【作者】 毛利铧；

【导师】 鲍旭东；

【作者基本信息】 东南大学 ， 生物医学工程， 2005， 硕士

【摘要】 随着交通运输的发展,沥青混凝土材料得到了越来越广泛地应用,同时对它的质量要求也越来越高。石子沥青混合料的构成和分布对混凝土的整体质量有非常大的影响,因此如何快速准确的获取混合料的构成和分布特性,成为当前迫切需要解决的问题。本文根据现实需求,引入了数字图像处理技术分析路面结构。在运用数字图像技术处理路面的过程中,图像分割是基础,它决定了图像处理的整个过程的好坏。本文在了解大量现有图像分割技术的前提下,结合路面图片的灰度特征,将阈值分割、区域生长算法和分水岭算法用于路面图片的分割,取得了一定的效果。在图像分割的基础之上,本文还对分割图像进行了后处理以改善分割效果。同时还计算了石子目标的构成和分布特性,为沥青石子混凝土材料的质量分析和性能预测提供技术支持。更多还原

【Abstract】 With the development of traffic, the asphalt-carpolite mixture is widely used. At the same time, the request of the quality of this mixture is increased. The form and the distributing of asphalt and carpolite in the mixture have great influence on projects. So it becomes an urgent problem to be solved how to know the form and the distributing of the mixture rapidly and nicely. In this paper, techniques of digital image processing are introduced to solve the problem.During the course of road image analysis, one of the key points is road image segmentation. The quality of road image segmentation determines the whole process. In this paper, based on the fully review of lots of image segmentation methods and carefully analysis on the road image characteristics, we choose three techniques to segment road images. They are thresholding technique, region growing method and watershed arithmetic. The results proved these three techniques are effective.After the segmenting of road images, we even use several techniques to improve the quality of these pictures. Then the form and the distributing characteristics of the carpolite are calculated. The outcomes will support to measure the quality of the mixture. By combining of other knowledge, such as mechanics and material, we can even forecast the capability of the mixture in the future.更多还原