基于GSR-YOLOv7的探地雷达回波图像识别的研究  被引量:4

Research on GPR Echo Image Recognition Based on GSR-YOLOv7

在线阅读下载全文

作  者:钟运杰 李康 莫思特[1] 李碧雄[2] ZHONG Yunjie;LI Kang;MO Site;LI Bixiong(School of Electrical Engineering,Sichuan University,Chengdu 610065,CHN;College of Water Resources and Hydropower,Sichuan University,Chengdu 610065,CHN)

机构地区:[1]四川大学电气工程学院,成都610065 [2]四川大学水利水电学院,成都610065

出  处:《半导体光电》2023年第5期767-774,共8页Semiconductor Optoelectronics

摘  要:针对探地雷达回波图像特征不突出,工作人员现场解译时间长,准确率不高的问题,提出一种基于YOLOv7改进的回波图像识别方法GSR-YOLOv7。首先使用GhostConv替换YOLOv7卷积层的卷积核以减小参数量;引入SimAM注意力机制提高特征学习能力;通过对感受野模块RFB卷积核的复用,增大感受野;同时引入改进EIoU损失函数提高模型分类能力和回归精度。基于道路回波图像数据集的实验结果表明,改进后的模型MAP_(50)达到了97.57%,MAP_(50∶95)达到了73.13%,较YOLOv7分别提高2.13%和8.46%,模型大小减小了35%。所提出的GSR-YOLOv7模型对于回波目标的检测效果较好,适用于移动端系统,对于体积小、算力低的平台具有较大应用价值。In order to address the problems of indistinct features,time-consuming field interpretation,and low accuracy in Ground Penetrating Radar(GPR) echo image recognition,an improved method called GSR-YOLOv7 based on YOLOv7 is proposed in this study.First,GhostConv was used to replace the convolutional kernels in the YOLOv7 convolutional layers,thus reducing the number of parameters.Second,the SimAM attention mechanism was introduced to improve the feature learning capabilities.Third,the reuse of the receptive field block(RFB) convolutional kernel was implemented to enlarge the receptive field.Finally,an improved EIoU loss function was employed to improve model classification and regression accuracy.Experimental results on a dataset of road echo images demonstrate the effectiveness of the proposed approach.The GSR-YOLOv7 model achieves a MAP_(50) score of 97.57% and a MAP_(50∶95) score of 73.13%,showing improvements of 2.13% and 8.46%,respectively,over YOLOv7.In addition,the model size is reduced by 35%.The GSR-YOLOv7 model exhibits excellent detection performance for echo targets and is suitable for use on mobile systems.It has significant value for platforms with limited processing power and small form factors.

关 键 词:YOLOv7 探地雷达 SimAM GHOST EIoU 目标检测 

分 类 号:TN957.52[电子电信—信号与信息处理] P631[电子电信—信息与通信工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象