Open-set Goat Face Recognition with MobileFaceNet Adaptation for Livestock Farming
Keywords:
Black Goat, Lightweight Model, High Similarity RecognitionAbstract
In the context of intelligent livestock farming, precise individual animal identification has become a critical requirement. To address the challenges posed by high facial similarity among black goats and the need for model retraining in incremental learning scenarios, this paper introduces GoatFaceNet, a lightweight goat face recognition model based on MobileFaceNet. GoatFaceNet incorporates MixConv to enhance its feature extraction capability. Furthermore, the CurricularFace loss function is employed to improve inter-class separability and intra-class compactness, thereby increasing the model's robustness in open-set recognition tasks. A comprehensive goat face dataset was constructed to support the evaluation. Experimental results show that GoatFaceNet achieves an accuracy of 94.2% on the test set. Additional evaluations involving 3,200 goat face pairs confirm the model's superior open-set discrimination performance, validating its practical applicability and deployment potential in real-world farming environments.
References
Ait-Saidi A, Caja G, Salama AAK, Carné S. Implementing electronic identification for performance recording in sheep: I. Manual versus semiautomatic and automatic recording systems in dairy and meat farms. J Dairy Sci. 2014;97:7505-7514.
Corkery GP, Gonzales-Barron UA, Butler F, et al. A preliminary investigation on face recognition as a biometric identifier of sheep. Trans ASABE. 2007;50(1):313-320.
Taigman Y, Yang M, Ranzato M, Wolf L. DeepFace: Closing the gap to human-level performance in face verification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2014; Columbus, OH, USA. p. 1701-1708. doi:10.1109/CVPR.2014.220.
Howard AG. MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861. 2017.
Sandler M, Howard A, Zhu M, et al. MobileNetV2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2018. p. 4510-4520.
Zhang X, Zhou X, Lin M, et al. ShuffleNet: An extremely efficient convolutional neural network for mobile devices. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2018. p. 6848-6856.
Ma N, Zhang X, Zheng HT, et al. ShuffleNet V2: Practical guidelines for efficient CNN architecture design. In: Proceedings of the European Conference on Computer Vision (ECCV); 2018. p. 116-131.
Chen S, Liu Y, Gao X, et al. MobileFaceNets: Efficient CNNs for accurate real-time face verification on mobile devices. In: Chinese Conference on Biometric Recognition (CCBR). Cham: Springer; 2018. p. 428-438.
Martindez-Diaz Y, Luevano LS, Mendez-Vazquez H, et al. ShuffleFaceNet: A lightweight face architecture for efficient and highly accurate face recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops; 2019.
Deng J, Guo J, Xue N, et al. ArcFace: Additive angular margin loss for deep face recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR); 2019. p. 4690-4699.
Zhang H.-M., Zhou L.-X., Li Y.-H., et al. Sheep Face Recognition Method Based on Improved MobileFaceNet. Transactions of the Chinese Society for Agricultural Machinery. 2022;53(05):267-274. (in Chinese)
Xu B, Wang W, Guo L, et al. CattleFaceNet: A cattle face identification approach based on RetinaFace and ArcFace loss. Computers and Electronics in Agriculture. 2022;193:106675.
Billah M, Wang X, Yu J, et al. Real-time goat face recognition using convolutional neural network. Computers and Electronics in Agriculture. 2022;194:106730.
Zhang X, Xuan C, Ma Y, et al. An efficient method for multi-view sheep face recognition. Eng Appl Artif Intell. 2024;134:108697.
Zhang X, Xuan C, Ma Y, et al. High-similarity sheep face recognition method based on a Siamese network with fewer training samples. Comput Electron Agric. 2024;225:109295.
Li X, Zhang Y, Li S. Rethinking lightweight sheep face recognition via network latency-accuracy tradeoff. Comput Electron Agric. 2024;227:109662.
Zhang X, Xuan C, Ma Y, et al. Lightweight model-based sheep face recognition via face image recording channel. J Anim Sci. 2024;102:skae066.
Tan M, Le QV. MixConv: Mixed depthwise convolutional kernels. arXiv preprint arXiv:1907.09595. 2019.
Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2016. p. 2818-2826.
Zhang X, Wu X, Xiao Z, et al. Mixed-decomposed convolutional network: A lightweight yet efficient convolutional neural network for ocular disease recognition. CAAI Trans Intell Technol. 2024;9(2):319-332.
Boutros F, Damer N, Fang M, et al. MixFaceNets: Extremely efficient face recognition networks. In: 2021 IEEE International Joint Conference on Biometrics (IJCB); 2021. p. 1-8.
Ramachandran P, Zoph B, Le QV. Searching for activation functions. arXiv:1710.05941. 2017.
Howard A, Sandler M, Chen B, et al. Searching for MobileNetV3. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV); 2019; Seoul, Republic of Korea. p. 1314-1324.
Wang X, Zhang S, Wang S, et al. Mis-classified vector guided softmax loss for face recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2020;34(7):12241-12248.
Huang Y, Wang Y, Tai Y, et al. CurricularFace: Adaptive curriculum learning loss for deep face recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; 2020. p. 5901-5910.
Yan M, Zhao M, Xu Z, et al. VarGFaceNet: An efficient variable group convolutional neural network for lightweight face recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops; 2019.
Mehta S, Rastegari M. MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer. arXiv preprint arXiv:2110.02178. 2021.
He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2016. p. 770-778.
