农业工程2025，Vol.15Issue(9)：41-48,8.DOI:10.19998/j.cnki.2095-1795.202509307
轻量级多目标牛行为识别模型EVH-YOLO11构建
Construction of lightweight multi-objective cattle behavior recognition model EVH-YOLO11
陈敏 1任瑞仙 1张志东 1李浩溥2
作者信息
- 1. 山西工程技术学院大数据与智能工程系,山西 阳泉 045000
- 2. 山西农业大学农业工程学院,山西 晋中 030801
- 折叠
摘要
Abstract
Behavior recognition is of great significance in cattle production and health management.Currently,there are some prob-lems in cattle behavior recognition,such as high complexity of model and low accuracy in multi-objective recognition.To solve this problem,a lightweight multi-objective cattle behavior recognition model EVH-YOLO11 was proposed.In order to solve complex inter-ference problems such as occlusion,overlap and small targets in actual cattle farm environment,a sandwich layout design was intro-duced for EfficientViT feature extraction module to reduce calculation redundancy.At same time,combined with Dynamic Head detec-tion head,objective detection performance was adaptively enhanced.Experiments showed that EVH-YOLO11 outperformed main-stream models,which could provide technical support for smart cattle farms.关键词
目标检测/YOLO11/EfficientViT/轻量化模型/智慧牛场/行为识别Key words
object detection/YOLO11/EfficientViT/lightweight model/smart cow farm/behavior recognition分类
农业科技引用本文复制引用
陈敏,任瑞仙,张志东,李浩溥..轻量级多目标牛行为识别模型EVH-YOLO11构建[J].农业工程,2025,15(9):41-48,8.
