The MCAA-YOLO + XPBI integrated model provides a hybrid intelligent measurement method for predicting the body weight of Hu sheep

In modern animal husbandry, accurately obtaining sheep body weight plays a crucial role in improving feeding management and optimizing economic benefits. Conventional manual measurement methods are inefficient and easily influenced by external factors such as stress responses, resulting in unstable outcomes that fail to meet the practical demands of intelligent farming. Therefore, this study proposes an automated method that estimates body weight from body-size features based on an improved YOLOv10 and an XPBI model, offering an efficient, accurate, and non-contact solution. In the feature extraction stage, a Multi-Component Attention Aggregation (MCAA) module is introduced into YOLOv10 to integrate the advantages of GAM, CBAM, CoordAtt, and ECA, thereby enhancing feature representation. The resulting MCAA-YOLOv10 model achieves notable performance improvements, with Precision, Recall, mAP50, and mAP50-95 reaching 0.995, 0.986, 0.991, and 0.987, respectively, ensuring accurate and stable acquisition of body-size data. Based on these features, an integrated prediction model, the XGBoost-PCA-Bagging Integrator (XPBI), is constructed, employing principal component analysis and guided bagging to improve dimensionality reduction and parameter selection. In the weight-prediction task, the XPBI model demonstrates excellent performance, achieving an MAE of 1.121, an RMSE of 1.490, a MAPE of 4.47%, and an R ² of 0.918, while requiring only 0.106 ms for single-sample inference, significantly outperforming comparison models. The results indicate that predicting Hu sheep body weight from body-size parameters enables efficient and accurate estimation, confirming the feasibility and effectiveness of this approach in practical applications. This method provides a reliable technological pathway for the digital and intelligent transformation of animal husbandry, enhances the scientific level of livestock management, and offers considerable social and economic value.