PB-DETR：群养生猪多行为识别的自适应液体神经网络编码器与动态监督框架
提出PB-DETR框架，实现群养生猪多行为识别，mAP达0.893。
阅读 4
关键词 · 相关资讯与研究
#生猪行为识别
沿着“生猪行为识别”继续探索，共 1 条内容。
你可能还想看
没有这个标签，内容接近Lightweight two-stage framework for multi-cow behavior recognition based on edge devices
Lightweight two-stage framework for multi-cow behavior recognition based on edge devices。Computers and Electronics in Agriculture
关注点
该框架将多牛行为识别拆为两阶段并面向边缘设备轻量化部署，关注点在于其精度与算力开销的权衡及实际牛舍场景的泛化能力。
阅读 1
Advancing pig lameness detection with a multi-task framework: Leveraging multi-view 3D pose estimation and wavelet convolution（多任务框架推进生猪跛行检测：利用多视角3D姿态估计与小波卷积）
提出PoseGait-MT多任务框架，用多视角3D姿态与小波卷积同时检测生猪跛行程度和患肢。
阅读 5
Identifying mounting behaviour in boars using deep learning-based instance segmentation and binary classification - a pilot study
提出基于Mask R-CNN和SVM的猪爬跨行为自动检测方法，实现群养公猪连续监测。
阅读 3
Weight-posture joint modeling for body weight estimation of unconstrained pigs
Weight-posture joint modeling for body weight estimation of unconstrained pigs。Smart Agricultural Technology
关注点
该研究针对无约束养殖场景，将体重与姿态联合建模以提升猪只估重精度，适用于非接触式养殖监测，但实际精度与数据规模需看原文验证。
阅读 0
Automated analysis of feline posture using deep learning and geometric modeling
提出一种结合深度学习和几何建模的猫姿态自动分析框架，从RGB图像提取临床可解释的姿态描述符。
农业遥感与作物表型智慧农业
阅读 3
Lightweight AI tracks every cow in the herd in real time
提出轻量化边缘AI模型，实时追踪牛群个体行为。
关注点
该模型在商业牛场以单帧24.5毫秒、40帧每秒连续运行，验证了低成本边缘部署的可行性；但结论限于视觉行为识别，多模态融合与长期稳定性尚待后续验证。
阅读 0
Non-contact backfat thickness prediction in finishing pigs via multimodal feature fusion for pig sorting systems
提出多模态特征融合方法，非接触预测育肥猪背膘厚度，用于猪只分拣系统。
阅读 3
From behavioural monitoring to decision support in aquaculture: an interpretable analytical framework
From behavioural monitoring to decision support in aquaculture: an interpretable analytical framework。Biosystems Engineering
阅读 3