EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Automated computer vision analysis of pen-level feeder occupancy in post-weaning piglets under two pen configurations and by sex

Automatisierte Computer-Vision-Analyse der Futterstellenbelegung auf Buchtenebene bei Absatzferkeln unter zwei Buchtenkonfigurationen und nach Geschlecht

Victor Hugo Silva Souza, Mamunur Rahman, Maria Cristina de Souza Silva, Luana Maria Benicio, Vinicius de Souza Cantarelli, Alessandro Torres Campos, Isabella C F S Condotta iD

Translational Animal Science · 15.01.2026 · 10

DOI: 10.1093/tas/txag150 ↗

Auf einen Blick

Bei 80 abgesetzten PIC-Ferkeln lieferte die Videoanalyse eine hohe Erkennungsleistung. Die Futterstellenbelegung unterschied sich nach Beobachtungsfenster und Buchtenkonfiguration, nicht nach Geschlecht.

Fragestellung

Lässt sich die Futterstellenbelegung auf Buchtenebene bei Absatzferkeln automatisiert per Video erfassen und unterscheidet sie sich nach Buchtenkonfiguration und Geschlecht?

Methode

YOLOv8n-Detektor zur Klassifikation von Videoframes als at_feeder oder not_at_feeder; 16 Einzelgeschlechtsbuchten in 2 × 2-Design; Auswertung über drei 24-h-Fenster an Tag 1, 4 und 7.

Ergebnisse

Validierung auf 2000 Frames: mAP@0.5 = 0,995 und F1-Score = 0,993 für beide Klassen. Effekt des Beobachtungsfensters (P < 0,001) und Interaktion Buchtenkonfiguration × Fenster (P = 0,014); Geschlecht nicht signifikant. Jedes weitere Stunde Belegung: 0,023 kg/Ferkel mehr Futterverschwendung im 24-h-Intervall (95 %-KI 0,010 bis 0,037; P = 0,003).

Schlussfolgerung der Autor:innen

Automatisierte Videoanalyse lieferte ein hochdurchsatzfähiges Maß für Futterstellenbelegung auf Buchtenebene in der frühen Phase nach dem Absetzen und bildet eine Basis für spätere individuelle Fütterungsmaße.

Kernaussagen

  • 80 PIC-Ferkel, 16 Einzelgeschlechtsbuchten, Auswertung an Tag 1, 4 und 7
  • mAP@0.5 0,995; F1-Score 0,993 in der Validierung
  • Geschlecht kein signifikanter Faktor; Belegung nach Fenster und Konfiguration unterschiedlich

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Translational Animal Science, online erschienen am 23.10.2026. Verfasst von Victor Hugo Silva Souza und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Animal Sciences, University of Illinois Urbana-Champaign, Urbana, IL 61801, United States. Die Arbeit nennt 14 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Agriculture and Food Research Initiative.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The nursery phase is a challenging period in swine production, and automated monitoring of feeder use may provide additional information about piglet activity during early post-weaning. A YOLOv8n detector trained using whole-frame bounding boxes was used to classify video frames as at_feeder or not_at_feeder. These classifications were used to quantify pen-level feeder occupancy, defined as the percentage of analyzed frames in which at least one piglet had its head inserted into the feeder. Eighty PIC piglets weaned at 21 d were housed in 16 single-sex pens. Eight pens contained four piglets (0.39 m²/pig; 18.75 cm feeder length/pig) and eight contained six piglets (0.26 m²/pig; 12.5 cm feeder length/pig), resulting in two pen configurations that differed simultaneously in group size, floor-space allowance, and feeder-space allowance. Sex and pen configuration formed a 2 × 2 factorial arrangement with four pens per cell, and the 16 pens were distributed across four spatial blocks. Video was analyzed during three 24-h windows beginning at 0900 h on Day 1, the day of weaning, and on Days 4 and 7. On a balanced 2000-frame within-recording validation set, the model achieved mAP@0.5 = 0.995 and an F1-score of 0.993 for both classes. A mixed model with spatial block as a fixed effect and pen as a random intercept identified an observation-window effect (P < 0.001) and a pen configuration × window interaction (P = 0.014). Sex and interactions involving sex were not significant. Occupancy was higher in six-piglet pens during the later observation windows. An illustrative sensitivity analysis showed that the later differences between pen configurations were compatible with the expected mathematical effect of group size on a presence-based occupancy measure. Average daily gain and feed disappearance did not differ by sex, pen configuration, or their interaction. Feed conversion ratio showed a sex × pen-configuration interaction (P = 0.010). In a secondary within-pen analysis, each additional hour of feeder occupancy was associated with 0.023 kg/pig greater feed disappearance over the temporally matched 24-h interval (95% confidence interval, 0.010 to 0.037 kg/pig; P = 0.003). Automated video analysis provided a high-throughput measure of pen-level feeder occupancy across the early post-weaning period, establishing a foundation for subsequent development of individual-level feeding measures.

Schlagwörter

Stichwörter der Autor:innen

automated monitoringcomputer visiondeep learningfeeder occupancynurseryswine

Autor:innen und Institutionen

  • Department of Animal Sciences, University of Illinois Urbana-Champaign, Urbana, IL 61801, United States.
  • Department of Animal Science, Federal University of Lavras, Lavras, MG 37200-900, Brazil.
  • Department of Agricultural & Biological Engineering, University of Illinois Urbana-Champaign, Urbana, IL 61801, United States.
  • Department of Agricultural Engineering, Federal University of Lavras, Lavras, MG 37200-900, Brazil.

ORCID-Profile: Isabella C F S Condotta

Förderung

  • Agriculture and Food Research Initiative

Zitieren

APA

Souza, V. H. S., Rahman, M., Silva, M. C. S., Benicio, L. M., Cantarelli, V. S., Campos, A. T., & Condotta, I. C. F. S. (2026). Automated computer vision analysis of pen-level feeder occupancy in post-weaning piglets under two pen configurations and by sex. Translational Animal Science, 10. https://doi.org/10.1093/tas/txag150

Vancouver

Souza VHS, Rahman M, Silva MCS, Benicio LM, Cantarelli VS, Campos AT, et al. Automated computer vision analysis of pen-level feeder occupancy in post-weaning piglets under two pen configurations and by sex. Translational Animal Science. 2026;10. doi:10.1093/tas/txag150

BibTeX
@article{Souza2026,
  author = {Souza, Victor Hugo Silva and Rahman, Mamunur and Silva, Maria Cristina de Souza and Benicio, Luana Maria and Cantarelli, Vinicius de Souza and Campos, Alessandro Torres and Condotta, Isabella C F S},
  title = {Automated computer vision analysis of pen-level feeder occupancy in post-weaning piglets under two pen configurations and by sex},
  journal = {Translational Animal Science},
  year = {2026},
  volume = {10},
  doi = {10.1093/tas/txag150}
}
Kostenlos starten →