桂林电子科技大学学报2024，Vol.44Issue(6)：613-620,8.DOI:10.16725/j.1673-808X.2022340
基于改进YOLOv5的异型烟检测算法
Special-shaped cigarette detection algorithm based on improved YOLOv5
廖家设 1张彤 1谢青峰 1辛敏 1张祎楠1
作者信息
- 1. 桂林电子科技大学机电工程学院,广西桂林 541004
- 折叠
摘要
Abstract
In order to solve the problem of heavy workload in checking and detecting special-shaped cigarettes,reduce the false de-tection rate,and improve the working efficiency and intelligent level of the tobacco logistics industry,a special-shaped cigarette de-tection algorithm based on improved YOLOv5 is proposed.The algorithm takes YOLOv5 as the basic framework,introduces pyra-mid split attention(PSA)module to extract feature space domain information,and concatenate with feature channel domain informa-tion extracted by Focus module to obtain multi-dimensional feature information;Atrous Spatial Pyramid Pooling(ASPP)is used to replace SPP module,and use different ratio of cavity convolution operation to expand receptive field and reduce operation parame-ters while ensuring resolution;K-means++algorithm is used to optimize the anchor frame and improve the matching degree be-tween the anchor frame and the obtained image.The algorithm was tested with the special-shaped cigarette image set collected in the production line,and the mAP parameter of the experimental result is 94.14％,which is 5.04％higher than the result before optimiza-tion.In addition,the improved model does not introduce complex modules,and can be deployed in the AI edge computing integrat-ed development board.The reasoning and recognition time is about 75 ms,which can meet the real-time requirements of the produc-tion line.关键词
YOLOv5/异型烟检测/K-means++/金字塔切分注意力/空洞空间金字塔池化Key words
YOLOv5/special-shaped cigarette detection/K-means++/pyramid split attention/atrous spatial pyramid pooling分类
信息技术与安全科学引用本文复制引用
廖家设,张彤,谢青峰,辛敏,张祎楠..基于改进YOLOv5的异型烟检测算法[J].桂林电子科技大学学报,2024,44(6):613-620,8.
