Algorithm Study of Edge Location and Feature Measurement of Building Surface Crack Images

Abstract

Old buildings surface cracks are more obvious with the accumulation of its disrepair, which will impact aesthetics, durability and even threaten the safety of persons and properties if feature values surpass the security thresholds. In this paper, firstly, the algorithm of the improved Canny operator gray threshold edge detection and location, based on traditional differential operators and sub-pixel positioning algorithms, achieve coarse location, furthermore, the sub-pixel location algorithm of Gaussian curve fitting based on neighbor gradient features realize sub-pixel precision positioning. Finally, crack features are calculated and comparative analysis through the measurement algorithm. The experiments prove the algorithms possess high accuracy and feasibility with important application values of assessment, identification and maintenance.