이미지프로세싱기법을 이용한 균열인식 알고리즘 및 프로그램 개발

초록

The focus of this study is to develop a program of crack detection on asphalt pavement for the pavement management system of small network. Image processing routines for crack detection on asphalt pavement were developed based on the process of segmentation and interpretation algorithm. Regression-based thresholding method that was derived from ARAN images was shown excellent for segmentation algorithm of the pavement images. The developed algorithm was verified by applying it into other images taken by small digital handy camera. Crack ratio was defined by number of pixel over area of images concerned, which could be required for performance indicator of asphalt pavement. Finally the program of crack detection, called HEART-Crack, was coded with selected algorithms and graphic-user interface.