EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Confounder-Aware Pre-Training Quality Control Workflow for Mechanically Scanned Sonar Echoes in Offshore Aquaculture Cages

Shujie Wan, Haopeng Song, Guodong Li, Jing Cheng, Hang Yang, Zhijun Wang, Yutao Zhang

Journal of Marine Science and Engineering · 12.11.2026 · 14 (22) · 2145

DOI: 10.3390/jmse14222145 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Marine Science and Engineering, online erschienen am 12.11.2026. Verfasst von Shujie Wan und 6 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Project of the Ministry of Agriculture and Rural Affairs of the People’s Republic of China, National Key Research and Development Program of China, Central Public-interest Scientific Institution Basal Research Fund, Chinese Academy of Fishery Sciences.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mechanically scanned sonar can support offshore aquaculture sensing, but acoustic candidate quality and biological identifiability must be evaluated before species modeling. We analyzed 70,421 frames from 22 records and 336 complete rotations acquired on one day in two cages, one per species. A historical top-20-per-rotation detector yielded 6647 candidates, of which 4801 passed morphology rules; an uncapped rerun yielded 27,980 candidates and 14,402 passes. The 5 m between-cage pass-fraction gap fell from 26.9 to 0.4 percentage points after uncapping, revealing selection amplification. A separate probability sample of 128 passes, failures, and lower-ranked candidates yielded 83.6% agreement (κ = 0.627) between two blinded reviewers. Rater-specific, sampling-weighted proportions of automatic passes judged visually isolated were 39.2% and 55.3%; these labels do not establish single-fish identity. In the historical capped 142-case matched cohort, the mixed acoustic/detector set showed a record-level domain association (balanced accuracy 0.919; six-test Holm-adjusted p = 0.0030), conditional on ranking and selection. Species effects cannot be separated from cage effects in this one-cage-per-species design. This case study audits design eligibility, candidate selection, and reviewer dependence before biological classifier development.

Förderung

  • Science and Technology Project of the Ministry of Agriculture and Rural Affairs of the People’s Republic of China
  • National Key Research and Development Program of China
  • Central Public-interest Scientific Institution Basal Research Fund, Chinese Academy of Fishery Sciences

Zitieren

APA

Wan, S., Song, H., Li, G., Cheng, J., Yang, H., Wang, Z., & Zhang, Y. (2026). A Confounder-Aware Pre-Training Quality Control Workflow for Mechanically Scanned Sonar Echoes in Offshore Aquaculture Cages. Journal of Marine Science and Engineering, 14(22), 2145. https://doi.org/10.3390/jmse14222145

Vancouver

Wan S, Song H, Li G, Cheng J, Yang H, Wang Z, et al. A Confounder-Aware Pre-Training Quality Control Workflow for Mechanically Scanned Sonar Echoes in Offshore Aquaculture Cages. Journal of Marine Science and Engineering. 2026;14(22):2145. doi:10.3390/jmse14222145

BibTeX
@article{Wan2026,
  author = {Wan, Shujie and Song, Haopeng and Li, Guodong and Cheng, Jing and Yang, Hang and Wang, Zhijun and Zhang, Yutao},
  title = {A Confounder-Aware Pre-Training Quality Control Workflow for Mechanically Scanned Sonar Echoes in Offshore Aquaculture Cages},
  journal = {Journal of Marine Science and Engineering},
  year = {2026},
  volume = {14},
  number = {22},
  pages = {2145},
  doi = {10.3390/jmse14222145}
}
Kostenlos starten →