Pruning deep convolutional neural networks for efficient edge computing in condition assessment of infrastructures
Pruning deep convolutional neural networks for efficient edge computing in condition assessment of infrastructures
复制标题
DOI：
10.1111/mice.12449
复制
发表时间：
2019-05
期刊：
影响因子：
--
通讯作者：
Rih-Teng Wu;Ankush Singla;M. Jahanshahi;E. Bertino;B. J. Ko;D. Verma
中科院分区：
文献类型：
--
作者：
Rih-Teng Wu;Ankush Singla;M. Jahanshahi;E. Bertino;B. J. Ko;D. Verma
民用基础设施的健康监测是物联网（IoT）的关键应用，而边缘计算是物联网的重要组成部分。在这种情况下，可以取代当前人工检查的自动检查机器人群就是边缘设备的例子。将预先训练的深度学习算法整合到这些机器人中进行自主损伤检测是一个具有挑战性的问题，因为这些设备通常在计算和存储资源方面受到限制。这项研究介绍了一种基于网络修剪的解决方案，该解决方案使用泰勒扩展来利用预训练的深度卷积神经网络进行有效的边缘计算并将其纳入检查机器人。在两个预先训练的网络上进行综合实验的结果（即，VGG 16和ResNet 18）和两种类型的普遍表面缺陷（即，裂纹和腐蚀）的性能，内存需求，和损伤检测的推理时间方面进行了详细的介绍和讨论。结果表明，所提出的方法显着提高资源效率，而不降低损伤检测性能。
Health monitoring of civil infrastructures is a key application of Internet of things (IoT), while edge computing is an important component of IoT. In this context, swarms of autonomous inspection robots, which can replace current manual inspections, are examples of edge devices. Incorporation of pretrained deep learning algorithms into these robots for autonomous damage detection is a challenging problem since these devices are typically limited in computing and memory resources. This study introduces a solution based on network pruning using Taylor expansion to utilize pretrained deep convolutional neural networks for efficient edge computing and incorporation into inspection robots. Results from comprehensive experiments on two pretrained networks (i.e., VGG16 and ResNet18) and two types of prevalent surface defects (i.e., crack and corrosion) are presented and discussed in detail with respect to performance, memory demands, and the inference time for damage detection. It is shown that the proposed approach significantly enhances resource efficiency without decreasing damage detection performance.