KarstSeg3D: Deep Learning for Characterizing Paleokarst Collapse Features in 3-D Seismic Images

This is a Keras version of KarstSeg implemented by Xinming Wu for Paleokarst segmentation in 3D seismic images

As described in Deep Learning for Characterizing Paleokarst Collapse Features in 3-D Seismic Images by Xinming Wu¹, Shangsheng Yan¹, Jie Qi² and Hongliu Zeng³. ¹Computational Interpretation Group, USTC; ²The University of Oklahoma; ³BEG, UT Austin.

Getting Started with Example Model for paleokarst prediction

If you would just like to try out a pretrained example model, then you can download the pretrained model and use the <apply.py> script to run a demo. I recommend to run the prediction on CPU <./cpurun apply.py>, which is fast enough.

Note

The apply.py performs the 3D visualization of results by using Java libraries, which requires install Java (1.8 is recommended). You can simply mute the visualization codes if you got error messages regarding to the plotting.

Dataset

To train our CNN network, we automatically created 120 pairs of synthetic seismic and corresponding karst volumes, which were shown to be sufficient to train a good karst segmentation network.

The training and validation datasets can be downloaded here

Training

Run <train.py> to start training a new karstSeg model by using the 120 synthetic datasets

Publications

If you find this work helpful in your research, please cite:

@article{wu2020karstSeg,
    author = {Xinming Wu and Shangsheng Yan and Jie Qi and Hongliu Zeng},
    title = {Deep Learning for Characterizing Paleokarst Collapse Features in 3-{D} Seismic Images},
    journal = {Journal of Geophysical Research: Solid Earth},
    volume = {125},
    number = { },
    doi = {doi.org/10.1029/2020JB019685},
    year = {2020},
}

License

This extension to the Keras library is released under a creative commons license which allows for personal and research use only. For a commercial license please contact the authors. You can view a license summary here: http://creativecommons.org/licenses/by-nc/4.0/