计算机与数字工程2023，Vol.51Issue(12)：2852-2858,7.DOI:10.3969/j.issn.1672-9722.2023.12.016
基于YOLOv5架构的大幅面SAR图像车辆目标识别方法
Vehicle Target Recognition of Large-scale SAR Images Based on YOLOv5
李庆 1田甜 1田金文1
作者信息
- 1. 华中科技大学多谱信息处理技术国家级重点实验室 武汉 430074||华中科技大学人工智能与自动化学院 武汉 430074
- 折叠
摘要
Abstract
Vehicle target recognition in SAR image is a challenging frontier research field.A vehicle target recognition method in large-scale SAR image based on YOLOv5 is proposed.Taking the convolutional neural network YOLOv5 as the basic model of ve-hicle target recognition in large-scale SAR image,the transfer learning method is used to obtain the initial parameters of the model,which effectively reduces the number of training samples and improves the convergence speed of the model.In order to test the per-formance of the algorithm,a large-scale SAR image dataset containing vehicle targets is constructed.Simulation experiments are carried out on this dataset and compared with some classical deep learning networks.The experimental results show that the pro-posed vehicle target recognition algorithm in large-scale SAR image has higher recognition accuracy and faster speed.关键词
SAR图像/车辆目标识别/卷积神经网络/迁移学习/YOLOv5Key words
SAR images/vehicle target recognition/convolutional neural network(CNN)/transfer learning/YOLOv5分类
信息技术与安全科学引用本文复制引用
李庆,田甜,田金文..基于YOLOv5架构的大幅面SAR图像车辆目标识别方法[J].计算机与数字工程,2023,51(12):2852-2858,7.
