[论文解读] Automatic individual pig detection and tracking in surveillance videos

Lei Zhang, Helen Gray+3 more|arXiv (Cornell University)|Dec 12, 2018

Video Surveillance and Tracking Methods参考文献 29被引用 13

一句话总结

该论文提出了一种实时、端到端的方法，用于在光照变化、遮挡和外观相似等挑战性条件下，自动检测和跟踪监控视频中的个体猪只。该方法通过分层数据关联算法，将多尺度CNN检测器与相关滤波跟踪器相结合，实现了具有在线检测优化和漂移校正能力的鲁棒跟踪，在无需人工标记的真实农场数据集上实现了高精度。

ABSTRACT

Individual pig detection and tracking is an important requirement in many video-based pig monitoring applications. However, it still remains a challenging task in complex scenes, due to problems of light fluctuation, similar appearances of pigs, shape deformations and occlusions. To tackle these problems, we propose a robust real time multiple pig detection and tracking method which does not require manual marking or physical identification of the pigs, and works under both daylight and infrared light conditions. Our method couples a CNN-based detector and a correlation filter-based tracker via a novel hierarchical data association algorithm. The detector gains the best accuracy/speed trade-off by using the features derived from multiple layers at different scales in a one-stage prediction network. We define a tag-box for each pig as the tracking target, and the multiple object tracking is conducted in a key-points tracking manner using learned correlation filters. Under challenging conditions, the tracking failures are modelled based on the relations between responses of detector and tracker, and the data association algorithm allows the detection hypotheses to be refined, meanwhile the drifted tracks can be corrected by probing the tracking failures followed by the re-initialization of tracking. As a result, the optimal tracklets can sequentially grow with on-line refined detections, and tracking fragments are correctly integrated into respective tracks while keeping the original identifications. Experiments with a dataset captured from a commercial farm show that our method can robustly detect and track multiple pigs under challenging conditions. The promising performance of the proposed method also demonstrates a feasibility of long-term individual pig tracking in a complex environment and thus promises a commercial potential.

研究动机与目标

• 在无需人工标记或物理标签的情况下，实现在复杂农场监控环境中对个体猪只进行自动、实时的检测与跟踪。
• 解决基于视频的猪只监测中光照波动、猪只外观相似、形状形变及遮挡等挑战。
• 开发一种能够通过检测优化和故障建模来纠正漂移并整合断裂轨迹的鲁棒跟踪系统。
• 通过真实世界视频数据，在商业农场环境中证明长期个体猪只跟踪的可行性。

提出的方法

• 一种单阶段基于CNN的检测器，利用不同尺度下多层特征，实现检测精度与速度之间的最优权衡。
• 采用基于相关滤波器的跟踪器，以关键点跟踪方式实现猪只跟踪，确保实时性能。
• 提出一种新颖的分层数据关联算法，通过建模检测器与跟踪器响应关系来识别跟踪失败，从而连接检测结果与跟踪轨迹。
• 通过探测失败并必要时重新初始化跟踪器，对检测假设进行优化，并对漂移轨迹进行校正。
• 为每头猪定义标签框作为跟踪目标，实现身份保持的最优轨迹段的顺序增长。
• 该方法在日光和红外条件下均能运行，增强了在光照变化环境下的鲁棒性。

实验结果

研究问题

• RQ1 能否开发一种无需人工标记或物理标签的实时、自动猪只检测与跟踪系统？
• RQ2 在存在遮挡和外观变化的复杂视频场景中，如何有效校正跟踪漂移和断裂轨迹？
• RQ3 通过整合检测优化与故障建模，分层数据关联算法在多大程度上能提升跟踪的鲁棒性？
• RQ4 所提出的方法在包括红外在内的多种光照条件下，能否保持高精度与稳定性？
• RQ5 该系统在真实农场监控视频数据上的长期个体猪只跟踪性能如何？

主要发现

• 所提方法在光照显著变化和遮挡等挑战性条件下，实现了对多头猪只的鲁棒检测与跟踪。
• 分层数据关联算法通过在线检测优化，成功将断裂的轨迹段整合为连贯且身份保持的轨迹段。
• 通过响应分析建模跟踪器故障，并在必要时重新初始化跟踪器，有效校正了跟踪漂移。
• 该系统在真实商业农场数据集上表现出色，证明了其在长期个体猪只监测中的可行性。
• 该方法可实时运行，并在日光和红外条件下均表现有效，显著提升了在商业养殖中的实际应用价值。