雾天条件下改进YOLOv4的目标检测

doi:10.16182/j.issn1004731x.joss.22-0423

系统仿真学报 ›› 2023, Vol. 35 ›› Issue (8): 1681-1691.doi: 10.16182/j.issn1004731x.joss.22-0423

雾天条件下改进YOLOv4的目标检测

刘书刚(), 张林坤(), 杜昊东, 王洪涛

华北电力大学，河北保定 071000

收稿日期:2022-04-27 修回日期:2022-07-18 出版日期:2023-08-25 发布日期:2023-08-15
作者简介:
刘书刚(1969-)，男，副教授，硕士，研究方向为人工智能。E-mail：lsg69@qq.com
基金资助:
国家自然科学基金(61802124)

Improved Object Detection of YOLOv4 in Foggy Conditions

Shugang Liu(), Linkun Zhang(), Haodong Du, Hongtao Wang

North China Electric Power University, Baoding 071000, China

Received:2022-04-27 Revised:2022-07-18 Online:2023-08-25 Published:2023-08-15

摘要/Abstract

关键词: 生成对抗网络, DeblurGANv2, Coordinate Attention, ShuffleNet V2, 去雾目标检测

Key words: generative adversarial nets, DeblurGANv2, Coordinate Attention, ShuffleNet V2, defogging target detection

刘书刚, 张林坤, 杜昊东, 王洪涛. 雾天条件下改进YOLOv4的目标检测[J]. 系统仿真学报, 2023, 35(8): 1681-1691.

Shugang Liu, Linkun Zhang, Haodong Du, Hongtao Wang. Improved Object Detection of YOLOv4 in Foggy Conditions[J]. Journal of System Simulation, 2023, 35(8): 1681-1691.

图/表 15

图1

图2

图3

图4

图5

图6

表1

图7

图8

表2

图9

图10

图11

图12

表3

参考文献 23

1	Liu Wei, Anguelov D, Erhan D, et al. SSD: Single Shot MultiBox Detector[C]//Computer Vision-ECCV 2016. Cham: Springer International Publishing, 2016: 21-37.
2	Redmon J, Divvala S, Girshick R, et al. You Only Look Once: Unified, Real-time Object Detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2016: 779-788.
3	Redmon J, Farhadi A. YOLO9000: Better, Faster, Stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2017: 6517-6525.
4	Redmon J, Farhadi A. YOLOv3: An Incremental Improvement[EB/OL]. (2018-04-08)[2022-05-24].
5	Bochkovskiy A, Wang C Y, Liao Hongyuan. YOLOv4: Optimal Speed and Accuracy of Object Detection[EB/OL]. (2020-04-23)[2022-06-04].
6	Lin T Y, Goyal P, Girshick R, et al. Focal Loss for Dense Object Detection[C]//2017 IEEE International Conference on Computer Vision(ICCV). Piscataway, NJ, USA: IEEE, 2017: 2999-3007.
7	Lin T Y, Dollár Piotr, Girshick R, et al. Feature Pyramid Networks for Object Detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2017: 936-944.
8	Girshick R, Donahue J, Darrell T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2014: 580-587.
9	Li Chongyi, Guo Chunle, Guo Jichang, et al. PDR-net: Perception-inspired Single Image Dehazing Network With Refinement[J]. IEEE Transactions on Multimedia, 2020, 22(3): 704-716.
10	Ren Shaoqing, He Kaiming, 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.
11	Li Boyi, Peng Xiulian, Wang Zhangyang, et al. AOD-net: All-in-one Dehazing Network[C]//2017 IEEE International Conference on Computer Vision(ICCV). Piscataway, NJ, USA: IEEE, 2017: 4780-4788.
12	Li Boyi, Ren Wenqi, Fu Dengpan, et al. Benchmarking Single Image Dehazing and Beyond[J]. IEEE Transactions on Image Processing, 2019, 28(1): 492-505.
13	Liu Yu, Zhao Guanlong, Gong Boyuan, et al. Improved Techniques for Learning to Dehaze and Beyond: A Collective Study[EB/OL]. (2018-07-30)[2022-06-06].
14	Yang Wenhan, Yuan Ye, Ren Wenqi, et al. Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study[J]. IEEE Transactions on Image Processing, 2020, 29: 5737-5752.
15	Kupyn O, Martyniuk T, Wu Junru, et al. DeblurGAN-v2: Deblurring (Orders-of-magnitude) Faster and Better[C]//2019 IEEE/CVF International Conference on Computer Vision(ICCV). Piscataway, NJ, USA: IEEE, 2019: 8877-8886.
16	Ma Ningning, Zhang Xiangyu, Zheng Haitao, et al. ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design[C]//Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 122-138.
17	Kupyn O, Budzan V, Mykhailych M, et al. DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 8183-8192.
18	Zhang Xiangyu, Zhou Xinyu, Lin Mengxiao, et al. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 6848-6856.
19	Hu Jie, Shen Li, Sun Gang. Squeeze-and-excitation Networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 7132-7141.
20	Woo S, Park J, Lee J Y, et al. CBAM: Convolutional Block Attention Module[C]//Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 3-19.
21	Hou Qibin, Zhou Daquan, Feng Jiashi. Coordinate Attention for Efficient Mobile Network Design[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2021: 13708-13717.
22	He Kaiming, Sun Jian, Tang Xiaoou. Single Image Haze Removal Using Dark Channel Prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.
23	李梦蕊, 柳晓鸣, 常婧. 基于颜色衰减先验的自适应Retinex去雾算法[J]. 计算机仿真, 2021, 38(1): 354-358.
	Li Mengrui, Liu Xiaoming, Chang Jing. An Adaptive Retinex of Image Haze Removal Based on Color Attenuation Prior[J]. Computer Simulation, 2021, 38(1): 354-358.

数据集	交通信号灯	卡车	人	汽车	自行车
合计	8 422	7 123	10 825	14 413	298
训练集	7 586	6 458	9 632	12 561	263
测试集	836	665	1 193	1 552	35

图像	指数	暗通道先验法	自适应 Retinex	DeblurGANv2
住宅	PSNR/dB	23.70	26.68	28.78
住宅	SSIM/%	69	74	76
公路	PSNR/dB	23.50	26.09	28.23
公路	SSIM/%	68	74	75
广场	PSNR/dB	23.10	26.61	27.54
广场	SSIM/%	63	70	72

算法	mAP/%	AP/%					检测速度/(帧/s)
算法	mAP/%	交通信号灯	卡车	人	汽车	自行车	检测速度/(帧/s)
YOLOv3-暗通道	76.54	67.21	81.56	82.98	82.50	68.45	35
YOLOv3-自适应Retinex	80.38	69.95	85.23	87.96	88.34	70.45	33
YOLOv3-DeblurGANv2	81.69	71.23	85.27	86.54	84.78	81.63	34
YOLOv4-暗通道	84.45	73.12	90.36	91.71	90.7	76.36	41
YOLOv4-自适应Retinex	85.53	74.95	91.45	90.54	91.29	79.42	39
YOLOv4-DeblurGANv2	87.52	77.86	90.18	92.86	93.05	83.65	40
本文算法	86.56	75.26	90.17	92.08	91.88	83.41	46

[1]	蔡兴泉, 李治均, 奚梦瑶, 孙海燕. 基于神经网络的手绘服饰图纹上色及风格迁移[J]. 系统仿真学报, 2023, 35(3): 604-615.
[2]	曹建芳, 贾一鸣, 闫敏敏, 田晓东. 稳定增强生成对抗网络在壁画的超分辨率重建[J]. 系统仿真学报, 2022, 34(5): 1076-1089.
[3]	吴曦, 孟祥林, 杨镜宇. 下一代战略博弈推演系统研究[J]. 系统仿真学报, 2021, 33(9): 2017-2024.
[4]	程文聪, 史小康, 王志刚. 基于生成对抗网络的仿真卫星云图生成方法[J]. 系统仿真学报, 2021, 33(6): 1297-1306.

雾天条件下改进YOLOv4的目标检测

Improved Object Detection of YOLOv4 in Foggy Conditions

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 23

相关文章 4

编辑推荐

Metrics

本文评价