用新流程实现奶牛密集队列中精准识别，准确率超98%。
Automated Re-Identification of Holstein-Friesian Cattle in Dense Crowds
- 先检测、再分割、后识别，结合开源模型提升复杂场景鲁棒性。
- 在密集群组中达98.93%识别准确率，较基线提升超27%。
- 适合农业自动化监控，无需人工干预，代码数据可复现。
当荷斯坦牛个体空间分离时，现有检测与重识别（Re-ID）方法表现良好；但在密集群体中，基于YOLO的物种检测等方法失效，尤其对轮廓断裂的毛色图案物种。为提升有效性和可迁移性，本文提出一种新的「检测-分割-识别」流程，利用开放词汇无权重定位与通用图像分割模型作为预处理，结合重识别网络。为评估该方法，我们发布了一个在实际奶牛场拍摄的九天闭路电视数据集。所提方法克服了密集群组中的检测失效问题，实现98.93%的识别准确率，显著优于基于定向边界框的方法（提升47.52%）及SAM物种检测基线（提升27.13%）。此外，无监督对比学习进一步将重识别准确率提升至94.82%。结果表明，在无需人工干预的农场环境中，密集场景下的重识别既可行又可靠。代码与数据集已公开以支持复现。
原文摘要 · Abstract (English)
Holstein-Friesian detection and re-identification (Re-ID) methods capture individuals well when targets are spatially separate. However, existing approaches, including YOLO-based species detection, break down when cows group closely together. This is particularly prevalent for species which have outline-breaking coat patterns. To boost both effectiveness and transferability in this setting, we propose a new detect-segment-identify pipeline that leverages the Open-Vocabulary Weight-free Localisation and the Segment Anything models as pre-processing stages alongside Re-ID networks. To evaluate our approach, we publish a collection of nine days CCTV data filmed on a working dairy farm. Our methodology overcomes detection breakdown in dense animal groupings, resulting in a 98.93% accuracy. This significantly outperforms current oriented bounding box-driven, as well as SAM species detection baselines with accuracy improvements of 47.52% and 27.13%, respectively. We show that unsupervised contrastive learning can build on this to yield 94.82% Re-ID accuracy on our test data. Our work demonstrates that Re-ID in crowded scenarios is both practical as well as reliable in working farm settings with no manual intervention. Code and dataset are provided for reproducibility.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。