华侨大学学报（自然科学版）2025，Vol.46Issue(3)：255-263,9.DOI:10.11830/ISSN.1000-5013.202412006
改进YOLOv8n模型的火灾场景火焰检测方法
Flame Detection Method in Fire Scene With Improved YOLOv8n Model
摘要
Abstract
Aiming at the problem of low accuracy in flame detection caused by complex smoke and dust envi-ronments in fire scenes,an efficient and precise flame detection method based on the YOLOv8n model was pro-posed.First,a variety of fire scene images were selected as the original images for the dataset,and random noise,such as salt and pepper noise,was added to simulate a smoke and dust environment.Second,a median filtering module was embedded at the front of the model's network framework to enhance the network's capa-bility to handle interference noise in smoke and dust environments.Finally,by utilizing Ghost convolution modules and designing cross layer connection networks at different lay levels,the number of parameters was re-duced while the generalization capability of the network was optimized.This enable real-time and high-preci-sion flame detection in fire scene with noise interference.Experimental results show that the improved YOLOv8n model had superior real-time performance and detection accuracy performance.关键词
火焰检测/椒盐噪声/YOLOv8n模型/中值滤波模块/轻量级Ghost卷积Key words
flame detection/random noise/YOLOv8n model/median filtering module/lightweight Ghost convolution分类
计算机与自动化引用本文复制引用
乐其河,陈炜,郑祥盘,许亦镜,林立霖..改进YOLOv8n模型的火灾场景火焰检测方法[J].华侨大学学报（自然科学版）,2025,46(3):255-263,9.基金项目
福建省自然科学基金资助项目(2022J05235) （2022J05235）
福建省技术创新重点攻关及产业化项目(校企联合类)(2023XQ018) （校企联合类）
闽江学院"揭榜挂帅"项目(ZD202303) （ZD202303）
闽江学院预研项目(MJY22022) （MJY22022）
