| 吕禾丰,陆华才.基于 YOLOv5 算法的交通标志识别技术研究[J].电子测量与仪器学报,2021,35(10):137-144 |
| 基于 YOLOv5 算法的交通标志识别技术研究 |
| Research on traffic sign recognition technology based on YOLOv5 algorithm |
| |
| DOI： |
| 中文关键词: 深度学习 YOLOv5 Cluster NMS EIOU 交通标志识别 |
| 英文关键词:deep learning YOLOv5 Cluster-NMS EIOU traffic sign recognition |
| 基金项目:皖江高端装备制造协同创新中心开放基金项目（GCKJ2018013）、安徽工程大学基金项目（项目编号：Xjky2020022）资助 |
|
|
|
|
| 摘要点击次数: 362 |
| 全文下载次数: 771 |
| 中文摘要: |
| 针对传统方式识别交通标志算法存在的检测精度较低的问题,提出了一种改进 YOLOv5 算法的交通标志识别方法。 首
先改进 YOLOv5 算法的损失函数,使用 EIOU 损失函数代替 YOLOv5 算法所使用的 GIOU 损失函数来优化训练模型,提高算法
的精度,实现对目标更快速的识别;然后使用加权 Cluster 非极大值抑制(NMS)改进 YOLOv5 本身所使用的加权 NMS 算法,提高
生成检测框的准确率。 实验结果表明,改进后的 YOLOv5 算法在由长沙理工大学制作的 CCTSDB 交通标志数据集上训练的模
型的 mAP 值达到了 84. 35%,比原始的 YOLOv5 算法提高了 6. 23%。 所以改进 YOLOv5 算法在交通标志识别中有更高的精度,
能够更好的应用到实践当中。 |
| 英文摘要: |
| Aiming at the low detection accuracy of traditional traffic sign recognition algorithms,a traffic sign recognition method with
improved YOLOv5 algorithm is proposed. First,improve the loss function of the YOLOv5 algorithm,use the EIOU loss function instead of
the GIOU loss function used by the YOLOv5 algorithm to optimize the training model,improve the accuracy of the algorithm, and achieve
faster identification of the target,then use the weighted Cluster NMS to improve the YOLOv5 itself. The weighted NMS algorithm improves
the accuracy of generating the detection frame. The experimental results show that the mAP value of the model trained on the CCTSDB
traffic sign dataset produced by Changsha University of Science and Technology by the improved YOLOv5 algorithm reaches 84. 35%,
which is 6. 23% higher than the original YOLOv5 algorithm. Therefore,the improved YOLOv5 algorithm has higher accuracy in traffic
sign recognition and can be better applied to practice. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
