What question did this study set out to answer?

To evaluate a non-invasive computer vision framework combining instance segmentation, deep feature extraction, and machine learning regression to estimate pig body weight in farm environments.

September 17, 2026Frontiers in Artificial IntelligenceOpen Access

Evaluation of hybrid models based on image segmentation and inference for pig weight estimation

View Full Paper

Ask AI

Bookmark

Authors

MVMiguel Angel Valles-Coral KRKelvin Lleins Rojas-Córdova LPLloy Pinedo

Discussion

Member takes

Overview

Computational study demonstrates accurate non-invasive weight estimation in pigs using computer vision and support vector regression, suggesting viable tools for precision livestock farming.

Key Points

To evaluate a non-invasive computer vision framework combining instance segmentation, deep feature extraction, and machine learning regression to estimate pig body weight in farm environments.
Collected 3,800 lateral images of pigs paired with ground-truth body weights under real farm conditions in San Martín, Peru.
Standardized images geometrically, performed instance segmentation using YOLOv8n-seg, and extracted morphological feature embeddings via EfficientNet-B0.
Trained supervised regression models (SVR, XGBoost, and CatBoost) and evaluated performance via repeated stratified cross-validation and an independent test set, using Friedman and Wilcoxon post hoc tests with Holm correction.
Support vector regression (SVR) demonstrated the best predictive performance on the independent test set, achieving an RMSE of 2.68 kg, an MAE of 1.81 kg, and an R² of 0.85.
SVR performance was statistically superior compared to XGBoost and CatBoost models based on post hoc Wilcoxon analysis.

Cite This Study

Valles-Coral et al. (2026) studied this question.

synapsesocial.com/papers/6aabb6445f706d05830e4a9d https://doi.org/10.3389/frai.2026.1876396

View Full Paper

Ask AI

Bookmark

September 17, 2026Frontiers in Artificial IntelligenceOpen Access

Evaluation of hybrid models based on image segmentation and inference for pig weight estimation

View Full Paper

Ask AI

Bookmark

Authors

MVMiguel Angel Valles-Coral KRKelvin Lleins Rojas-Córdova LPLloy Pinedo

Discussion

Member takes

Overview

Computational study demonstrates accurate non-invasive weight estimation in pigs using computer vision and support vector regression, suggesting viable tools for precision livestock farming.

Key Points

To evaluate a non-invasive computer vision framework combining instance segmentation, deep feature extraction, and machine learning regression to estimate pig body weight in farm environments.
Collected 3,800 lateral images of pigs paired with ground-truth body weights under real farm conditions in San Martín, Peru.
Standardized images geometrically, performed instance segmentation using YOLOv8n-seg, and extracted morphological feature embeddings via EfficientNet-B0.
Trained supervised regression models (SVR, XGBoost, and CatBoost) and evaluated performance via repeated stratified cross-validation and an independent test set, using Friedman and Wilcoxon post hoc tests with Holm correction.
Support vector regression (SVR) demonstrated the best predictive performance on the independent test set, achieving an RMSE of 2.68 kg, an MAE of 1.81 kg, and an R² of 0.85.
SVR performance was statistically superior compared to XGBoost and CatBoost models based on post hoc Wilcoxon analysis.

Cite This Study

Valles-Coral et al. (2026) studied this question.

synapsesocial.com/papers/6aabb6445f706d05830e4a9d https://doi.org/10.3389/frai.2026.1876396

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark