Abstract: Digital image processing technology has the advantages of high efficiency, real-time, quantification and low cost, which provides a new research direction for asphalt pavement quality detection and evaluation. In view of the difference between the size of the actual area and the number of aggregate particles corresponding to different acquisition heights, the self-made image acquisition equipment was used to collect 20 pieces of asphalt mixture digital images under the image acquisition height of 50~100 cm. Through the regression analysis of the collected images at different heights and the actual corresponding pavement width, the functional relationship between the acquisition height and the actual road size corresponding to the image was proposed. And through the study on increase of aggregate particles over 9.5 mm and size accuracy in asphalt mixture binary image after the digital image preprocessing, the best image acquisition height was proposed. The research results show that:with the increase of image acquisition height, the actual road width of the corresponding area of the image increases proportionally. Based on the results of the increase of aggregate particles and the recognition accuracy of aggregate particles under different acquisition heights, the optimal image acquisition height range is determined as 70~80 cm.

Key words: highway engineering, paving quality, digital image technology, asphalt mixture, uniformity

摘要： 数字图像处理技术具有高效、实时、定量化、成本低的优点，为沥青路面摊铺质量检测与评价提供了新的研究方向。针对不同采集高度对应数字图像实际区域的尺寸和集料颗粒数量存在的差异性问题，通过自制图像采集设备，在50~100 cm图像采集高度下，分别采集20张沥青混合料数字图像，通过对现场不同高度采集图像与实际对应路面宽度进行回归分析，提出采集高度与图像对应实际路面尺寸的函数关系；并通过对数字图像预处理后的沥青混合料二值图像中9.5 mm以上集料颗粒增加量和尺寸精确性进行研究，提出最佳的图像采集高度。研究结果表明：随着图像采集高度的增加，图像对应区域实际路面宽度成比例增加；综合不同采集高度下图像集料颗粒增加量结果和集料颗粒识别精确度结果，确定70~80 cm的高度范围为最佳图像采集高度。

关键词: 道路工程, 摊铺质量, 数字图像技术, 沥青混合料, 均匀性