It is distinctly important to precisely detect and quantify the micro-cracks in shale rock sample, which serve as the potential fluid flow conduits. In the SEM scanning images, because of the low contrast of the micro-cracks and background and extremely central grayscale value distribution, the segmentation is always insufficient. In this paper, a transformed operator, incomplete beta function, is applied and the simulated annealing method is used to optimize the parameters --a and ß. Simulated annealing method, which is a self-adaptive algorithm, can provide different optimal values depending on different input images. The result of the classical segmentation technique – Otsu’s method, acts as the evaluation parameter. After the processing, the target becomes more obvious by human vision. In addition, the grayscale value distribution is expanded, and is not restricted to the high value range only. Most importantly, the contrast of target and background increases from 0.1216 to 0.2118 and the volume fraction of micro-cracks increases from 1.25 % to 4.02 %, which show that the segmentation after processing is more accurate. All the observation and evaluation parameters show the strong ability of image enhancement studied in this paper.
Micro-Crack Segmentation of Middle Bakken Shale Rock Sample With High-Resolution SEM – The Application of Self-Adaptive Image Enhancement Technique
- Share Icon Share
- Search Site
Wang, Hongsheng, Rabiei, Minou, and Sai Wang. "Micro-Crack Segmentation of Middle Bakken Shale Rock Sample With High-Resolution SEM – The Application of Self-Adaptive Image Enhancement Technique." Paper presented at the 52nd U.S. Rock Mechanics/Geomechanics Symposium, Seattle, Washington, June 2018.
Download citation file: