摘要
针对森林火灾检测对实时性要求较高的问题,提出了一种基于改进YOLOv8的森林火灾检测方法 .在YOLOv8的基础上,采用轻量化特征提取网络EfficientNet优化YOLOv8原主干网络CSPDarknet53,以减少计算量并提高模型的收敛速度,进而提高火灾检测速度.此外,融入了SENet注意力机制模块,以增强网络对火灾检测的准确性.使用α-IoU损失函数代替YOLOv8原始损失函数中的CIoU损失函数来计算定位损失,该函数能够自适应地调整IoU的阈值,更好地处理不同大小和形状的目标,同时提高模型对小目标的检测性能.结果表明:该方法的平均准确率(mA@0.5P)达到了87.2%,帧率(FPS)提升了17帧,显著提高了火灾检测的实时性.
An improved forest fire detection method based on YOLOv8 is proposed to solve the challenge of achieving high real-time performance in forest fire detection.Building upon YOLOv8,the lightweight feature extraction network EfficientNet is utilized to optimize the original YOLOv8 backbone network CSPDarknet53 to diminish computational requirements and accelerate model convergence,thereby accelerating the fire detection speed.Additionally,the SENet attention mechanism module is integrated to bolster the model's accuracy in fire detection.Theα-IoU loss function is implemented to supplant the CIoU loss function from YOLOv8's original loss function for calculating positioning loss.This function can adaptively fine-tune the IoU threshold to more effectively handle targets of varying sizes and shapes,while also enhancing the model′s capability to detect small targets.The outcomes demonstrate that the proposed method achieves an average accuracy of 87.2%at mA@0.5P,with a 17-frame increase in the frame rate per second(FPS),significantly enhancing the real-time capabilities of fire detection.
作者
雷建云
田祚汉
夏梦
雷瑞璠
LEI Jianyun;TIAN Zuohan;XIA Meng;LEI Ruifa(South-Central Minzu University,School of Computer Science,Wuhan 430074,China;South-Central Minzu University,Hubei Provincial Engineering Research Center for Intelligent Management of Manufacturing Enterprises,Wuhan 430074,China;South-Central Minzu University,Hubei Provincial Engineering Research Center of Agricultural Blockchain and Intelligent Management,Wuhan 430074,China)
出处
《中南民族大学学报(自然科学版)》
2026年第1期97-105,共9页
Journal of South-Central Minzu University(Natural Science Edition)
基金
湖北省技术创新计划重点研发资助专项(2023BAB087)
中央引导地方科技发展资金资助项目(ZYYD2024QY08)
武汉市重点研发计划资助项目(2023010402010614)
武汉东湖新技术开发区“揭榜挂帅”资助项目(2023KJB204)。
