[1]应雨洁,丁飞,余耀元,等.M-YOLO:基于多尺度特征融合的轻量级恶劣天气目标检测模型[J].智能系统学报,2026,21(5):1292-1299.[doi:10.11992/tis.202601011]
YING Yujie,DING Fei,YU Yaoyuan,et al.M-YOLO: a lightweight multi-scale feature fusion-based model for adverse weather object detection[J].CAAI transactions on intelligent systems,2026,21(5):1292-1299.[doi:10.11992/tis.202601011]
点击复制
YING Yujie,DING Fei,YU Yaoyuan,et al.M-YOLO: a lightweight multi-scale feature fusion-based model for adverse weather object detection[J].CAAI transactions on intelligent systems,2026,21(5):1292-1299.[doi:10.11992/tis.202601011]
M-YOLO:基于多尺度特征融合的轻量级恶劣天气目标检测模型
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
21
期数:
2026年第5期
页码:
1292-1299
栏目:
学术论文—智能系统
出版日期:
2026-09-05
- Title:
- M-YOLO: a lightweight multi-scale feature fusion-based model for adverse weather object detection
- Keywords:
- YOLOv5; target detection; multi-access edge computing; attention mechanism; intelligent transportation; image enhancement; lightweight model; adverse weather
- 分类号:
- TP391.41
- 摘要:
- 为提高复杂气象环境下智能交通系统的交通目标识别精度，提出基于多接入边缘计算(multi-access edge computing，MEC)架构的增强型YOLOv5模型(a lightweight multi-scale feature fusion-based model for adverse weather object detection，M-YOLO)。设计双阶段图像增强机制，实现雾/雨/雪多源气象特征的融合；引入通道与空间协同的卷积块注意力机制(convolutional block attention module，CBAM)以提升模型在复杂背景下的目标识别和定位能力；在YOLOv5架构基础上新增P2/4小目标检测层，提高对小尺寸目标的检测精度；同时，M-YOLO还采用了MPDIoU(minimum point distance intersection over union)损失函数优化边界框回归。最后，通过Foggy-Cityscapes、Rain-Cityscapes和Snow-Cityscapes数据集进行实验验证，结果表明M-YOLO均表现出更优的平均检测精度。在混合天气条件下，M-YOLO模型的mAP50达到0.485，相比于基准模型(0.415)大幅提高，为边缘计算环境下的实时交通监控提供了高效解决方案。
- Abstract:
- To improve the traffic target recognition accuracy of intelligent transportation systems in complex meteorological environments, an enhanced YOLOv5 model (M-YOLO) based on multi-access edge computing (MEC) architecture is proposed. A two-stage image enhancement mechanism was designed to realize the fusion of fog/rain/snow multi-source meteorological features; a channel and spatial synergistic Convolutional Block Attention Mechanism (CBAM) was introduced to enhance the model’s target recognition and localization ability in complex backgrounds; a P2/4 small-target detection layer was added based on the YOLOv5 architecture to improve the detection accuracy of small-sized targets; meanwhile, M-YOLO also adopted the MPDIoU loss function to optimize the bounding box regression. Finally, experimental validation was performed with Foggy-Cityscapes, Rain-Cityscapes, and Snow-Cityscapes datasets, and the results show that M-YOLO exhibits better average detection accuracy. Under mixed weather conditions, the mAP50 of the M-YOLO model reaches 0.485, which represents a significant improvement over the benchmark model (0.415), providing an efficient solution for real-time traffic monitoring in edge computing environments.
- 参考文献/References:
-
[1] HU Yunchao, PATEL M, SABELLA D, et al. Mobile edge computing—a key technology towards 5G[J]. ETSI white paper, 2015, 11(11): 1-16
[2] YANG H F, CAI Jiarui, LIU Chenxi, et al. Cooperative multi-camera vehicle tracking and traffic surveillance with edge artificial intelligence and representation learning[J]. Transportation Research Part C: Emerging Technologies, 2023, 148: 103982.
[3] MCMAHAN H B, MOORE E, RAMAGE D, et al. Communication-efficient learning of deep networks from decentralized data[C]//Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. Fort Lauderdale: PMLR, 2017, 54: 1273–1282.
[4] 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. Las Vegas: IEEE, 2016: 779–788.
[5] GAO Chenqiang, MENG Deyu, YANG Yi, et al. Infrared patch-image model for small target detection in a single image[J]. IEEE transactions on image processing, 2013, 22(12): 4996-5009
[6] CHEN Weiting, HUANG Zhikai, TSAI C C, et al. Learning multiple adverse weather removal via two-stage knowledge learning and multi-contrastive regularization: toward a unified model[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 17632–17641.
[7] FU Ying, HONG Yang, CHEN Linwei, et al. LE-GAN: Unsupervised low-light image enhancement network using attention module and identity invariant loss[J]. Knowledge-based systems, 2022, 240: 108010
[8] JOSE VALANARASU J M, YASARLA R, PATEL V M. TransWeather: transformer-based restoration of images degraded by adverse weather conditions[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022: 2343–2353.
[9] 王新蕾, 王硕, 翟嘉政, 等. 多任务联合学习下的复杂天气航拍图像目标检测算法[J]. 计算机工程与应用, 2025, 61(2): 97-111 WANG Xinlei, WANG Shuo, ZHAI Jiazheng, et al. Object detection algorithm of aerial image in complex weather based on multi-task joint learning[J]. Computer engineering and applications, 2025, 61(2): 97-111
[10] 董中华, 刘恒. 基于域适应的真实场景图像超分辨方法[J]. 计算机科学与应用, 2023, 13(6): 1279-1288 DONG Zhonghua, LIU Heng. Real-world image super-resolution method based on domain adaptation[J]. Computer science and application, 2023, 13(6): 1279-1288
[11] 王珏, 洪敏轩, 夏叶桐, 等. 基于条件生成对抗网络的低照度彩色图像增强算法[J]. 红外与激光工程, 2024, 53(11): 20240238 WANG Jue, HONG Minxuan, XIA Yetong, et al. Low-light color image enhancement algorithm based on conditional generative adversarial network[J]. Infrared and laser engineering, 2024, 53(11): 20240238
[12] HU Jie, SHEN Li, SUN Gang. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 7132–7141.
[13] HU Jie, SHEN Li, ALBANIE S, et al. Gather-excite: exploiting feature context in convolutional neural networks[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems. Montréal: Curran Associates Inc., 2018: 9423–9433.
[14] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Computer Vision–ECCV 2018. Cham: Springer, 2018: 3–19.
[15] 许云峰, 刘畅达, 梅卫, 等. 一种轻量级空间位置注意力模块及其在图像分类网络中的应用[J]. 信息与控制, 2025, 54(2): 321-335, 352 XU Yunfeng, LIU Changda, MEI Wei, et al. Lightweight spatial location attention module and its application to image classification networks[J]. Information and control, 2025, 54(2): 321-335, 352
[16] HINTON G, VINYALS O, DEAN J. Distilling the knowledge in a neural network[EB/OL]. (2015–03–09)[2026–01–06]. https://doi.org/10.48550/arXiv.1503.02531.
[17] ZHAO Liang, WEI Zhengjie, LI Yanting, et al. SEDG-Yolov5: a lightweight traffic sign detection model based on knowledge distillation[J]. Electronics, 2023, 12(2): 305
[18] DONG Chenhang, PANG Chengxin, LI Zhengwei, et al. PG-YOLO: a novel lightweight object detection method for edge devices in industrial Internet of Things[J]. IEEE access, 2022, 10: 123736-123745
[19] 李昂, 孙士杰, 张朝阳, 等. 改进YOLOv5s的轨道障碍物检测模型轻量化研究[J]. 计算机工程与应用, 2023, 59(4): 197-207 LI Ang, SUN Shijie, ZHANG Zhaoyang, et al. Research on lightweight of improved YOLOv5s track obstacle detection model[J]. Computer engineering and applications, 2023, 59(4): 197-207
[20] 杨睿宁, 惠飞, 金鑫, 等. 改进YOLOv5s的复杂交通场景路侧目标检测算法[J]. 计算机工程与应用, 2023, 59(16): 159-169 YANG Ruining, HUI Fei, JIN Xin, et al. Roadside target detection algorithm for complex traffic scene based on improved YOLOv5s[J]. Computer engineering and applications, 2023, 59(16): 159-169
[21] 李子姝, 谢人超, 孙礼, 等. 移动边缘计算综述[J]. 电信科学, 2018(1): 87-101 LI Zishu, XIE Renchao, SUN Li, et al. A survey of mobile edge computing[J]. Telecommunications science, 2018(1): 87-101
[22] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2015–04–10)[2026–01–06]. https://doi.org/10.48550/arXiv.1409.1556.
[23] VAN DEN OORD A, LI Yazhe, VINYALS O. Representation learning with contrastive predictive coding[EB/OL]. (2019–01–22)[2026–01–06]. https://doi.org/10.48550/arXiv.1807.03748.
[24] SAITO K, USHIKU Y, HARADA T, et al. Strong-weak distribution alignment for adaptive object detection[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 6949–6958.
[25] MA Siliang, XU Yong. MPDIoU: a loss for efficient and accurate bounding box regression[EB/OL]. (2023–07–14)[2026–01–06]. https://doi.org/10.48550/arXiv.2307.07662.
[26] SAKARIDIS C, DAI Dengxin, VAN GOOL L. Semantic foggy scene understanding with synthetic data[J]. International journal of computer vision, 2018, 126(9): 973-992
[27] CORDTS M, OMRAN M, RAMOS S, et al. The cityscapes dataset for semantic urban scene understanding[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 3213–3223.
[28] SINDAGI V A, OZA P, YASARLA R, et al. Prior-based domain adaptive object detection for hazy and rainy conditions[C]//Computer Vision–ECCV 2020. Cham: Springer, 2020: 763–780.
[29] WANG Lucai, QIN Hongda, ZHOU Xuanyu, et al. R-YOLO: a robust object detector in adverse weather[J]. IEEE transactions on instrumentation and measurement, 2023, 72: 5000511
[30] ZHAO Ganlong, LI Guanbin, XU Ruijia, et al. Collaborative training between region proposal localization and classification for domain adaptive object detection[C]//Computer Vision–ECCV 2020. Cham: Springer, 2020: 86–102.
[31] WANG Kunpeng, CAI Jingxiang, YAO Juan, et al. Co-teaching based pseudo label refinery for cross-domain object detection[J]. IET image processing, 2021, 15(13): 3189-3199
- 相似文献/References:
-
[1]胡光龙,秦世引.动态成像条件下基于SURF和Mean shift的运动目标高精度检测[J].智能系统学报,2012,7(1):61.
HU Guanglong,QIN Shiyin.High precision detection of a mobile object under dynamic imaging based on SURF and Mean shift[J].CAAI transactions on intelligent systems,2012,7():61.
[2]韩峥,刘华平,黄文炳,等.基于Kinect的机械臂目标抓取[J].智能系统学报,2013,8(2):149.[doi:10.3969/j.issn.1673-4785.201212038]
HAN Zheng,LIU Huaping,HUANG Wenbing,et al.Kinect-based object grasping by manipulator[J].CAAI transactions on intelligent systems,2013,8():149.[doi:10.3969/j.issn.1673-4785.201212038]
[3]韩延彬,郭晓鹏,魏延文,等.RGB和HSI颜色空间的一种改进的阴影消除算法[J].智能系统学报,2015,10(5):769.[doi:10.11992/tis.201410010]
HAN Yanbin,GUO Xiaopeng,WEI Yanwen,et al.An improved shadow removal algorithm based on RGB and HSI color spaces[J].CAAI transactions on intelligent systems,2015,10():769.[doi:10.11992/tis.201410010]
[4]曾宪华,易荣辉,何姗姗.流形排序的交互式图像分割[J].智能系统学报,2016,11(1):117.[doi:10.11992/tis.201505037]
ZENG Xianhua,YI Ronghui,HE Shanshan.Interactive image segmentation based on manifold ranking[J].CAAI transactions on intelligent systems,2016,11():117.[doi:10.11992/tis.201505037]
[5]葛园园,许有疆,赵帅,等.自动驾驶场景下小且密集的交通标志检测[J].智能系统学报,2018,13(3):366.[doi:10.11992/tis.201706040]
GE Yuanyuan,XU Youjiang,ZHAO Shuai,et al.Detection of small and dense traffic signs in self-driving scenarios[J].CAAI transactions on intelligent systems,2018,13():366.[doi:10.11992/tis.201706040]
[6]莫宏伟,汪海波.基于Faster R-CNN的人体行为检测研究[J].智能系统学报,2018,13(6):967.[doi:10.11992/tis.201801025]
MO Hongwei,WANG Haibo.Research on human behavior detection based on Faster R-CNN[J].CAAI transactions on intelligent systems,2018,13():967.[doi:10.11992/tis.201801025]
[7]宁欣,李卫军,田伟娟,等.一种自适应模板更新的判别式KCF跟踪方法[J].智能系统学报,2019,14(1):121.[doi:10.11992/tis.201806038]
NING Xin,LI Weijun,TIAN Weijuan,et al.Adaptive template update of discriminant KCF for visual tracking[J].CAAI transactions on intelligent systems,2019,14():121.[doi:10.11992/tis.201806038]
[8]伍鹏瑛,张建明,彭建,等.多层卷积特征的真实场景下行人检测研究[J].智能系统学报,2019,14(2):306.[doi:10.11992/tis.201710019]
WU Pengying,ZHANG Jianming,PENG Jian,et al.Research on pedestrian detection based on multi-layer convolution feature in real scene[J].CAAI transactions on intelligent systems,2019,14():306.[doi:10.11992/tis.201710019]
[9]刘召,张黎明,耿美晓,等.基于改进的Faster R-CNN高压线缆目标检测方法[J].智能系统学报,2019,14(4):627.[doi:10.11992/tis.201905026]
LIU Zhao,ZHANG Liming,GENG Meixiao,et al.Object detection of high-voltage cable based on improved Faster R-CNN[J].CAAI transactions on intelligent systems,2019,14():627.[doi:10.11992/tis.201905026]
[10]单义,杨金福,武随烁,等.基于跳跃连接金字塔模型的小目标检测[J].智能系统学报,2019,14(6):1144.[doi:10.11992/tis.201905041]
SHAN Yi,YANG Jinfu,WU Suishuo,et al.Skip feature pyramid network with a global receptive field for small object detection[J].CAAI transactions on intelligent systems,2019,14():1144.[doi:10.11992/tis.201905041]
[11]白健鹏,王巍,陈雨溪,等.基于轻量型YOLOv5的风机桨叶检测与空间定位[J].智能系统学报,2022,17(6):1173.[doi:10.11992/tis.202204016]
BAI Jianpeng,WANG Wei,CHEN Yuxi,et al.Detection and spatial location of wind turbine blades based on lightweight YOLOv5[J].CAAI transactions on intelligent systems,2022,17():1173.[doi:10.11992/tis.202204016]
[12]冯晗,姜勇.使用改进Yolov5的变电站绝缘子串检测方法[J].智能系统学报,2023,18(2):325.[doi:10.11992/tis.202201027]
FENG Han,JIANG Yong.A substation insulator string detection method based on an improved Yolov5[J].CAAI transactions on intelligent systems,2023,18():325.[doi:10.11992/tis.202201027]
备注/Memo
收稿日期:2026-1-6。
基金项目:工业和信息化部产业技术基础公共服务平台项目(2019-00892-3-1)；江苏省高等学校基础科学(自然科学)研究重大项目(25KJA510003)；江苏省研究生科研创新计划项目(SJCX24_0336)；江苏省大学生创新创业训练计划项目(202410293166Y).
作者简介:应雨洁，硕士研究生，主要研究方向为车联网、智能网联交通。E-mail：b22080607@njupt.edu.cn。;丁飞，教授，博士，主要研究方向为群智感知计算、混合信息物理系统。E-mail：dingfei@njupt.edu.cn。;余耀元，硕士研究生，主要研究方向为人工智能与智能控制。E-mail：1822393465@qq.com。
通讯作者:丁飞. E-mail：dingfei@njupt.edu.cn
更新日期/Last Update:
2026-09-05