发表机构
Tribhuvan University; Fort Valley State University（特里布文大学; 山谷堡州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对现有牛鼻纹识别闭集假设的局限，提出开放集牛鼻纹识别任务，构建泄漏控制评估协议，对比混合CNN-ViT模型与MegaDescriptor-L模型，验证了阈值校准等对开放集识别的关键作用。
AI 中文摘要
可靠的个体牛识别支持疾病监测、疫苗接种记录、繁殖管理和牲畜保险。尽管牛鼻纹是一种稳定的非接触式生物特征，但现有的鼻纹识别系统大多假设已登记动物为闭集，限制了其实际部署。我们将牛鼻纹生物特征重新定义为开放集、基于图库的识别问题，该问题可对未见过的动物进行弃权（不执行），并支持无需模型重新训练的增量登记。我们引入了一种基于身份不相交拆分、每折重新训练、保留阈值校准、验证重复移除和自举置信区间的泄漏控制评估协议。我们使用两种对比嵌入配置评估该框架：混合CNN-ViT度量学习模型和MegaDescriptor-L基础模型。在最优阈值选择下，当目标误接受率为10^(-1)、10^(-2)、10^(-3)时，混合模型的检测与识别率分别达到98.3%、96.4%、93.6%，而MegaDescriptor-L则达到99.3%、98.1%、96.1%。然而，可部署的阈值校准显示最优性能与校准性能之间存在显著差异：在1%的目标下，混合模型的误接受率为1.03%，而MegaDescriptor-L达到2.44%。增量登记在单张参考图像下Rank-1准确率超过91%，在八张参考图像下达到97.3%，且无需重新训练模型或降低现有图库性能。这些结果表明，阈值校准、泄漏控制和嵌入质量对于可靠的开放集牛识别至关重要，并为面向部署的动物生物特征系统提供了实用评估框架。
英文摘要
Reliable individual cattle identification supports disease surveillance, vaccination records, breeding management, and livestock insurance. Although the bovine muzzle provides a stable, non-contact biometric, existing muzzle-recognition systems largely assume a closed set of enrolled animals, limiting their practical deployment. We reformulate cattle muzzle biometrics as an open-set, gallery-based identification problem that can reject previously unseen animals and support incremental enrollment without model retraining. We introduce a leakage-controlled evaluation protocol based on identity-disjoint splits, per-fold retraining, held-out threshold calibration, verified duplicate removal, and bootstrap confidence intervals. We evaluate the framework using two contrasting embedding configurations: a hybrid CNN-ViT metric-learning model and the MegaDescriptor-L foundation model. Under oracle threshold selection, the hybrid model achieves detection-and-identification rates of 98.3%, 96.4%, and 93.6% at target false-acceptance rates of 10^(-1), 10^(-2), and 10^(-3), respectively, while MegaDescriptor-L achieves 99.3%, 98.1%, and 96.1%. However, deployable threshold calibration reveals a substantial difference between oracle and calibrated performance: the hybrid model achieves a false-acceptance rate of 1.03% at a 1% target, whereas MegaDescriptor-L reaches 2.44%. Incremental enrollment further achieves Rank-1 accuracy above 91% with a single reference image and up to 97.3% with eight reference images, without retraining the model or degrading the existing gallery. These results demonstrate that threshold calibration, leakage control, and embedding quality are critical for reliable open-set cattle identification and provide a practical evaluation framework for deployment-oriented animal biometric systems.
