【论文推荐】武汉大学 李可欣等：边端轻量级YOLOv4改进算法及其在电力现场作业中的应用

为实现电网现场作业的实时安全管控，采用“云-边-端”协同的系统架构。利用移动边端设备搭载轻量化的智能识别算法实现现场作业风险的就地判断。YOLOv4-Tiny算法具有轻量级、检测速度快等特点，经常部署在移动边端上用于实时视频流的检测，但该算法的检测精度较低。本文提出了一种基于注意力机制和优化训练方法的改进YOLOv4-Tiny算法，在不损失算法检测速度的前提下，可以提高算法的检测精度。具体而言，在主干网络中加入了卷积块注意模块分支以提高特征提取能力，在颈部网络中加入了高效通道注意机制以提高特征利用率。此外，为了提高该改进算法的训练效果，采用了转移学习、mosaic数据增强和标签平滑三种优化训练方法。最后，在搭载了NVIDIA Jetson Xavier NX芯片的边缘计算装置上对改进算法进行测试。结果表明，改进后的YOLOv4-Tiny算法在检测现场着装规范数据集时的速度为17.25FPS，平均平均精度（mAP）从70.89%提升到85.03%。

A “cloud-edge-end” collaborative system architecture is adopted for real-time security management of power system on-site work, and mobile edge computing equipment utilizes lightweight intelligent recognition algorithms for onsite risk assessment and alert.Owing to its lightweight and fast speed, YOLOv4-Tiny is often deployed on edge computing equipment for real-time video stream detection; however, its accuracy is relatively low.This study proposes an improved YOLOv4-Tiny algorithm based on attention mechanism and optimized training methods, achieving higher accuracy without compromising the speed.Specifically, a convolution block attention module branch is added to the backbone network to enhance the feature extraction capability and an efficient channel attention mechanism is added in the neck network to improve feature utilization.Moreover, three optimized training methods: transfer learning, mosaic data augmentation, and label smoothing are used to improve the training effect of this improved algorithm.Finally, an edge computing equipment experimental platform equipped with an NVIDIA Jetson Xavier NX chip is established and the newly developed algorithm is tested on it.According to the results, the speed of the improved YOLOv4-Tiny algorithm in detecting on-site dress code compliance datasets is 17.25 FPS, and the mean average precision (mAP) is increased from 70.89% to 85.03%.

On-site power system work, YOLOv4-Tiny, Convolution block attention mechanism, Efficient channel attention, Optimized training methods.