Date of Award

January 2018

Document Type


Degree Name

Master of Science (MS)


Petroleum Engineering

First Advisor

Minou Rabiei


As the preferential flow channels in the shale reservoir, the fracture systems including the natural micro-cracks and hydraulic fractures have received great attention from the whole energy industry worldwide. However, it is challenging to quantify the fracture systems in the shale rocks precisely because most of well-developed “histogram-based” image processing techniques cannot handle the case of small target segmentation. Because the fracture apertures are very thin, the over-segmentation or insufficient segmentation would lead to significant error in the quantification, including the fracture porosity, aperture, length, tortuosity etc., which would lead to serious mistakes to the property calculation.

In this research, two novel image processing methods are proposed. The self-adaptive image enhancement method employs incomplete beta function and simulated annealing algorithm to modify the grayscale intensity histogram. The contrast between the target and the background of the transformed gray image reaches the maximum. Also, “self-adaptive” means the enhancement process is specified by the input images. The comparison of segmentation results before and after the image enhancement show that the target becomes more obvious to the naked eyes and the precise fracture porosity of the test image is 4.02 %.

The multi-stage image segmentation (MSS) method combines the global and local information of the image to finish the segmentation. The generated three-dimensional model provides visualization of the fracture systems existing in the core. Also, the important parameters of the fractures can be obtained, including aperture, length, tortuosity, and porosity. Compared with the real permeability from the core-flooding experiments, the permeability calculated from the MSS method has the minimum error of 22.1 %. The results show that the proposed methods in this research can be effective tools for the precise quantification of the thin fracture systems.