arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.03501cs.CVcs.AIcs.RO
收藏

面向野生动物保护的基于移动平台的实时多线程波斑鸨检测与分割

Real-Time Threaded Houbara Detection and Segmentation for Wildlife Conservation using Mobile Platforms

Khalifa University Center for Autonomous Robotic Systems (KUCARS)(卡利法大学自主机器人系统中心);Khalifa University(卡利法大学);RENECO International Wildlife Consultants LLC(RENECO国际野生动物咨询公司)

Lyes Saad Saoud, Loic Lesobre, Enrico Sorato, Irfan Hussain

AI总结:

针对移动平台资源有限及隐蔽物种检测难的问题,提出集成TDM的两阶段深度学习框架,通过多线程并行YOLOv10检测与MobileSAM分割,在4万张标注波斑鸨数据集上实现高精度实时检测分割,并开源代码与数据集。

摘要:

自然环境下的实时动物检测与分割对野生动物保护至关重要,可通过远程摄像头流实现非侵入式监测。然而,由于计算资源有限且许多物种具有隐蔽外观,这些任务仍颇具挑战。我们提出一种面向移动设备优化的两阶段深度学习框架,该框架集成了线程检测模型(Threading Detection Model, TDM),以并行化基于YOLOv10的检测和基于MobileSAM的分割。与以往的YOLO+SAM流水线不同,我们的方法通过多线程降低延迟,提升了实时性能。YOLOv10负责检测,MobileSAM执行轻量级分割,二者并发运行以高效利用资源。针对保护优先级物种——具有隐蔽性的波斑鸨(Houbara Bustard),我们的模型取得了0.9627的mAP50、0.7731的mAP75、0.7178的mAP95,以及0.7421的MobileSAM mIoU。YOLOv10每帧运行耗时43.7毫秒,证实其具备实时可用性。我们构建了一个包含40000张标注图像的精选Houbara数据集,以支持在多样环境下的模型训练与评估。本研究使用的代码和数据集已在GitHub公开,地址为https://github.com/LyesSaadSaoud/mobile-houbara-detseg。交互式演示及更多资源可访问https://lyessaadsaoud.github.io/LyesSaadSaoud-Threaded-YOLO-SAM-Houbara。

英文摘要:

Real-time animal detection and segmentation in natural environments are vital for wildlife conservation, enabling non-invasive monitoring through remote camera streams. However, these tasks remain challenging due to limited computational resources and the cryptic appearance of many species. We propose a mobile-optimized two-stage deep learning framework that integrates a Threading Detection Model (TDM) to parallelize YOLOv10-based detection and MobileSAM-based segmentation. Unlike prior YOLO+SAM pipelines, our approach improves real-time performance by reducing latency through threading. YOLOv10 handles detection while MobileSAM performs lightweight segmentation, both executed concurrently for efficient resource use. On the cryptic Houbara Bustard, a conservation-priority species, our model achieves mAP50 of 0.9627, mAP75 of 0.7731, mAP95 of 0.7178, and a MobileSAM mIoU of 0.7421. YOLOv10 operates at 43.7 ms per frame, confirming real-time readiness. We introduce a curated Houbara dataset of 40,000 annotated images to support model training and evaluation across diverse conditions. The code and dataset used in this study are publicly available on GitHub at https://github.com/LyesSaadSaoud/mobile-houbara-detseg. For interactive demos and additional resources, visit https://lyessaadsaoud.github.io/LyesSaadSaoud-Threaded-YOLO-SAM-Houbara.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑