Articles | Volume 40, issue 2
https://doi.org/10.5194/jm-40-163-2021
https://doi.org/10.5194/jm-40-163-2021
Research article
 | 
22 Oct 2021
Research article |  | 22 Oct 2021

Semantic segmentation of vertebrate microfossils from computed tomography data using a deep learning approach

Yemao Hou, Mario Canul-Ku, Xindong Cui, Rogelio Hasimoto-Beltran, and Min Zhu

Related authors

A novel specimen-based mid-Paleozoic dataset of antiarch placoderms (the most basal jawed vertebrates)
Zhaohui Pan, Zhibin Niu, Zumin Xian, and Min Zhu
Earth Syst. Sci. Data, 15, 41–51, https://doi.org/10.5194/essd-15-41-2023,https://doi.org/10.5194/essd-15-41-2023, 2023
Short summary

Related subject area

Taxonomy
Novel heterococcolithophores, holococcolithophores and life cycle combinations from the families Syracosphaeraceae and Papposphaeraceae and the genus Florisphaera
Sabine Keuter, Jeremy R. Young, Gil Koplovitz, Adriana Zingone, and Miguel J. Frada
J. Micropalaeontol., 40, 75–99, https://doi.org/10.5194/jm-40-75-2021,https://doi.org/10.5194/jm-40-75-2021, 2021
Short summary

Cited articles

Adams, R. and Bischof, L.: Seeded region growing, IEEE Trans. Pattern Anal. Mach. Intell., 16, 641–647, https://doi.org/10.1109/34.295913, 1994. 
Andreev, P., Coates, M. I., Karatajute-Talimaa, V., Shelton, R. M., Cooper, P. R., Wang, N., and Sansom, I. J.: The systematics of the Mongolepidida (Chondrichthyes) and the Ordovician origins of the clade, PeerJ, 4, e1850, https://doi.org/10.7717/peerj.1850, 2016. 
Bhardwaj, S. and Mittal, A.: A survey on various edge detector techniques, Proc. Technol., 4, 220–226, https://doi.org/10.1016/j.protcy.2012.05.033, 2012. 
Botella, H., Blom, H., Dorka, M., Ahlberg, P. E., and Janvier, P.: Jaws and teeth of the earliest bony fishes, Nature, 448, 583–586, https://doi.org/10.1038/nature05989, 2007. 
Bourel, B., Marchant, R., de Garidel-Thoron, T., Tetard, M., Barboni, D., Gally, Y., and Beaufort, L.: Automated recognition by multiple convolutional neural networks of modern, fossil, intact and damaged pollen grains, Comput. Geosci., 140, 104498, https://doi.org/10.1016/j.cageo.2020.104498, 2020. 
Download
Short summary
In this study, we constructed an open dataset, which contains computed tomography (CT) data on nearly 500 vertebrate microfossils. We propose a semantic segmentation method for CT fish microfossil data based on deep learning (DL). We expect that our proposed method could be applied to CT data on other fossils with good performance. We also believe the fast-accumulating CT data on vertebrate microfossils might become a source of information-rich datasets for deep learning.