The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects

Tan M, Langenkämper D, Nattkemper TW (2022)
Sensors 22(14): 5383.

Zeitschriftenaufsatz | Veröffentlicht | Englisch
 
Download
OA 2.90 MB
Abstract / Bemerkung
Data augmentation is an established technique in computer vision to foster the generalization of training and to deal with low data volume. Most data augmentation and computer vision research are focused on everyday images such as traffic data. The application of computer vision techniques in domains like marine sciences has shown to be not that straightforward in the past due to special characteristics, such as very low data volume and class imbalance, because of costly manual annotation by human domain experts, and general low species abundances. However, the data volume acquired today with moving platforms to collect large image collections from remote marine habitats, like the deep benthos, for marine biodiversity assessment and monitoring makes the use of computer vision automatic detection and classification inevitable. In this work, we investigate the effect of data augmentation in the context of taxonomic classification in underwater, i.e., benthic images. First, we show that established data augmentation methods (i.e., geometric and photometric transformations) perform differently in marine image collections compared to established image collections like the Cityscapes dataset, showing everyday traffic images. Some of the methods even decrease the learning performance when applied to marine image collections. Second, we propose new data augmentation combination policies motivated by our observations and compare their effect to those proposed by the AutoAugment algorithm and can show that the proposed augmentation policy outperforms the AutoAugment results for marine image collections. We conclude that in the case of small marine image datasets, background knowledge, and heuristics should sometimes be applied to design an effective data augmentation method.
Stichworte
marine objects classification; underwater computer vision; deep learning; data augmentation
Erscheinungsjahr
2022
Zeitschriftentitel
Sensors
Band
22
Ausgabe
14
Art.-Nr.
5383
eISSN
1424-8220
Finanzierungs-Informationen
Open-Access-Publikationskosten wurden durch die Universität Bielefeld gefördert.
Page URI
https://pub.uni-bielefeld.de/record/2964556

Zitieren

Tan M, Langenkämper D, Nattkemper TW. The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects. Sensors. 2022;22(14): 5383.
Tan, M., Langenkämper, D., & Nattkemper, T. W. (2022). The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects. Sensors, 22(14), 5383. https://doi.org/10.3390/s22145383
Tan, Mingkun, Langenkämper, Daniel, and Nattkemper, Tim Wilhelm. 2022. “The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects”. Sensors 22 (14): 5383.
Tan, M., Langenkämper, D., and Nattkemper, T. W. (2022). The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects. Sensors 22:5383.
Tan, M., Langenkämper, D., & Nattkemper, T.W., 2022. The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects. Sensors, 22(14): 5383.
M. Tan, D. Langenkämper, and T.W. Nattkemper, “The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects”, Sensors, vol. 22, 2022, : 5383.
Tan, M., Langenkämper, D., Nattkemper, T.W.: The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects. Sensors. 22, : 5383 (2022).
Tan, Mingkun, Langenkämper, Daniel, and Nattkemper, Tim Wilhelm. “The Impact of Data Augmentations on Deep Learning-Based Marine Object Classification in Benthic Image Transects”. Sensors 22.14 (2022): 5383.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Creative Commons Namensnennung 4.0 International Public License (CC-BY 4.0):
Volltext(e)
Access Level
OA Open Access
Zuletzt Hochgeladen
2022-07-21T08:48:59Z
MD5 Prüfsumme
fcb2d9405f66070c5af7be689b399711


Link(s) zu Volltext(en)
Access Level
Restricted Closed Access

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Web of Science

Dieser Datensatz im Web of Science®
Quellen

PMID: 35891060
PubMed | Europe PMC

Suchen in

Google Scholar