基于YOLOv5s的复杂场景下高效烟火检测算法YOLOv5s-MRD  

YOLOv5s-MRD:efficient fire and smoke detection algorithm for complex scenarios based on YOLOv5s

作　　者：侯阳 张琼[2] 赵紫煊 朱正宇 张晓博[2] HOU Yang;ZHANG Qiong;ZHAO Zixuan;ZHU Zhengyu;ZHANG Xiaobo(Department of Electronic Information Engineering,Chengdu Jincheng College,Chengdu Sichuan 611731,China;School of Computing and Artificial Intelligence,Southwest Jiaotong University,Chengdu Sichuan 611756,China)

机构地区：[1]成都锦城学院电子信息学院,成都611731 [2]西南交通大学计算机与人工智能学院,成都611756

出　　处：《计算机应用》2025年第4期1317-1324,共8页journal of Computer Applications

基　　金：国家自然科学基金资助项目(61976247);四川省重点研发计划项目(2023YFS0404)。

摘　　要：现有的烟火检测方法主要依赖员工现场巡视,效率低且实时性差,因此,提出一种基于YOLOv5s的复杂场景下的高效烟火检测算法YOLOv5s-MRD(YOLOv5s-MPDIoU-RevCol-Dyhead)。首先,采用MPDIoU(Maximized Position-Dependent Intersection over Union)方法改进边框损失函数,以适应重叠或非重叠的边界框回归(BBR),从而提高BBR的准确性和效率;其次,利用可逆柱状结构RevCol(Reversible Column)网络模型思想重构YOLOv5s模型的主干网络,使它具有多柱状网络架构,并在模型的不同层之间加入可逆链接,从而最大限度地保持特征信息以提高网络的特征提取能力;最后,引入Dynamic head检测头,以统一尺度感知、空间感知和任务感知,从而在不额外增加计算开销的条件下显著提高目标检测头的准确性和有效性。实验结果表明:在DFS(Data of Fire and Smoke)数据集上,与原始YOLOv5s算法相比,所提算法的平均精度均值(mAP@0.5)提升了9.3%,预测准确率提升了6.6%,召回率提升了13.8%。可见,所提算法能满足当前烟火检测应用场景的要求。Current fire and smoke detection methods mainly rely on site inspection by staff,which results in low efficiency and poor real-time performance,so an efficient fire and smoke detection algorithm for complex scenarios based on YOLOv5s,called YOLOv5s-MRD(YOLOv5s-MPDIoU-RevCol-Dyhead),was proposed.Firstly,the MPDIoU(Maximized Position-Dependent Intersection over Union)method was employed to modify the border loss function,thereby enhancing the accuracy and efficiency of Bounding Box Regression(BBR)by adapting to BBR in overlapping or non-overlapping scenarios.Secondly,the RevCol(Reversible Column)network model concept was applied to reconstruct the backbone of YOLOv5s,transforming it into a backbone network with multi-column network architecture.At the same time,by incorporating reversible links across various layers of the model,so that the retention of feature information was maximized,thereby improving the network’s feature extraction capability.Finally,with the integration of Dynamic head detection heads,scale awareness,spatial awareness,and task awareness were unified,thereby improving detection heads’accuracy and effectiveness significantly without additional computational cost.Experimental results demonstrate that on DFS(Data of Fire and Smoke)dataset,compared to the original YOLOv5s algorithm,the proposed algorithm achieves a 9.3%increase in mAP@0.5(mean Average Precision),a 6.6%improvement in prediction accuracy,and 13.8%increase in recall.It can be seen that the proposed algorithm can meet the requirements of current fire and smoke detection application scenarios.

关 键 词：目标检测 RevCol网络 YOLOv5 Dynamic head检测头 MPDIoU 烟火检测 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]