| 注册
首页|期刊导航|内蒙古农业大学学报(自然科学版)|基于改进YOLOv8n的水稻病害检测方法

基于改进YOLOv8n的水稻病害检测方法

庞超 王传安 苏煜 张中杰

内蒙古农业大学学报(自然科学版)2024,Vol.45Issue(2):62-68,7.
✕
内蒙古农业大学学报(自然科学版)2024,Vol.45Issue(2):62-68,7.DOI:10.16853/j.cnki.1009-3575.2024.02.008

基于改进YOLOv8n的水稻病害检测方法

Rice Disease Detection Method Based on Improved YOLOv8n

庞超 1王传安 2苏煜 1张中杰1

作者信息

  • 1. 安徽科技学院机械工程学院,凤阳 233100
  • 2. 安徽科技学院信息与网络工程学院,凤阳 233100
  • 折叠

摘要

Abstract

Aiming at the problems of multiple diseases and low detection efficiency in rice production,an improved YOLOv8n model for rice disease identification was proposed.First,the BiFPN pyramid structure was incorporated for the integration of effective fea-ture layers.Second,the GAM attention mechanism was introduced after the original backbone network SPPF module to rationally al-locate resources.Finally,the WIoU loss function was used to improve the detection accuracy.The experimental results showed that the MAP of BGW-YOLOv8n in the rice disease detection task reached 91.6%,which was 3.6%higher than that of YOLOv8n.This method was superior to other common algorithms such as YOLOv5,and improved the phenomenon of disease missing detection.

关键词

水稻病害/YOLO/目标检测/全局注意力机制

Key words

Rice disease/YOLO/Object detection/Global attention mechanism

分类

信息技术与安全科学

引用本文复制引用

庞超,王传安,苏煜,张中杰..基于改进YOLOv8n的水稻病害检测方法[J].内蒙古农业大学学报(自然科学版),2024,45(2):62-68,7.

基金项目

专业服务安徽省十大新兴产业项目(Xj2022009) (Xj2022009)

内蒙古农业大学学报(自然科学版)

OA北大核心

1009-3575

访问量0
|
下载量0
段落导航相关论文