家畜生态学报2025，Vol.46Issue(10)：108-113,6.DOI:10.3969/j.issn.1673-1182.2025.10.016
基于改进YOLO v8的哺乳仔猪保温区温度适宜状态识别研究
Identifying the Optimal Temperature Status in Heat Preservation Area for Lactating Piglets Based on Improved YOLO v8
摘要
Abstract
Traditional methods for regulating the appropriate temperature during piglets' growth by monitoring the temperature in heating zones often face challenges,such as determining the optimal placement of temperature sen-sors and the issue of heating when no piglets are present.These methods fail to meet the growth temperature needs of piglets and result in unnecessary energy consumption.This study addresses these issues by defining five typical aggre-gation and dispersion states of suckling piglet groups in the heating zone,based on variations in environmental temper-ature.The study employs an improved YOLO v8 detection algorithm,utilizing optimized VanillaNet architecture,the Sim AM attention mechanism,and the Lion optimizer.The goal is to use machine vision to assess the optimal temper-ature conditions in the piglet heating zone and enable intelligent regulation of the heating system.The results show that the improved YOLO v8 maintains high accuracy,with a memory usage of 0.7MB and a recognition time of 1.495 ms/f,outperforming both YOLO v5 and the original YOLO v8 in terms of model size and recognition time.In com-plex farm environments,the average state recognition rate exceeds 97％,demonstrating high detection confidence.This research provides a critical basis for adjusting heating devices in piglet thermal systems which offer an optimal thermal environment for the piglets while effectively reducing energy consumption.关键词
哺乳仔猪/适宜生长环境温度/图像识别/聚散状态/改进YOLO v8Key words
lactating piglets/appropriate growth environment temperature/image recognition/the sta-tus of aggregation and dispersion/improved YOLO v8分类
农业科技引用本文复制引用
付明明,郑萍,侯郁硕,曹月,李抒憧,黎煊..基于改进YOLO v8的哺乳仔猪保温区温度适宜状态识别研究[J].家畜生态学报,2025,46(10):108-113,6.基金项目
农业农村部智慧养殖技术重点实验室开放课题(KLSFTAA-KF001,KLSFTAA-KF002) （KLSFTAA-KF001,KLSFTAA-KF002）
黑龙江省自然科学基金(LH2023C017) （LH2023C017）
黑龙江省教育厅新一轮黑龙江省"双一流"学科协同创新成果项目(LJGXCG2023-062,LJGXCG2024-F14) （LJGXCG2023-062,LJGXCG2024-F14）
