基于嵌入式和YOLOv8存储柜控制系统实验设计  

Experiment Design of the Storage Cabinet Control System Based on Embedded System and YOLOv8

作　　者：张浩鹏[1,2] 王海珍[1,2] 范梅花[3] 郭强[3] ZHANG Haopeng;WANG Haizhen;FAN Meihua;GUO Qiang(School of Computer and Control Engineering,Qiqihar University,Qiqihar 161006,Heilongjiang,China;Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis,Qiqihar University,Qiqihar 161006,Heilongjiang,China;School of Mechanical and Electrical Engineering,Qiqihar University,Qiqihar 161006,Heilongjiang,China)

机构地区：[1]齐齐哈尔大学计算机与控制工程学院,黑龙江齐齐哈尔161006 [2]齐齐哈尔大学黑龙江省重点实验室-大数据网络安全检测分析,黑龙江齐齐哈尔161006 [3]齐齐哈尔大学机电工程学院,黑龙江齐齐哈尔161006

出　　处：《实验室研究与探索》2024年第5期84-88,共5页Research and Exploration In Laboratory

基　　金：黑龙江省高等教育教学改革研究项目(SJGY20200770,SJGY20210969);黑龙江省省属高等学校基本科研业务费科研项目(145209126);黑龙江省省属高等学校基本科研业务费科研创新平台项目(135409421);黑龙江省教育科学规划重点课题(GJB1422296);黑龙江省教育厅基本科研业务费面上项目(145309613);齐齐哈尔大学教育科学研究项目(GJQTZA202301,GJQTYB202320,GJQTZX2021007);齐齐哈尔大学学位与研究生教育教学改革研究项目(JGXM_QUG_2022027);黑龙江省大学生创新创业训练计划项目(S202310232142)。

摘　　要：为实现存储柜高效和安全的管理,提出基于人脸检测的存储柜控制系统,系统包括终端、后端和前端。终端采用先训练后移植的方式将YOLOv8人脸检测算法部署在嵌入式Android设备,将人脸图片转化为人脸特征文本数据,Retrofit携带人脸特征和存储柜相关数据向后端发送网络请求。后端采用若依框架、Jwt认证权限、Spring Security管理动态权限菜单和Spring Boot实现API接口,MyBatis管理MySQL数据库和Redis缓存数据。前端也采用若依框架,采用Vue2和Element UI通过Axios访问后端获取数据进行可视化。测试结果表明,系统识别效果好,执行器工作正常,数据可视化展示准确,系统运行稳定。To achieve efficient and secure management of the storage cabinet,a face detection-based storage cabinet control system is proposed,the system includes terminals,back-end and front-end.The terminal deploys the YOLOv8 face detection algorithm on embedded Android devices by using a pre-training and post-transplantation approach.It converts face images into face feature text data,and Retrofit carries the face features and storage cabinet-related data to send network requests to the back-end.The back-end employs the RuoYi framework,JWT for authentication and authorization,Spring Security for managing dynamic permission menus,Spring Boot for implementing API interfaces,MyBatis for managing the MySQL database,and Redis for caching data.The front-end also adopts the RuoYi framework,and uses Vue2 and Element UI.It accesses the back-end through Axios to retrieve data for visualization.The results indicate that the system has good recognition performance,the actuators work normally,the data visualization is accurately displayed,and the system operates stably.

关 键 词：YOLOv8 人脸检测 边缘计算 嵌入式Android设备 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]