From video to encounter histories: individual identification and machine vision for salmonid capture-recapture monitoring
- 1Institute of Freshwater Research, Swedish University of Agricultural Sciences
Abstract
Individual encounter histories are central to capture-recapture models, but fisheries monitoring is often reduced to counts that do not account for variation in detection probability. This study presents a machine-vision workflow for converting long-term video surveillance of Atlantic salmon (Salmo salar) and sea trout (Salmo trutta) spawning runs into individual-level data for capture-recapture analysis. The workflow detects fish-positive frames, saves video clips, extracts fish-head regions of interest, and organizes cropped images for re-identification. A binary EfficientNetB0 fish detector achieved validation accuracy of 0.9918 and PR AUC of 0.9992; at a conservative threshold of 0.98, validation false positives were eliminated while retaining 94.3% recall. A YOLOv8n model localized fish-head regions, and an EfficientNetB0 ArcFace model trained on head images from 700 identities achieved 99.80% accuracy among accepted known matches and an image-weighted false-accept rate of 0.81% when non-training identities were treated as unknown. Zero-shot closed-set retrieval achieved 99.17% top-1 accuracy across 1,416 identities excluded from model training, demonstrating strong generalization to previously unseen identities.
Related
- Rationalizing the application of spatial capture-recapture versus space-to-event models for monitoring populations of an endangered social carnivore
- IoT-enabled individual poultry monitoring with lightweight step-counting for health assessment
- Identification of a pan-orthoebolavirus-reactive antibody from an rVSV-EBOV vaccinated individual
- Individual recognition and number prediction for Manx shearwater (Puffinus puffinus) by passive acoustic monitoring
- Non-invasive UV-A photo-identification for individual recognition in olms (genus Proteus)
BibTeX
@misc{karlsson2026from,
title = {{From video to encounter histories: individual identification and machine vision for salmonid capture-recapture monitoring}},
author = {K. Karlsson},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.03.749040},
url = {https://www.biorxiv.org/content/10.64898/2026.09.03.749040v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - From video to encounter histories: individual identification and machine vision for salmonid capture-recapture monitoring
AU - K. Karlsson
PY - 2026
DA - 2026/09/03
PB - bioRxiv
DO - 10.64898/2026.09.03.749040
UR - https://www.biorxiv.org/content/10.64898/2026.09.03.749040v1
AB - Individual encounter histories are central to capture-recapture models, but fisheries monitoring is often reduced to counts that do not account for variation in detection probability. This study presents a machine-vision workflow for converting long-term video surveillance of Atlantic salmon (Salmo salar) and sea trout (Salmo trutta) spawning runs into individual-level data for capture-recapture analysis. The workflow detects fish-positive frames, saves video clips, extracts fish-head regions of interest, and organizes cropped images for re-identification. A binary EfficientNetB0 fish detector achieved validation accuracy of 0.9918 and PR AUC of 0.9992; at a conservative threshold of 0.98, validation false positives were eliminated while retaining 94.3% recall. A YOLOv8n model localized fish-head regions, and an EfficientNetB0 ArcFace model trained on head images from 700 identities achieved 99.80% accuracy among accepted known matches and an image-weighted false-accept rate of 0.81% when non-training identities were treated as unknown. Zero-shot closed-set retrieval achieved 99.17% top-1 accuracy across 1,416 identities excluded from model training, demonstrating strong generalization to previously unseen identities.
ER -