- Volume 49 Issue 2
It is necessary to measure cracks on concrete slab surface accurately in concrete structure maintenance for the stability of the structure. However, in real world, the process is done by time consuming and ineffective manual inspection. Although there have been some studies to provide computerized inspection methods, they are vulnerable to rugged surface or noise due to the influence of the light or environmental reasons. In this paper, we propose a new method that extracts not only undistorted cracks but minute cracks that were often regarded as noise. We extract candidate crack areas by applying fuzzy method with R, G, and B channel values of concrete slab structure. Then further refinement processes are performed with SOM algorithm and density based cutoff to remove noise. Experiment verifies that the proposed method is sufficiently useful in various crack images.
- 강현구, 홍성걸, 콘크리트 학회지, 제20권, 제5호, 통권 106호, pp.61-68, 2008.
- 한국콘크리트학회, 최신 콘크리트 공학, 기문당, 1992.
- 강준묵, 오영철, 엄대용, "수치영상처리에 의한 콘크리트 구조물의 균열 정보 획득," 대한토목학회논문집, 22권, 5-D호, pp.1001-1010, 2002.
- K. B. Kim, K. B. Sim, S. H. Ahn, "Recognition of Concrete Surface Cracks using The ART1-based RBF Network," Lecture Notes in Computer Science, LNCS 3972, Springer, pp.669-675, 2006.
- 김광백, 조재현, "ART2 기반 양자화를 이용한 콘크리트 슬래브 표면의 균열 검출," 한국해양정보통신학회논문지, 12권, 10호, pp.1897-1902, 2008.
- A. Kandel, G. Langholz, Fuzzy Control Systems, CRC Press, Inc., 1994.
- W. Pedrycz, Fuzzy Control and Fuzzy Systems, Research Studies Press Ltd., 1989.
- 김광백, "Color Image Vector Quantization using Enhanced SOM Algorithm," 멀티미디어학회논문지, 7권, 12호, pp.1737-1744, 2004.