基于YOLOv7改进的喷漆表面缺陷检测算法研究  

Research on improved paint surface defect detection algorithm based on YOLOv7

作　　者：杨晓强[1] 蒋志鹏 YANG Xiaoqiang;JIANG Zhipeng(College of Aviation Engineering,Civil Aviation Flight University of China,Guanghan 618300,China)

机构地区：[1]中国民用航空飞行学院航空工程学院,四川广汉618300

出　　处：《电子设计工程》2025年第4期30-35,共6页Electronic Design Engineering

基　　金：中国民用航空飞行学院科研创新基金(JG2022-16)。

摘　　要：针对产品表面喷漆检测任务,提出了一种基于YOLOv7改进的喷漆表面缺陷检测算法,旨在提高检测准确性和效率。对YOLOv7算法进行改进,新增小目标检测层,增强模型对不同尺寸目标的检测能力。通过引入GAM注意力机制,提升模型在特征提取方面的能力。替换原有损失函数为SIoU损失函数,不仅能加速模型的收敛过程,还能有效提高其整体精度。实验结果显示,改进后的算法平均精度能达到92.84%,比YOLOv7算法提升5.4%,能够为喷漆表面缺陷检测提供帮助。Aiming at the task of product surface spray painting detection,an improved algorithm based on YOLOv7 was proposed to improve the detection accuracy and efficiency.The YOLOv7 algorithm is improved and a small target detection layer is added to enhance the detection ability of the model for targets of different sizes.By introducing GAM attention mechanism,the ability of the model in feature extraction is improved.Replacing the original loss function with SIoU loss function can not only accelerate the convergence process,but also effectively improve the overall accuracy of the model.Experimental results show that the average accuracy of the improved algorithm can reach 92.84%,which is 5.4%higher than that of the YOLOv7 algorithm,and can provide help for the detection of paint surface defects.

关 键 词：YOLOv7 喷漆表面缺陷检测 注意力机制 损失函数 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]