基于深度学习的羊面部身份识别研究进展
Research Progress on Sheep Facial Recognition Based on Deep Learning Method
王光普 1宣传忠 2刘苏慧 1唐朝辉 2苏赫2
作者信息
- 1. 内蒙古农业大学机电工程学院,呼和浩特 010018
- 2. 内蒙古农业大学机电工程学院,呼和浩特 010018;内蒙古自治区草业与养殖业智能装备工程技术研究中心,呼和浩特 010018
- 折叠
摘要
随着人工智能技术的发展,基于深度学习技术的羊面部身份识别已得到深入研究,其在养殖场的应用可提高养羊业的智能化与精准化水平.该研究归纳了目前羊只身份识别的传统方法和生物特征方法,分析了利用羊脸生物特征实现羊只身份识别的潜力,通过构建不同的深度学习网络模型实现羊脸身份识别,探讨了识别模型精度和稳定性提高的机理.论述了羊脸身份识别模型的轻量化和迁移学习研究,旨在增强识别模型在便携式设备上的部署能力,梳理了羊脸身份识别技术在羊只表情、品种识别等领域的拓展应用.在此基础上展望了羊面部身份识别技术的发展,需要进一步完善规模化的羊脸数据库,设计多视角的羊脸图像采集系统,研发先进的羊脸身份识别模型,解决跨生长周期的羊脸身份识别问题.研究可为规模化养殖场的羊脸身份识别研究提供理论参考.
Abstract
The development of artificial intelligence technology has significantly propelled the field of sheep facial identity recogni-tion,particularly through the application of deep learning techniques.This innovation holds the potential to transform the sheep in-dustry by enhancing the intelligence and precision of breeding operations.This study provided a comprehensive overview of the tradi-tional and biometric methods for sheep identity recognition.It also analyzed the potential of using sheep face biometric features to achieve sheep identity recognition,constructing different deep learning network models to achieve sheep face identity recognition,and exploring the mechanism of improving the accuracy and stability of recognition models.This paper also presented a comprehen-sive study on the lightweighting and transfer learning of sheep face identity recognition models,so as to enhance the deployment capa-bility of recognition models on portable devices.Additionally,the paper outlined the expanded applications of sheep face identity recognition technology in fields such as sheep expression and breed recognition.On this basis,the development of sheep facial iden-tity recognition technology was discussed.Building upon this foundation,the paper anticipated future developments in sheep facial identity recognition technology.It emphasized the necessity of establishing a comprehensive sheep facial database,designing a multi-perspective sheep facial image acquisition system,and developing advanced sheep facial identity recognition models to address the challenges associated with cross-growth cycle sheep facial identity recognition.The research provided a theoretical reference for the study of sheep face identification in large-scale breeding farms.
关键词
深度学习/身份识别/规模化养殖/精准畜牧业/羊脸识别Key words
Deep learning/Identity recognition/Large-scale farming/Precision livestock farming/Sheep face recognition引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据