The results showed that the characteristics of aggregate in the cross-section of asphalt mixture could be comprehensively described using annular segmentation combined with the OTSU threshold segmentation method. The virgin and RAP aggregates in the cross-section could be effectively identified owing to different gray values by retaining the green channel. The related macro qualitative test of the homogeneity of the cross-section of the specimen initially showed that these two homogeneity coefficients were reliable. The orthogonal experiment results show that the heating temperature of RAP and the heating temperature of virgin asphalt mixture had a significant effect on D, whereas the heating temperature of RAP and the mixing time of virgin and RAP mixture had a significant effect on H. Moreover, the homogeneity of HIR asphalt mixture was improved by increasing the heating temperature of RAP, heating temperature of virgin asphalt mixture, and mixing time of virgin and RAP mixture, which is consistent with previous research. The reliability of these two homogeneity coefficients was verified again.