An improved image segmentation algorithm and measurement methods for asphalt mixtures

Asphalt mixture is the most widely used pavement materials over the world, whose microstructure always plays an important role in construction which can be measured or studied by image analysis conveniently. However, there is no reliable segmentation or standard measurement for asphalt mixture images which blocks further researches. An improved multilevel threshold algorithm via Kapur entropy based on shuffled frog leaping algorithm is proposed which can appropriately solve the hot asphalt mixture images´ segmentation problem. In comparison with traditional methods, the experiments of segmenting images are illustrated to show that the proposed method can get ideal segmentation result with less computation cost using the shuffled frog leaping algorithm. A device that can capture the asphalt mixture´s standard images objective and quantitative the asphalt mixture microstructure indexes after segmentation are also proposed which can be a novel measurement of asphalt mixture in applications.