Abstract | ||
---|---|---|
In this paper, we propose a novel method that automatically extracts characteristics of cracks such as length, thickness and direction, etc., from a concrete surface image with a series of image processing techniques. We use the closing morphologic operation to adjust the effect of light extending over the whole concrete surface image. After applying the high-pass filtering operation to sharpen boundaries of cracks, we classify intensity values of the image into 8 groups and remove intensity values that belong to the highest frequency group among them for the removal of background. Then, we binarize the preprocessed image. The auxiliary lines used to measure cracks of concrete surface are removed from the binarized image with position information extracted by the histogram operation. Then cracks broken by the removal of background are extended to reconstruct an original crack with the 5x5 masking operation. We remove unnecessary information by applying three types of noise removal operations successively and extract areas of cracks from the binarized image. At last, the opening morphologic operation is applied to compensate extracted cracks for measuring characteristics of cracks correctly. We also propose a method that automatically recognizes the directions (horizontal, vertical, -45 degree, 45 direction degree) of the cracks with the FCM-based RBF(Radial Basis Function) neural network. The proposed RBF neural network applied FCM to the learning between the input layer and the middle layer and the Delta learning method to the learning between the middle layer and the output layer. Experiments using real images of concrete surface showed that the proposed method extracts cracks fairly well and measures characteristics of cracks precisely. Also, The proposed FCM-based RBF neural network was effective in recognizing the direction of the extracted cracks. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ISDA.2008.191 | ISDA (1) |
Keywords | Field | DocType |
middle layer,real image,image processing technique,concrete surface image,whole concrete surface image,identifying concrete slab surface,preprocessed image,intensity value,fcm-based rbf network,concrete surface,binarized image,closing morphologic operation,concrete,high pass filter,image processing,high pass filters,data mining,pixel,noise | Histogram,Radial basis function,Pattern recognition,Masking (art),Computer science,Filter (signal processing),Image processing,High-pass filter,Artificial intelligence,Real image,Artificial neural network,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 3 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Young Woon Woo | 1 | 31 | 8.39 |
Doo Heon Song | 2 | 17 | 6.52 |
kwangbaek kim | 3 | 110 | 43.94 |