With the rapid advancement of artificial intelligence, intelligent farming has become a key trend in modern agriculture. In particular, the application of intelligent systems in broiler farming is essential for enhancing production efficiency and optimizing management practices. Broiler farming is a complex process involving multiple interrelated components. However, existing knowledge graphs primarily focus on disease and prevention, making it difficult to capture the intricate interdependencies within the farming process. This limits the effectiveness of knowledge-based support in decision-making. To develop a high-quality broiler farming knowledge system, this study adopts large language modeling technology to integrate a Chinese corpus and construct a comprehensive knowledge graph dataset covering four core dimensions: broiler breeds, farming environment, feeding management, and disease prevention.

Facial surface temperature provides useful non-contact information for monitoring dairy-cow health, welfare, heat stress, and reproductive status. Infrared thermography can capture thermal information from facial regions such as the eyes, muzzle, nostrils, and ears; however, infrared images often contain weak texture and indistinct anatomical boundaries, which can hinder reliable cow-head localization during mobile robotic inspection. Visible-light images provide richer structural information but do not contain temperature data. This study developed YOLO11-AFE, a lightweight visible–thermal cow-head detection and heterogeneous-camera association method for quadruped inspection robots. The detector incorporates ADown for lightweight downsampling, C3k2_FE for adaptive feature enhancement, and SPPF_ECA for channel-aware multi-scale representation. An improved Hungarian matching algorithm was subsequently used to establish one-to-one correspondences between cow-head detections in synchronized visible-light and infrared pseudo-colour images. Across three independent runs, YOLO11-AFE achieved precision, mAP@0.5, and mAP@0.5:0.95 values of 97.59 ± 0.20%, 96.37 ± 0.24%, and 70.76 ± 0.51%, respectively, on the visible-light test subset, and 96.11 ± 0.24%, 99.07 ± 0.10%, and 91.50 ± 0.40%, respectively, on the infrared subset. The model required 2.14 million parameters and 5.27 GFLOPs, representing reductions of 17.37% and 18.17%, respectively, relative to YOLO11n. The association method achieved an overall accuracy of 98.11% across 371 ground-truth cow-head pairs in the combined validation and test evaluation. TensorRT FP16 deployment on the Jetson Orin NX achieved 36.71 FPS for the complete core processing pipeline. These results demonstrate that YOLO11-AFE provides an accurate and computationally efficient perception front end for future non-contact facial-temperature monitoring using mobile inspection robots.

Facial surface temperature can provide useful non-contact information for dairy-cow health, welfare, heat-stress, and reproductive-status monitoring. Infrared thermography can capture thermal information from facial regions such as the eyes, muzzle, nostrils, and ears, but infrared images alone often lack clear texture and anatomical boundaries, making reliable cow-head localisation difficult during mobile robotic inspection. Visible-light images provide richer structural cues but no temperature information. This study proposes YOLO11-AFE, a lightweight visible–thermal cow-head detection and heterogeneous-camera association framework for quadruped inspection robots. ADown was introduced for lightweight downsampling, C3k2_FE for adaptive feature enhancement, and SPPF_ECA for channel-aware multi-scale representation. An improved Hungarian matching algorithm was then used to establish one-to-one cow-head correspondence between synchronized visible-light and infrared pseudo-colour detections. YOLO11-AFE achieved Precision, mAP@0.5, and mAP@0.5:0.95 values of 97.51%, 96.26%, and 70.64% on the visible-light test subset, and 95.92%, 99.07%, and 91.41% on the infrared subset. Compared with YOLO11n, parameters and FLOPs were reduced by 17.37% and 18.17%. TensorRT deployment on Jetson Orin NX reached 36.71 FPS for the complete core pipeline. All 187 cow-head pairs were correctly associated, indicating that the framework provides an edge-deployable front end for future non-contact facial temperature monitoring.