|
二维码(扫一下试试看!) |
基于改进YOLOv5的纸病检测方法 |
Paper Defect Detection Method Based on Improved YOLOv5 |
投稿时间:2022-05-27 |
DOI:10.11980/j.issn.0254-508X.2022.10.012 |
关键词: 纸病检测 YOLOv5 批量归一化模块 坐标注意力机制 损失函数 |
Key Words:paper defect detection YOLOv5 batch normalization coordinate attention loss function |
基金项目:陕西省榆林市2020年科技计划项目(CXY-2020-090)。 |
|
摘要点击次数: 2249 |
全文下载次数: 1325 |
摘要:针对传统纸病检测算法中纸病特征提取困难、实时性差的问题,提出一种改进YOLOv5的纸病检测方法。该方法首先在批量归一化模块的首尾部分添加居中和缩放校准,形成更稳定的纸病有效特征分布;其次在骨干网络中添加坐标注意力机制,增强骨干网络的纸病特征提取能力;最后选用 CIoU_loss作为边界框回归的损失函数,实现高精度的定位。实验结果表明,改进后的算法平均精度达99.02%,实时检测速度达41.58 帧/s,相较于现有的基于CNN纸病分类算法,检测精度与检测速度都有较大的提升,且改进后的算法对光源的依赖程度更低,能对各类纸病实现精准辨识。 |
Abstract:Aiming at the problems of difficulty in extracting paper defect features and poor real time in current paper defect detection algorithms, an improved YOLOv5 paper defect detection method was proposed in this study. The method was firstly adding centering and scaling calibration to the head and tail of the batch normalization module to form a more stable effective feature distribution of paper defects. Secondly, a coordinate attention was added to the backbone network to enhance the paper defect feature extraction capability of the backbone network. Finally, CIoU_loss was selected as the loss function of bounding box regression to achieve high-precision positioning. The experimental results showed that the average accuracy of the improved algorithm reached 99.02%, the real-time detection speed reached 41.58 frames/s. Compared with the existing CNN-based paper defect classification algorithm, the detection accuracy and detection speed were greatly improved. The improved algorithm was less dependent on the light source and could accurately identify various paper defects. |
查看全文 查看/发表评论 下载PDF阅读器 |