电工电气杂志社

基于改进YOLOv7的变电设备红外图像轻量识别检测方法

来源：电工电气发布时间：2024-12-02 09:02 浏览次数：44

基于改进YOLOv7的变电设备红外图像轻量识别检测方法

陈海波，叶金翔，王生祺

(国网浙江省电力有限公司超高压分公司，浙江杭州 310000)

摘要：变电站设备准确的红外热图像识别与检测是其温度状态智能分析的先决条件。为了克服复杂背景的干扰，提出了改进的轻量级 YOLOv7 方法，以提高在复杂红外背景下变电站设备的识别效果。提出的方法引入了高分辨率 P2 检测头来改进小目标检测，无参数注意模块 SimAM 在复杂红外背景中更好地提取不同变电设备目标特征，CARAFE 模块在上采样过程中减少特征信息的损失，进一步增强算法的鲁棒性。实验及测试结果显示提出的模型比原始 YOLOv7-tiny 高出 2.6% 检测精度，实现了 101 FPS(帧数)的实时推理速度，证明了所提出的模型在变电站设备的红外图像目标识别方面的优势，特别是较小的变电设备，并且提出的模型比其他轻量级模型拥有更高的识别检测精度。

关键词： 变电设备；红外图像；目标识别与检测；计算机视觉；深度学习

中图分类号：TM63 ；TP391.41 文献标识码：B 文章编号：1007-3175(2024)11-0055-06

Lightweight Recognition and Detection Method for Infrared Images of

Substation Equipment Based on Improved YOLOv7

CHEN Hai-bo, YE Jin-xiang, WANG Sheng-qi

（State Grid Zhejiang Electric Power Co., Ltd. Ultra High Voltage Branch, Hangzhou 310000, China）

Abstract: The accurate recognition and detection of infrared thermal images of substation equipment is a prerequisite for intelligent analysis of its thermal status. To address the interference posed by complex backgrounds, this paper presents an improved light weight YOLOv7 method aimed at enhancing the recognition performance of substation equipment under intricate infrared conditions. The proposed approach introduces a high-resolution P2 detection head to improve small target detection, the parameter-free attention module SimAM effectively extracts target features of various substation equipment under the complex infrared backgrounds. Additionally, the CARAFE module minimizes the loss of feature information during the upsampling process, further bolstering the algorithm's robustness. Experimental results demonstrates that the proposed model surpasses the original YOLOv7-tiny by 2.6% in detection accuracy, achieving a real-time inference speed of 101 FPS. It is proved that the proposed model has advantages in infrared image target recognition of substation equipment, especially small substation equipment, and the proposed model has higher recognition and detection accuracy than other lightweight models.

Key words: substation equipment; infrared image; target recognition and detection; computer vision; deep learning

参考文献

[1] MA Jianchao , ZHENG Hanbo , SUN Yonghui,et al.Temperature Compensation Method for Infrared Detection of Live Equipment Under the Interferences of Wind Speed and Ambient Temperature[J].IEEE Transactions on Instrumentation and Measurement，2021，70 ：1-9.

[2] SADYKOVA Diana, PERNEBAYEVA Damira, BAGHERI Mehdi, et al.IN-YOLO: Real-Time Detection of Outdoor High Voltage Insulators Using UAV Imaging[J].IEEE Transactions on Power Delivery，2019，35(3) ：1599-1601.

[3] TU Y, GONG B, WANG C, et al.Effect of Moisture on Temperature Rise of Composite Insulators Operating in Power System[J].IEEE Transactions on Dielectrics and Electrical Insulation，2015，22(4) ：2207-2213.

[4] 张琦，欧嘉俊，谢剑翔，等. 基于图像监控的换流站变电设备智能巡检系统[J] . 电气开关，2023，61(6) ：36-42.

[5] 熊凯飞. 雾霾天气及遮挡场景下的变电设备视觉检测方法[D]. 长沙：长沙理工大学，2024.

[6] 岳晓彤. 基于图像识别的同塔多回线路带电作业状态诊断技术研究[D]. 鞍山：辽宁科技大学，2023.

[7] 赵振兵，冯烁，赵文清，等. 融合知识迁移和改进 YOLOv6 的变电设备热像检测方法[J]. 智能系统学报，2023，18(6) ：1213-1222.

[8] 黄悦华，杨楚睿，陈晨，等. 基于改进 Centernet 的变电设备红外检测方法[J] . 电子测量技术，2023，46(4) ：142-148.

[9] 郑文杰，杨祎，乔木，等. 基于改进 YOLO 和 Resnet 的变电设备热缺陷识别及诊断方法[J]. 重庆理工大学学报(自然科学)，2023，37(9) ：261-269.

[10] 周阳洋，胡俊华，徐华，等. 一种瓷支柱绝缘子红外图像目标检测算法[J] . 浙江电力，2023，42(11) ：78-85.

[11] YANG Lingxiao, ZHANG Ruyuan, LI Lida, et al.SimAM:A Simple, Parameter-Free Attention Module for Convolutional Neural Networks[C]//International Conference on Machine Learning，2021，139 ：11863-11874.

[12] WANG Jiaqi, CHEN Kai, XU Rui, et al.CARAFE:Content-Aware ReAssembly of FEatures[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV)，2020 ：3007-3016.

[13] ZHENG Hanbo, CUI Yaohui, YANG Wenqiang, et al. An Infrared Image Detection Method of Substation Equipment Combining Iresgroup Structure and CenterNet [J] . IEEE Transactions on Power Delivery，2022，37(6) ：4757-4765.

返回索引列表

Article retrieval

文章检索

基于改进YOLOv7的变电设备红外图像轻量识别检测方法