摘要
针对复杂场景下红外目标检测存在准确率低、召回率低的问题,为了提高红外图像中的小目标以及被遮挡目标的检测识别能力,提出基于Efficientnet的红外目标检测算法。首先,将高效轻量的Efficientnet作为模型的特征提取主干网,降低模型的参数量,提升训练速度。在Efficientnet主干网的最后一个输出层引入SPP模块,丰富特征图的表达能力,进行多尺度融合,扩大特征图的感受野;在模型特征融合部分,使用FPN特征金字塔网络,特征融合后增加CSPNet模块和ECA注意力机制,加强特征提取。检测部分使用YOLO Head,对目标进行分类和回归,并用CIoU Loss作为边界框回归损失函数,提高对被遮挡目标的识别能力。实验结果表明,基于Efficientnet的模型大小仅为YOLOv3的18.8%,并且在FLIR数据集上mAP达到80.74%,相比于YOLOv3算法提高10.12%,该模型在减少模型参数量的同时,提升了检测精度。该模型在FLIR数据集上具有良好的泛化能力,提高了对小目标和遮挡目标的检测能力。
In order to improve the detection and recognition of small targets and obscured targets in infrared images,the Efficientnet-based infrared target detection algorithm is proposed for the problem of low accuracy and low recall of infrared target detection in complex scenes.First,the efficient and lightweight Efficientnet is used as the feature extraction backbone of the model to reduce the number of parameters of the model and improve the training speed.In the last output layer of Efficientnet backbone,SPP module is introduced to enrich the expression capability of feature map,perform multi-scale fusion and expand the perceptual field of feature map;in the feature fusion part of the model,FPN feature pyramid network is used,and CSPNet module and ECA attention mechanism are added after feature fusion to enhance feature extraction.The detection part uses YOLO Head to classify and regress the targets,and uses CIoU Loss as the bounding box regression loss function to improve the recognition ability of the obscured targets.The experimental results show that the Efficientnet-based model is only 18.8%of the size of YOLOv3,and the mAP reaches 80.74%on the FLIR dataset,which is 10.12%better than the YOLOv3algorithm,and the model improves the detection accuracy while reducing the number of model parameters.The model has good generalization ability on the FLIR dataset and improves the detection of small and occluded targets.
作者
侯艳丽
王娟
Hou Yanli;Wang Juan(School of Information Science and Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China)
出处
《电子测量技术》
北大核心
2023年第16期64-72,共9页
Electronic Measurement Technology
基金
河北省重点研发计划(21355901D)项目资助
