基于YOLOv5的物联网草莓病虫害监测系统设计  

Design of IoT Strawberry Disease and Pest Monitoring System Based on YOLOv5

作　　者：黄伟州 周小杰 汪婵 冯智 陈子琪 HUANG Weizhou;ZHOU Xiaojie;WANG Chan;FENG Zhi;CHEN Ziqi(School of Intelligent Manufacturing,Anhui Science and Technology University,Fengyang Anhui 233100,China;School of Electrical and Electronic Engineering,Anhui Science and Technology University,Fengyang Anhui 233100,China)

机构地区：[1]安徽科技学院智能制造学院,安徽凤阳233100 [2]安徽科技学院电气与电子工程学院,安徽凤阳233100

出　　处：《兰州工业学院学报》2024年第6期55-60,共6页Journal of Lanzhou Institute of Technology

基　　金：安徽省高校自然科学研究项目(2023AH051866)。

摘　　要：针对传统农业模式下草莓生长周期监测存在依赖人工经验,无法实时掌握病虫害,以及生长环境监测耗费大量人力物力的问题,设计了一种基于物联网与深度学习算法(YOLOv5)的草莓病虫害监测系统。系统通过物联网传感器,实时采集大棚的土壤PH值、温湿度、光照、空气质量等数据,将传感器监测的数据传输至STM32上,并通过Wi-Fi通信方式将数据上传至云平台。同时,结合改良的YOLOv5算法,将草莓植株的分类图像识别平均准确率mAP提升至84.5%,从而迅速发现病虫害并检测草莓成熟度。Aiming at the problems that strawberry growth cycle monitoring in traditional agricultural mode relies on manual experience,cannot grasp diseases and pests in real time,and the monitoring of growth environment costs a lot of manpower and material resources,a strawberry pest monitoring system based on the Internet of Things and deep learning algorithm(YOLOv5)is designed.The system collects soil PH value,temperature and humidity,light,air quality and other data of the greenhouse in real time through the Internet of Things sensor,transmits the data monitored by the sensor to the STM32,and uploads the data to the cloud platform through Wi-Fi communication.Meanwhile,combined with the improved YOLOv5 algorithm,the average accuracy of mAP of classification image recognition of strawberry plants is increased to 84.5%,so as to quickly detect diseases and pests and detect strawberry maturity.

关 键 词：物联网 深度学习 病虫害监测 STM32 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]