EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments

Pablo Ruiz-Ponce iD, David Ortiz-Perez, Jose Garcia-Rodriguez iD, Benjamin Kiefer

Sensors · 02.04.2023 · 23 (7) · 3691

DOI: 10.3390/s23073691 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 02.04.2023. Verfasst von Pablo Ruiz-Ponce und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Technology and Computation, University of Alicante, 03690 San Vicente del Raspeig, Spain. Die Arbeit wurde bisher 53-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: MCIN/AEI/10.13039/501100011033, European Union NextGenerationEU/PRTR, HORIZON-MSCA-2021-SE-0.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Certain fields present significant challenges when attempting to train complex Deep Learning architectures, particularly when the available datasets are limited and imbalanced. Real-time object detection in maritime environments using aerial images is a notable example. Although SeaDronesSee is the most extensive and complete dataset for this task, it suffers from significant class imbalance. To address this issue, we present POSEIDON, a data augmentation tool specifically designed for object detection datasets. Our approach generates new training samples by combining objects and samples from the original training set while utilizing the image metadata to make informed decisions. We evaluate our method using YOLOv5 and YOLOv8 and demonstrate its superiority over other balancing techniques, such as error weighting, by an overall improvement of 2.33% and 4.6%, respectively.

Schlagwörter

Stichwörter der Autor:innen

YOLOaerial imagesdata augmentationdata imbalancemaritime environmentsobject detection

Autor:innen und Institutionen

  • Department of Computer Technology and Computation, University of Alicante, 03690 San Vicente del Raspeig, Spain.
  • Faculty of Science, University of Tuebingen, 72076 Tuebingen, Germany.

ORCID-Profile: Pablo Ruiz-Ponce, Jose Garcia-Rodriguez

Förderung

  • MCIN/AEI/10.13039/501100011033
  • European Union NextGenerationEU/PRTR
  • HORIZON-MSCA-2021-SE-0

Zitieren

APA

Ruiz-Ponce, P., Ortiz-Perez, D., Garcia-Rodriguez, J., & Kiefer, B. (2023). POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments. Sensors, 23(7), 3691. https://doi.org/10.3390/s23073691

Vancouver

Ruiz-Ponce P, Ortiz-Perez D, Garcia-Rodriguez J, Kiefer B. POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments. Sensors. 2023;23(7):3691. doi:10.3390/s23073691

BibTeX
@article{RuizPonce2023,
  author = {Ruiz-Ponce, Pablo and Ortiz-Perez, David and Garcia-Rodriguez, Jose and Kiefer, Benjamin},
  title = {POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {7},
  pages = {3691},
  doi = {10.3390/s23073691}
}
Kostenlos starten →