Image-Based Weight Estimation of Sheep Using Multitask Convolutional Neural Networks

• Anthony Irlan Marques Luz UFPI
• Julio Vítor Monteiro Marques UFPI
• Rafael Luz Araújo UFPI
• Romuere Rodrigues Veloso e Silva UFPI

DOI: https://doi.org/10.5753/eniac.2026.29513

Resumo

This study evaluates a multitask convolutional neural network for joint sheep segmentation and weight estimation from outdoor RGB images. The dataset contains 507 annotated frames from 42 animals acquired with a Kinect V2 sensor and evaluated using animal-level five-fold cross-validation. U-Net, U-Net++, and FPN were compared. Non-augmented U-Net obtained the best segmentation performance, with a Dice score of 0.93 ± 0.01 and an MAE of 10.70 ± 5.00 kg. Augmented U-Net++ achieved the lowest MAE (10.15 ± 3.91 kg), but lower segmentation performance. Given the limited sample and error magnitude, the results indicate that the proposed approach is feasible for image-based weight estimation under the experimental conditions considered.

Referências

Agliari, E., Alessandrelli, A., Barra, A., and Ricci-Tersenghi, F. (2023). Parallel learning by multitasking neural networks. Journal of Statistical Mechanics: Theory and Experiment, 2023(11):113401.

Banda, L. J. and Tanganyika, J. (2021). Livestock provide more than food in smallholder production systems of developing countries. Animal Frontiers, 11(2):7–14.

Buslaev, A., Iglovikov, V. I., Khvedchenya, E., Parinov, A., Druzhinin, M., and Kalinin, A. A. (2020). Albumentations: Fast and flexible image augmentations. Information, 11(2).

Camacho-Pérez, E., Chay-Canul, A. J., Garcia-Guendulain, J. M., and Rodríguez-Abreo, O. (2022). Towards the estimation of body weight in sheep using metaheuristic algorithms from biometric parameters in microsystems. Micromachines, 13(8):1325.

Conceição, A. R., Braga, P. H. A., and Schultz, E. B. (2023). Tecnologias aplicadas ao monitoramento de parâmetros fisiológicos na produção de ruminantes. Revista Agrária Acadêmica, 6(2):27–37.

Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition, pages 248–255.

Detlefsen, N. S., Borovec, J., Schock, J., Harsh, A., Koker, T., Liello, L. D., Stancl, D., Quan, C., Grechkin, M., and Falcon, W. (2022). Torchmetrics - measuring reproducibility in pytorch. Journal of Open Source Software, 7(70):4101.

EMBRAPA (2023). Ciência abre novas oportunidades na produção de caprinos e ovinos. Accessed: 2025-01-17.

Evan-Amos (2014). Xbox-one-kinect. Wikimedia Commons. Public domain.

He, C., Qiao, Y., Mao, R., Li, M., and Wang, M. (2023). Enhanced litehrnet based sheep weight estimation using rgb-d images. Computers and Electronics in Agriculture, 206:107667.

Iakubovskii, P. (2019). Segmentation models pytorch. [link].

Kashiha, M., Bahr, C., Ott, S., Moons, C. P., Niewold, T. A., Ödberg, F. O., and Berckmans, D. (2014). Automatic weight estimation of individual pigs using image analysis. Computers and Electronics in Agriculture, 107:38–44.

Kato, S. and Hotta, K. (2021). Mse loss with outlying label for imbalanced classification. arXiv preprint arXiv:2107.02393.

Kirillov, A., He, K., Girshick, R., and Dollár, P. (2017). A unified architecture for instance and semantic segmentation. In Computer Vision and Pattern Recognition Conference. CVPR.

Menesatti, P., Costa, C., Antonucci, F., Steri, R., Pallottino, F., and Catillo, G. (2014). A low-cost stereovision system to estimate size and weight of live sheep. Computers and Electronics in Agriculture, 103:33–38.

Ojala, M. and Garriga, G. C. (2010). Permutation tests for studying classifier performance. Journal of machine learning research, 11(6).

Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019). Pytorch: An imperative style, high performance deep learning library. Advances in neural information processing systems, 32.

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830.

Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation.

Samperio, E., Lidón, I., Rebollar, R., Castejón-Limas, M., and Álvarez-Aparicio, C. (2021). Lambs’ live weight estimation using 3d images. Animal, 15(5):100212.

Sant’Ana, D. A., Pache, M. C. B., Martins, J., Soares, W. P., de Melo, S. L. N., Garcia, V., de Moares Weber, V. A., da Silva Heimbach, N., Mateus, R. G., and Pistori, H. (2021). Weighing live sheep using computer vision techniques and regression machine learning. Machine Learning with Applications, 5:100076.

Sarbolandi, H., Lefloch, D., and Kolb, A. (2015). Kinect range sensing: Structured-light versus time-of-flight kinect. Computer Vision and Image Understanding, 139:1–20.

Weber, V. A. M., de Lima Weber, F., da Silva Oliveira, A., Astolfi, G., Menezes, G. V., de Andrade Porto, J. V., Rezende, F. P. C., de Moraes, P. H., Matsubara, E. T., Mateus, R. G., de Araújo, T. L. A. C., da Silva, L. O. C., de Queiroz, E. Q. A., de Abreu, U. G. P., da Costa Gomes, R., and Pistori, H. (2020). Cattle weight estimation using active contour models and regression trees bagging. Computers and Electronics in Agriculture, 179:105804.

Zhao, R., Qian, B., Zhang, X., Li, Y., Wei, R., Liu, Y., and Pan, Y. (2020). Rethinking dice loss for medical image segmentation. In 2020 IEEE International Conference on Data Mining (ICDM), pages 851–860. IEEE.

Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., and Liang, J. (2018). Unet++: A nested u-net architecture for medical image segmentation.