改进YOLOv8n的群养生猪目标检测算法研究  

Research on group pig target detection with improve YOLOv8n algorithm

作　　者：钟长华 宋弘 吴浩 江俊卓 ZHONG Changhua;SONG Hong;WU Hao;JIANG Junzhuo(Sichuan University of Science and Engineering,School of Automation and Information Engineering,Yibing Sichuan 643000,China;Key Laboratory of Artificial Intelligence in Sichuan Province,Sichuan University of Science and Engineering,Yibing Sichuan 643000,China;Aba Teachers University,Aba Sichuan 623002,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000 [2]人工智能四川省重点实验室,四川宜宾644000 [3]阿坝师范学院,四川阿坝623002

出　　处：《河北农业大学学报》2024年第6期89-96,共8页Journal of Hebei Agricultural University

基　　金：四川省科技厅项目(2022YFS0518,2022ZHCG0035);人工智能四川省重点实验室项目(2023RYY06);四川轻化工大学人才引进项目(2021RC12);四川轻化工大学研究生创新基金项目(Y2023296).

摘　　要：为满足嵌入式设备对模型轻量化和高精度检测的需求,本研究提出了一种基于YOLOv8n模型改进的群养生猪目标检测算法。首先在主干网络中引入C2fFB结构,减少对内存的访问量与冗余计算;然后以BiFPN结构构成新的Neck网络并引入C2fSC模块,进一步实现更深层次的特征融合,减少融合的空间冗余和通道冗余;最后采用SIoU替换原来的CIoU,提高了模型的精度。实验结果表明,本算法的F1分数、查准率、查全率、平均精确度相比改进之前分别提升了3%、1.8%、3.5%、1.5%,参数量、计算量、模型大小相比改进之前分别下降了46.84%、27.16%、44.71%。因此,本算法模型为群养生猪的智慧养殖提供了1种高效的目标检测解决方案。In order to meet the needs of for model lightweight and high-precision detection for embedded devices,this paper proposed a group-raising pig target detection algorithm based on the improved YOLOv8n model.First,the C2fFB structure was introduced in the backbone network to reduce the amount of memory access and redundant calculations.Then the new Neck network was constructed with the BiFPN structure and the C2fSC module was introduced to further achieve deeper feature fusion and reduce of the spatial redundancy and channel redundancy of the fusion.Finally,SIoU was used to replace the original CIoU to improve the accuracy of the model.The experimental results showed that the F1 score,precision,recall rate,and average precision of the proposed algorithm were improved compared with the original algorithm by 3%,1.8%,3.5%,and 1.5%,respectively.And the number of parameters,calculation amount,and model size were reduced by 46.84%,27.16%,and 44.71%respectively.Therefore,the algorithm model in this paper provides an efficient target detection solution for the intelligent breeding of group-raising pigs.

关 键 词：目标检测 YOLOv8n 群养生猪 BiFPN 交并比 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]