摘要
由于无人机巡检图像中复杂背景的干扰以及航拍角度等外在因素的影响,会给绝缘子目标的识别带来一定的难度。常用的Faster R-CNN模型在进行复杂背景下绝缘子目标检测时,存在远处的或被遮挡的小目标绝缘子的漏检问题,所以本文在现有的Faster R-CNN模型上选择ResNet101作为骨干网络,引入FPN结构提高对被遮挡的小目标绝缘子的检测精度,降低了受遮挡影响的目标的漏检率,并增加通道注意力机制SENet以增强绝缘子特征,提高特征表达能力。实验结果表明,该基于Faster R-CNN的改进模型在复杂背景下绝缘子目标检测中达到精度AP^(50)为93.2%,相较于基线模型AP^(50)提高了6.4%,并且优于目前一些先进的目标检测模型,对复杂背景下绝缘子的检测精度高,解决小目标绝缘子误检和漏检问题。
Due to the interference of complex background in UAV inspection image and the influence of external factors such as aerial shooting Angle,insulator object recognition will bring certain difficulty.When the commonly used Faster R-CNN model is used for insulator object detection under complex background,there is the problem of missing detection of small object insulators that are distant or blocked.Therefore,this paper selects ResNet101as the backbone network on the existing Faster R-CNN model.The FPN structure is introduced to improve the detection accuracy of the occluded small object insulator,reduce the missed detection rate of occluded targets,and the channel attention mechanism SENet is added to enhance the insulator characteristics.The experimental results show that the improved model based on Faster R-CNN achieves an accuracy of 93.2%in insulator object detection under complex background,which is 6.4%higher than that of the baseline model AP^(50),and is superior to some advanced object detection models at present,with high detection accuracy for insulators under complex background,and can solve the problem of false detection and missing detection of small object insulators.
作者
翟永杰
王璐瑶
郭聪彬
Zhai Yongjie;Wang Luyao;Guo Congbin(Department of Automation,North China Electric Power University,Baoding 071003,China)
出处
《电子测量技术》
北大核心
2023年第16期187-194,共8页
Electronic Measurement Technology
基金
国家自然科学基金(61871182)
河北省自然科学基金(F2020502009,F2021502008)
中央高校基本科研业务费专项资金(2021MS081)项目资助
作者简介
翟永杰,博士,教授,主要研究方向为模式识别与计算机视觉、机器学习与人工智能、分散控制系统设计与应用等。E-mail:zhaiyongjie@ncepu.edu.cn;通信作者:王璐瑶,硕士研究生,主要研究方向为模式识别与计算机视觉。E-mail:wangluyao@ncepu.edu.cn;郭聪彬,硕士研究生,主要研究方向为电力视觉与知识推理。E-mail:guocongbin@ncepu.edu.cn
