Mulero-Pázmány Margarita, Hurtado Sandro, Barba-González Cristóbal, Antequera-Gómez María Luisa, Díaz-Ruiz Francisco, Real Raimundo, Navas-Delgado Ismael, Aldana-Montes José F
Department of Animal Biology, University of Málaga, 29071, Málaga, Spain.
KHAOS Research Group, ITIS Software, University of Málaga, 29071, Málaga, Spain.
Sci Rep. 2025 May 9;15(1):16191. doi: 10.1038/s41598-025-90249-z.
Wildlife biologists increasingly use camera traps for monitoring animal populations. However, manually sifting through the collected images is expensive and time-consuming. Current deep learning studies for camera trap images do not adequately tackle real-world challenges such as imbalances between animal and empty images, distinguishing similar species, and the impact of backgrounds on species identification, limiting the models' applicability in new locations. Here, we present a novel two-stage deep learning framework. First, we train a global deep-learning model using all animal species in the dataset. Then, an agglomerative clustering algorithm groups animals based on their appearance. Subsequently, we train a specialized deep-learning expert model for each animal group to detect similar features. This approach leverages Transfer Learning from the MegaDetectorV5 (YOLOv5 version) model, already pre-trained on various animal species and ecosystems. Our two-stage deep learning pipeline uses the global model to redirect images to the appropriate expert models for final classification. We validated this strategy using 1.3 million images from 91 camera traps encompassing 24 mammal species and used 120,000 images for testing, achieving an F1-Score of 96.2% using expert models for final classification. This method surpasses existing deep learning models, demonstrating improved precision and effectiveness in automated wildlife detection.
野生动物生物学家越来越多地使用相机陷阱来监测动物种群。然而，手动筛选收集到的图像既昂贵又耗时。目前针对相机陷阱图像的深度学习研究未能充分应对现实世界中的挑战，如动物图像与空图像之间的不平衡、区分相似物种以及背景对物种识别的影响，限制了模型在新地点的适用性。在此，我们提出了一种新颖的两阶段深度学习框架。首先，我们使用数据集中的所有动物物种训练一个全局深度学习模型。然后，一种凝聚聚类算法根据动物的外观对其进行分组。随后，我们为每个动物组训练一个专门的深度学习专家模型，以检测相似特征。这种方法利用了从MegaDetectorV5（YOLOv5版本）模型进行的迁移学习，该模型已经在各种动物物种和生态系统上进行了预训练。我们的两阶段深度学习管道使用全局模型将图像重定向到适当的专家模型进行最终分类。我们使用来自91个相机陷阱的130万张图像对该策略进行了验证，这些图像涵盖24种哺乳动物物种，并使用120,000张图像进行测试，使用专家模型进行最终分类时F1分数达到了96.2%。该方法超越了现有的深度学习模型，在自动野生动物检测中展示了更高的精度和有效性。
