Abstract
Obtaining an accurate segmentation of images obtained by computed microtomography (micro-CT) techniques is a non-trivial process due to the wide range of noise types and artifacts present in these images. Current methodologies are often time-consuming, sensitive to noise and artifacts, and require skilled people to give accurate results. Motivated by the rapid advancement of deep learning-based segmentation techniques in recent years, we have developed a tool that aims to fully automate the segmentation process in one step, without the need for any extra image processing steps such as noise filtering or artifact removal. To get a general model, we train our network using a dataset made of high-quality three-dimensional micro-CT images from different scanners, rock types, and resolutions. In addition, we use a domain-specific augmented training pipeline with various types of noise, synthetic artifacts, and image transformation/distortion. For validation, we use a synthetic dataset to measure accuracy and analyze noise/artifact sensitivity. The results show a robust and accurate segmentation performance for the most common types of noises present in real micro-CT images. We also compared the segmentation of our method and five expert users, using commercial and open software packages on real rock images. We found that most of the current tools fail to reduce the impact of local and global noises and artifacts. We quantified the variation on human-assisted segmentation results in terms of physical properties and observed a large variation. In comparison, the new method is more robust to local noises and artifacts, outperforming the human segmentation and giving consistent results. Finally, we compared the porosity of our model segmented images with experimental porosity measured in the laboratory for ten different untrained samples, finding very encouraging results.
Highlights
Obtaining an accurate segmentation of images obtained by computed microtomography techniques is a non-trivial process due to the wide range of noise types and artifacts present in these images
The recent advances in deep-learning technologies based on neural networks have led to the emergence of high-performance automatic segmentation techniques, with the highest accuracy rates on popular b enchmarks[7]
The high degree of abstraction of deep learning methods has proven to be very effective compared to other segmentation techniques
Summary
Obtaining an accurate segmentation of images obtained by computed microtomography (micro-CT) techniques is a non-trivial process due to the wide range of noise types and artifacts present in these images. Some of the conventional methods are global multi-Otsu t hresholding[5], Marker-controlled Watershed[6], and converging active contours These methods require the use of different types of filters to deal with noise and artifacts. Several forms of multivariant classifiers using machine learning have been used to segment 2D micro-CT rock images[13,14] These last methods require the user to define/paint some areas with the desired labels to train with and perform a segmentation of the whole image. These studies concluded that the more complex architecture U-ResNet-3D performed better for the majority of the cases in terms of accuracy and topological similarity properties As reported in this last work, the authors used a single dataset (1100×1100×2200 voxels) for both training and testing. In this work, unresolved porosity regions are referred to as micro-phase
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
