基于语义信息距离解耦的变电运维多类别缺陷图像检测  

Decoupled Sematic Distance Based Multi-class Defect Scene Detecting for Substations

作　　者：张鑫 叶俊杰 崔瑶 黄鑫 仲林林 ZHANG Xin;YE Junjie;CUI Yao;HUANG Xin;ZHONG Linlin(NARI Group Corporation(State Grid Electric Power Research Institute),Nanjing 211106,China;State Key Laboratory of Smart Grid Protection and Control,Nanjing 211106,China;School of Electrical Engineering,Southeast University,Nanjing 210096,China)

机构地区：[1]南瑞集团有限公司(国网电力科学研究院有限公司),江苏南京211106 [2]智能电网保护和运行控制国家重点实验室,江苏南京211106 [3]东南大学电气工程学院,江苏南京210096

出　　处：《中国电力》2023年第6期209-218,共10页Electric Power

基　　金：国网电力科学研究院有限公司科技项目(524606210002)。

摘　　要：变电站设备种类繁多、缺陷类型复杂、特征差异大,传统的基于深度学习的缺陷图像检测模型难以同时有效处理不同设备的多种缺陷。为此,提出了一种基于语义信息距离解耦的缺陷图像检测模型(sematic-distance based decoupling detection model,SDB-DDM)。首先对缺陷类别进行语义信息聚簇,构建解耦式网络结构,然后对网络输出进行加权锚框融合,并在损失函数中加入局部预测损失以提升预测能力,同时提出解耦式非极大值抑制策略以加快模型推理速度。该模型可根据缺陷类别进行自适应调整,以适用变电运维多类别缺陷图像检测的应用场景。实验结果显示,该模型的平均精度均值达到了69.68%。同平台下相较于目前性能最佳的目标检测模型(YOLOX),精度提升了1.36个百分点,参数量下降了5%,推理速度提升了34%。Due to the complexity and differences of defect types in substations,traditional deep learning models for defects detection lack comprehensive response ability.It proposes a sematic distance based decoupling detection model.Firstly,the decoupled model structure is determined by clustering defect classes according to the semantic information distance between each other.Then,the weighted anchor fusion and local prediction loss techniques are used to improve the model performance.Meanwhile,the decoupled non-maximum suppression strategy is proposed to accelerate the model inference process.The experiment results show that the mean average precision of the model reaches 69.68%.Compared with YOLOX,which has been recognized as the best real-time object detection model,the accuracy of proposed model is improved by 1.36 percentage points,the parameter quantity is reduced by 5%,and the inference speed is improved by 34%.

关 键 词：变电运维场景 缺陷检测 深度学习 语义信息距离 解耦式模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TM63[自动化与计算机技术—计算机科学与技术]