School of Technology, Beijing Forestry University, Beijing 100083, China.
School of Microelectronics and Control Engineering, Changzhou University, Changzhou 213164, China.
Comput Intell Neurosci. 2022 Feb 7;2022:8615374. doi: 10.1155/2022/8615374. eCollection 2022.
Wild animals are essential for ecosystem structuring and stability, and thus they are important for ecological research. Since most wild animals have high athletic or concealable abilities or both, it is used to be relatively difficult to acquire evidence of animal appearances before applications of camera traps in ecological researches. However, a single camera trap may produce thousands of animal images in a short period of time and inevitably ends up with millions of images requiring classification. Although there have been many methods developed for classifying camera trap images, almost all of them follow the pattern of a very deep convolutional neural network processing all camera trap images. Consequently, the corresponding surveillance area may need to be delicately controlled to match the network capability, and it may be difficult to expand the area in the future. In this study, we consider a scenario in which camera traps are grouped into independent clusters, and images produced by a cluster are processed by an edge device installed with a customized network. Accordingly, edge devices in this scenario may be highly heterogeneous due to cluster scales. Resultantly, networks popular in the classification of camera trap images may not be deployable for edge devices without modifications requiring the expertise which may be hard to obtain. This motivates us to automatize network design via neural architecture search for edge devices. However, the search may be costly due to the evaluations of candidate networks, and its results may be infeasible without considering the resource limits of edge devices. Accordingly, we propose a search method using regression trees to evaluate candidate networks to lower search costs, and candidate networks are built based on a meta-architecture automatically adjusted regarding to the resource limits. In experiments, the search consumes 6.5 hours to find a network applicable to the edge device Jetson X2. The found network is then trained on camera trap images through a workstation and tested on Jetson X2. The network achieves competitive accuracies compared with the automatically and the manually designed networks.
野生动物对于生态系统的结构和稳定性至关重要，因此它们对于生态研究也很重要。由于大多数野生动物都具有很高的运动能力或隐藏能力，或者两者兼而有之，因此在生态研究中应用相机陷阱之前，获取动物出现的证据相对较为困难。然而，单个相机陷阱在短时间内可能会产生数千张动物图像，并且不可避免地会产生需要分类的数百万张图像。虽然已经开发了许多用于对相机陷阱图像进行分类的方法，但几乎所有方法都遵循非常深的卷积神经网络处理所有相机陷阱图像的模式。因此，相应的监控区域可能需要精细控制以匹配网络能力，并且将来可能难以扩展该区域。在本研究中，我们考虑了一种场景，其中相机陷阱被分组到独立的集群中，并且由集群生成的图像由安装有定制网络的边缘设备处理。因此，由于集群规模的原因，这种情况下的边缘设备可能具有高度的异构性。因此，如果不进行修改，在分类相机陷阱图像方面很流行的网络可能无法在没有专业知识的情况下部署在边缘设备上，而这种专业知识可能难以获得。这促使我们通过神经架构搜索为边缘设备自动化网络设计。然而，由于候选网络的评估，搜索可能会很昂贵，并且如果不考虑边缘设备的资源限制，其结果可能不可行。因此，我们提出了一种使用回归树评估候选网络以降低搜索成本的搜索方法，并且根据资源限制自动构建候选网络。在实验中，搜索需要 6.5 个小时才能找到适用于边缘设备 Jetson X2 的网络。然后，通过工作站在相机陷阱图像上训练找到的网络，并在 Jetson X2 上进行测试。与自动和手动设计的网络相比，该网络的准确率具有竞争力。
