| Title | Reconstructing high-fidelity digital rock images using deep convolutional neural networks |
| |
| Author | Bizhani, M ; Haeri
Ardakani, O; Little, E |
| Source | Scientific Reports vol. 12, 2022 p. 1-14, https://doi.org/10.1038/s41598-022-08170-8  Open Access |
| Image |  |
| Year | 2022 |
| Alt Series | Natural Resources Canada, Contribution Series 20210362 |
| Publisher | Nature Research |
| Document | serial |
| Lang. | English |
| Media | paper; on-line; digital |
| File format | pdf |
| Subjects | Science and Technology; general geology; rock analyses; scanning electron microscope analyses; scanning electron microscopy |
| Illustrations | 3-D images; digital images; figures |
| Program | Energy Geoscience Clean Energy Resources - Decreasing Environmental Risk |
| Released | 2022 03 11 |
| Abstract | Imaging methods have broad applications in geosciences. Scanning electron microscopy (SEM) and micro-CT scanning have been applied for studying various geological problems. Despite significant advances
in imaging capabilities, and image processing algorithms, acquiring high-quality data from images is still challenging and time-consuming. Obtaining a 3D representative volume for a tight rock sample takes days to weeks. Image artifacts such as noise
further complicate the use of imaging methods for the determination of rock properties. In this study, we present applications of several convolutional neural networks (CNN) for rapid image denoising, deblurring and super-resolving digital rock
images. Such an approach enables rapid imaging of larger samples, which in turn improves the statistical relevance of the subsequent analysis. We demonstrate the application of several CNNs for image restoration applicable to scientific imaging. The
results show that images can be denoised without a priori knowledge of the noise with great confidence. Furthermore, we show how attaching several CNNs in an end-to-end fashion can improve the final quality of reconstruction. Our experiments with SEM
and CT scan images of several rock types show image denoising, deblurring and super-resolution can be performed simultaneously. |
| Summary | (Plain Language Summary, not published)
Artificial intelligence (AI) is rapidly changing the landscape of many traditional fields of study. In geosciences, microscopic imaging of specimens is a
common method of studying various aspects of geological formation. Despite significant advances in imaging capabilities, it is sometimes difficult to obtain representative high-quality images due to various reasons. In this work, we employ several
state-of-the-arts deep neural networks to enhance microscopic images of various rock samples. Our work demonstrates the use of AI for improving the performance of current imaging methods both in terms of speed of processing data, as well as
automating the workflow. |
| GEOSCAN ID | 329084 |
|
|