Concrete may be eroded by carbonization in nature.After carbonization,the volume of concrete shrinkage,and even produce cracks,seriously affecting the durability of concrete.Micron X-ray computed tomography(μm Computed Tomography,μCT)is used to non-destructively detect the internal characteristics of cement-based materials.Starting from the internal cracks caused by carbonation erosion of cement-based materials,the carbonation cracks of cement-based materials are extracted to quantify the crack development of cement-based materials in the process of carbonation erosion.The traditional threshold segmentation technology is suitable for extracting coarse cracks,but it is not effective for extracting fine cracks,and the interference of pores and uncarbonized areas cannot be avoided.Therefore,in this study,the ocular vessel segmentation technology(B-COSFIRE)was introduced into μCT images to try to segment and extract cracks.The results show that the introduction of eyeball blood vessel segmentation technology can accurately identify coarse and fine cracks and extract them without interference from pores and uncarbonated areas.The introduction of this method makes the efficiency of crack extraction significantly improved.
维普
万方数据
