多尺度特征融合的轻量化口罩佩戴检测算法

doi:10.19678/j.issn.1000-3428.0062231

摘要/Abstract

摘要： 科学规范地佩戴口罩是预防新冠、流感等呼吸道传染病的有效方法，在当前疫情形势下，正确佩戴口罩显得尤为重要。已有的口罩佩戴检测算法多数存在结构复杂、训练难度较高和特征提取不足等问题，为此，提出一种多尺度特征融合的轻量化口罩佩戴检测算法L-MFFN-YOLO。以YOLOv4-Tiny网络为基础，L-MFFN-YOLO改进原始残差结构，使用轻量化残差模块促进模型快速收敛，在有效降低模型计算量的同时保证检测精度。在原网络、这2个尺度的基础上增加特征分支，以增强低特征层的信息表达能力并降低小目标的漏检率。通过多层级交叉融合结构最大程度地提取有用信息，从而提高特征利用率。除佩戴和未佩戴口罩2种情况外，在数据集中新增口罩佩戴不正确的类别并进行手工标注，实验结果表明，L-MFFN-YOLO算法的模型大小仅为5.8 MB，较原始网络YOLOv4-Tiny，其模型规模减小76%，mAP提高5.25个百分点，CPU下的处理时间快14 ms，能在资源受限的设备中满足口罩佩戴检测任务对准确率和实时性的要求。

关键词: 口罩佩戴检测, 轻量化检测算法, 残差结构, 低特征层, 多层级交叉融合

Abstract: Standardized usage of face masks is effective as a non-pharmaceutical intervention to prevent the spread of infectious respiratory diseases, such as COVID-19 and influenza.In the current epidemic situation, wearing face masks correctly is especially important.Most existing mask-wearing detection algorithms involve problems such as complex structures, high training difficulty, and insufficient feature extraction.Therefore, this study proposes a lightweight mask-wearing detection algorithm based on multi-scale feature fusion and the YOLOv4-Tiny network, called L-MFFN-YOLO.L-MFFN-YOLO improves on the original residual structure and uses a lightweight residual module to promote rapid convergence.Moreover, it reduces the computational load while ensuring detection accuracy. Based on the original network's 13×13 and 26×26 feature maps, 52×52 feature branches are added to enhance the ability of the lower feature layer to express information and reduce the false negative rate for small targets.On this basis, a Multi-level Cross Fusion(MCF) structure is used to maximally extract useful information so as to improve feature utilization.In addition to detecting mask-wearing, a category of masks worn incorrectly is added to the dataset and manually labeled.The experimental results show that the size of the proposed L-MFFN-YOLO model is only 5.8 MB, which is 76% smaller than that of the original YOLOv4-Tiny.Moreover, the mean Average Precision(mAP) of the proposed approach is 5.25 percentage points higher, and its processing time is 14 ms faster on an equivalent CPU.These results demonstrate that the proposed approach can meet the requirements of accuracy and real-time operation in resource-constrained devices to detect faces wearing masks.

Key words: mask-wearing detection, lightweight detection algorithm, residual structure, lower characteristic layer, Multi-level Cross Fusion(MCF)

中图分类号:

TP391

叶茂, 马杰, 王倩, 武麟. 多尺度特征融合的轻量化口罩佩戴检测算法[J]. 计算机工程, 2022, 48(7): 42-50.

YE Mao, MA Jie, WANG Qian, WU Lin. Lightweight Mask-Wearing Detection Algorithm with Multi-Scale Feature Fusion[J]. Computer Engineering, 2022, 48(7): 42-50.

http://www.ecice06.com/CN/Y2022/V48/I7/42

图/表 14

20220808083814

20220808083818

20220808083823

20220808083833

20220808083838

20220808083842

20220808083846

20220808083849

20220808083853

20220808083857

20220808083909

20220808083913

20220808083923

20220808083926

参考文献

[1] LIU Y, GAYLE A A, WILDER-SMITH A, et al.The reproductive number of COVID-19 is higher compared to SARS coronavirus[J].Journal of Travel Medicine, 2020, 27(2):1-4.
[2] REDMON J, FARHADI A.YOLOv3:an incremental improvement[EB/OL].[2021-07-05].https://arxiv.org/abs/1804.02767.
[3] 王艺皓, 丁洪伟, 李波, 等.复杂场景下基于改进YOLOv3的口罩佩戴检测算法[J].计算机工程, 2020, 46(11):12-22. WANG Y H, DING H W, LI B, et al.Mask wearing detection algorithm based on improved YOLOv3 in complex scenes[J].Computer Engineering, 2020, 46(11):12-22.(in Chinese)
[4] WANG C Y, MARK LIAO H Y, WU Y H, et al.CSPNet:a new backbone that can enhance learning capability of CNN[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:1571-1580.
[5] HE K M, ZHANG X Y, REN S Q, et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9):1904-1916.
[6] LIU S, QI L, QIN H F, et al.Path aggregation network for instance segmentation[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:8759-8768.
[7] 叶子勋, 张红英.YOLOv4口罩检测算法的轻量化改进[J].计算机工程与应用, 2021, 57(17):157-168. YE Z X, ZHANG H Y.Lightweight improvement of YOLOv4 mask detection algorithm[J].Computer Engineering and Applications, 2021, 57(17):157-168.(in Chinese)
[8] BOCHKOVSKIY A, WANG C Y, LIAO H Y M.YOLOv4:optimal speed and accuracy of object detection[EB/OL].[2021-07-05].https://arxiv.org/abs/2004.10934.
[9] HOWARD A, SANDLER M, CHU G, et al.Searching for MobileNetv3[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2019:1314-1324.
[10] 曹城硕, 袁杰.基于YOLO-Mask算法的口罩佩戴检测方法[J].激光与光电子学进展, 2021, 58(8):211-218. CAO C S, YUAN J.Mask-wearing detection method based on YOLO-Mask[J].Laser & Optoelectronics Progress, 2021, 58(8):211-218.(in Chinese)
[11] LIN T Y, DOLLÁR P, GIRSHICK R, et al.Feature pyramid networks for object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:936-944.
[12] 余阿祥, 李承润, 于书仪, 等.多注意力机制的口罩检测网络[J].南京师范大学学报(工程技术版), 2021, 21(1):23-29. YU A X, LI C R, YU S Y, et al.Multi-attention mechanism of mask wearing detection network[J].Journal of Nanjing Normal University(Engineering and Technology Edition), 2021, 21(1):23-29.(in Chinese)
[13] TAN M, PANG R, LE Q V, et al.EfficientDet:scalable and efficient object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:10781-10790.
[14] HU J, SHEN L, SUN G.Squeeze-and-excitation networks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:7132-7141.
[15] BODLA N, SINGH B, CHELLAPPA R, et al.Soft-NMS-improving object detection with one line of code[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:5562-5570.
[16] 彭成, 张乔虹, 唐朝晖, 等.基于YOLOv5增强模型的口罩佩戴检测方法研究[J].计算机工程, 2022, 48(4):39-49. PENG C, ZHANG Q H, TANG Z H, et al.Research on mask wearing detection method based on YOLOv5 enhancement model[J].Computer Engineering, 2022, 48(4):39-49.(in Chinese)
[17] HAN K, WANG Y H, TIAN Q, et al.GhostNet:more features from cheap operations[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:1577-1586.
[18] MA N N, ZHANG X Y, ZHENG H T, et al.ShuffleNet V2:practical guidelines for efficient CNN architecture design[EB/OL].[2021-07-05].https://arxiv.org/pdf/1807.11164.pdf.
[19] 何涛, 俞舒曼, 徐鹤.基于条件生成对抗网络与知识蒸馏的单幅图像去雾方法[J].计算机工程, 2022, 48(4):165-172. HE T, YU S M, XU H.Single image dehazing method based on conditional generative adversarial network and knowledge distillation[J].Computer Engineering, 2022, 48(4):165-172.(in Chinese)
[20] FAN S T, ZHANG X M, SONG Z H.Reinforced knowledge distillation:multi-class imbalanced classifier based on policy gradient reinforcement learning[J].Neurocomputing, 2021, 463:422-436.
[21] 曹远杰, 高瑜翔.基于GhostNet残差结构的轻量化饮料识别网络[J].计算机工程, 2022, 48(3):310-314. CAO Y J, GAO Y X.Lightweight beverage recognition network based on GhostNet residual structure[J].Computer Engineering, 2022, 48(3):310-314.(in Chinese)
[22] SANDLER M, HOWARD A, ZHU M L, et al.MobileNetV2:inverted residuals and linear bottlenecks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:4510-4520.
[23] SUN K, XIAO B, LIU D, et al.Deep high-resolution representation learning for human pose estimation[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2019:5686-5696.
[24] REN S Q, HE K M, GIRSHICK R, et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.
[25] HOWARD A G, ZHU M L, CHEN B, et al.MobileNets:efficient convolutional neural networks for mobile vision applications[EB/OL].[2021-07-05].https://arxiv.org/abs/1704.04861.

选择文件类型/文献管理软件名称

选择包含的内容