The detection of colony-forming units (CFUs) is a time-consuming but essential task in mulberry bacterial blight research. To overcome the problem of inaccurate small-target detection and high computational consumption in mulberry bacterial blight colony detection task, a mulberry bacterial blight colony dataset (MBCD) consisting of 310 images and 23,524 colonies is presented. Based on the MBCD, a colony detection model named Colony-YOLO is proposed. Firstly, the lightweight backbone network StarNet is employed, aiming to enhance feature extraction capabilities while reducing computational complexity. Next, C2f-MLCA is designed by embedding MLCA (Mixed Local Channel Attention) into the C2f module of YOLOv8 to integrate local and global feature information, thereby enhancing feature representation capabilities. Furthermore, the Shape-IoU loss function is implemented to prioritize geometric consistency between predicted and ground truth bounding boxes. Experiment results show that the Colony-YOLO achieved an mAP of 96.1% on MBCDs, which is 4.8% higher than the baseline YOLOv8n, with FLOPs and Params reduced by 1.8 G and 0.8 M, respectively. Comprehensive evaluations demonstrate that our method excels in detection accuracy while maintaining lower complexity, making it effective for colony detection in practical applications.

The accurate recognition of pig behaviors in intensive farming is crucial for health monitoring and growth assessment. To address multi-scale recognition challenges caused by perspective distortion (non-frontal camera angles), this study proposes MACA-Net, a YOLOv8n-based model capable of detecting four key behaviors: eating, lying on the belly, lying on the side, and standing. The model incorporates a Mamba Global–Local Extractor (MGLE) Module, which leverages Mamba to capture global dependencies while preserving local details through convolutional operations and channel shuffle, overcoming Mamba’s limitation in retaining fine-grained visual information. Additionally, an Adaptive Multi-Path Attention (AMPA) mechanism integrates spatial-channel attention to enhance feature focus, ensuring robust performance in complex environments and low-light conditions. To further improve detection, a Cross-Layer Feature Pyramid Transformer (CFPT) neck employs non-upsampled feature fusion, mitigating semantic gap issues where small target features are overshadowed by large target features during feature transmission. Experimental results demonstrate that MACA-Net achieves a precision of 83.1% and mAP of 85.1%, surpassing YOLOv8n by 8.9% and 4.4%, respectively. Furthermore, MACA-Net significantly reduces parameters by 48.4% and FLOPs by 39.5%. When evaluated in comparison to leading detectors such as RT-DETR, Faster R-CNN, and YOLOv11n, MACA-Net demonstrates a consistent level of both computational efficiency and accuracy. These findings provide a robust validation of the efficacy of MACA-Net for intelligent livestock management and welfare-driven breeding, offering a practical and efficient solution for modern pig farming.

In modern intensive pig farming, accurately monitoring piglet behavior is crucial for health management and improving production efficiency. However, the complexity of existing models demands high computational resources, limiting the application of piglet behavior recognition in farming environments. In this study, the piglet multi-behavior-recognition approach is divided into three stages. In the first stage, the LAMP pruning algorithm is used to prune and optimize redundant channels, resulting in the lightweight YOLOv8-Prune. In the second stage, based on YOLOv8, the AIFI module and the Gather–Distribute mechanism are incorporated, resulting in YOLOv8-GDA. In the third stage, using YOLOv8-GDA as the teacher model and YOLOv8-Prune as the student model, knowledge distillation is employed to further enhance detection accuracy, thus obtaining the YOLOv8-Piglet model, which strikes a balance between the detection accuracy and speed. Compared to the baseline model, YOLOv8-Piglet significantly reduces model complexity while improving detection performance, with a 6.3% increase in precision, 11.2% increase in recall, and an mAP@0.5 of 91.8%. The model was deployed on the NVIDIA Jetson Orin NX edge computing platform for the evaluation. The average inference time was reduced from 353.9 ms to 163.2 ms, resulting in a 53.8% reduction in the processing time. This study achieves a balance between model compression and recognition accuracy through the collaborative optimization of pruning and knowledge extraction.