Name
Abstract | ||
|---|---|---|
Ring artifacts are inevitable in microtomographic images. In a Digital Rock workflow, such defects might affect the subsequent segmentation and flow simulation. We propose a correction of ring artifacts in reconstructed microtomographic images by inpainting. Our blind inpainting method uses a 3D convolutional network U-net. For the creation of training and validation datasets, we suggest an algorithm for transferring real ring artifacts to an arbitrary place in the undistorted slices of 8 big images of sandstones and sand. The parameters of the deep neural network and loss functions are analyzed. A loss function based on the multi-scale structural similarity index (MS-SSIM) allows to achieve the best performance. The developed solution corrects ring artifacts perfectly from a point of view of visual assessment and outperforms existing inpainting methods according to quality metrics based on MS-SSIM and mean absolute error (MAE). |
Year | DOI | Venue |
|---|---|---|
2020 | 10.23919/FRUCT48808.2020.9087422 | 2020 26th Conference of Open Innovations Association (FRUCT) |
Keywords | DocType | ISSN |
ring artifacts,reconstructed microtomographic images,blind inpainting method,3D convolutional network U-net,loss function,image segmentation,flow simulation,3D CNN,digital rock workflow,deep neural network,loss functions,multiscale structural similarity index,sandstones | Conference | 2305-7254 |
ISBN | Citations | PageRank |
978-1-7281-4257-9 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Anton S. Kornilov | 1 | 0 | 0.34 |
| Ilia Safonov | 2 | 0 | 0.34 |
| Ivan Yakimchuk | 3 | 0 | 1.01 |