Achieving accurate cattle detection in complex barn environments is a critical technical challenge for smart livestock farming. Cattle exhibit highly similar appearances, severe occlusion, and significant multi-scale variations, making it difficult for existing detection methods to balance accuracy with model efficiency. This paper proposes a lightweight cattle detection model, DMFRNet, with YOLO11 as the baseline. To address these challenges, DMFRNet introduces three targeted improvements. First, C3K2-DIMB is designed to enhance multi-scale feature extraction by adaptively reweighting multi-branch depthwise convolution features, thereby improving the representation of cattle with different body sizes, poses, and viewing distances. Second, SimAM is embedded after the SPPF layer to refine high-level semantic features without introducing additional parameters, which improves feature discrimination under occlusion, low contrast, and complex backgrounds. Third, LSCDH replaces the original decoupled detection head to reduce parameter redundancy through cross-scale shared convolution while preserving multi-scale prediction capability. These designs jointly address the key challenges of multi-scale cattle appearance, occlusion, and lightweight model construction in complex barn scenes. Experimental results on the combined CBVD-5 and Dairy Cow dataset demonstrate that DMFRNet achieves a Precision of 93.49%, F1 of 89.84%, mAP50 of 93.85%, and mAP50–95 of 61.74%, with only 2.16 M parameters, 5.10 GFLOPs, and a model size of 4.5 MB. Comparative experiments demonstrate that DMFRNet provides a favorable accuracy–efficiency trade-off for lightweight dairy cow detection.

Reliable cow detection in barn environments remains challenging due to visual occlusion,overlapping animals,and cluttered backgrounds.To improve accuracy and real-time performance under such conditions,this study presents YOLO11-DML,an enhanced detection framework built upon YOLO11.The proposed model introduces targeted improvements in feature extraction,attention design,and lightweight head architecture.First,an upgraded C3K2-DIMB module replaces the original C3K2 block,using dynamically weighted convolution branches to strengthen multi-scale representation.Second,a Mixed Local Channel Attention (MLCA) mechanism is added to the end of the backbone,improving robustness against illumination variation and partial occlusion.Finally,a Lightweight Shared Convolution Detection Head (LSCDH) is designed to reduce parameter redundancy and computation costs while maintaining detection accuracy.Experiments on the CBVD-5 and Dairy Cow datasets show that YOLO11-DML achieves stable improvements,reaching a precision of 92.57% and an F1-score of 89.07%,with only 2.16M parameters,5.1 GFLOPs,and a model size of 4.5 MB.These results indicate that the model provides an efficient and lightweight solution for real-time multi-object cow monitoring in intelligent dairy production.