|
| | Abstract | | In geo-investigation, identifying cracks from the spread image of a borehole wall is one of the most common uses of borehole imaging and that picking up the cracks manually is time-consuming and easily affected by objective judgment. We have studied this problem and developed a method that identifies cracks automatically by using image-processing techniques. In our study, first, we translate the image from RGB color model to HSV color model which is nearer to the color recognition of human sight, second, we apply various filtering processes to improve the contrast between cracks and the background, then, we do segmentation and edging on the improved image, finally, we estimated the crack parameters such as depth, direction, dip angle and opening width by Hough Transform. We tested this method using real borehole image data. The result shown that the HSV color model can give better S/N ratio than RGB and with help of various image processing, the Hough Transform can extract the cracks from the image successfully. |
|
|
|
| |
|