| 一种高精度优化Faster-RCNN变电站安全帽检测方法 |
| |
| 引用本文： | 张彦凯,余飞鸿,夏 岩,刘 勇,张 蕊. 一种高精度优化Faster-RCNN变电站安全帽检测方法[J]. 四川电力技术, 2023, 46(1): 62-68 |
| |
| 作者姓名： | 张彦凯 余飞鸿 夏 岩 刘 勇 张 蕊 |
| |
| 作者单位： | 兰州陇能电力科技有限公司;四川轻化工大学自动化与信息工程学院; 智能电网四川省重点实验室 |
| |
| 摘 要： | 针对传统变电站人工视频安全帽检测效率低和错误率高的问题，文中提出了一种优化的高精度Faster-RCNN安全帽检测方法。首先，以RPN为主干网络引入特征金字塔，增强检测算法浅层和深层信息表征能力；接着，引入K-Means++聚类算法优化Anchor参数，提升网络对小目标的检测能力；然后，以ROI Align代替RoI池化，消除量化误差和原始图片与特征图的映射偏差，从而提高检测精度；最后，采用开源安全帽数据集对改进的网络进行训练与测试，并将该模型与YOLOv3、RFBnet和传统Faster-RCNN进行了对比。实验结果表明：优化的Faster-RCNN相比RFBnet、YOLOv3和传统Faster-RCNN模型，将mAP值分别提高了6.81%、9.57%和5.09%,达到了92.43%;检测速度为18 frame/s,同时增强了变电站安全帽高精度识别能力。
|
| 关 键 词： | 变电站 目标检测 Faster-RCNN K-Means++聚类 ROI Align |
A Detection Method for Safety Helmet in Substation Based on Improved High-precision Faster-RCNN |
| |
| ZHANG Yankai,YU Feihong,XIA Yan,LIU Yong,ZHANG Rui. A Detection Method for Safety Helmet in Substation Based on Improved High-precision Faster-RCNN[J]. Sichuan Electric Power Technology, 2023, 46(1): 62-68 |
| |
| Authors: | ZHANG Yankai YU Feihong XIA Yan LIU Yong ZHANG Rui |
| |
| Affiliation: | Lanzhou Longneng Electric Power Science & Technology Ltd.;School of Automation and Information Engineering, Sichuan University of Science & Engineering; Intelligent Electric Power Grid Key Laboratory of Sichuan Province |
| |
| Abstract: | Aiming at low efficiency and a high error rate of safety helmet detectionin substations by the traditional manual video detection method, a helmet detection method based on the improved high-precision Faster-RCNN is proposed. Firstly, the feature pyramid network module is introduced into the RPN backbone network, so that the representation |
| |
| Keywords: | substation edge detection Faster-RCNN K-Means++ clustering ROI Align |
|
| 点击此处可从《四川电力技术》浏览原始摘要信息 |
|
点击此处可从《四川电力技术》下载免费的PDF全文 |
