1.桂林电子科技大学 电子工程与自动化学院,广西 桂林 541004
2.桂林电子科技大学 智能综合自动化广西高校重点实验室,广西 桂林 541004
[ "伍锡如(1981-),男,湖南新化人,博士, 教授,博士生导师,2007年、2012年于湖南大学分别获得硕士和博士学位,主要从事机器人视觉与环境感知、多机器人的协调与编队控制、深度学习与模式识别方面的研究。E-mail: xiru@guet.edu.cn" ]
[ "郝家琦(2001-),男,山西长治人,硕士研究生,2023年于华北科技学院获得学士学位,主要研究方向为深度学习、目标检测、图像处理和计算机视觉方面的研究。E-mail: haojq@mails.guet.edu.cn" ]
收稿:2025-06-18,
修回:2025-06-27,
纸质出版:2025-10-10
移动端阅览
伍锡如,郝家琦,赵一波等.复杂天气条件下基于YOLO-CGT的自动驾驶车辆检测[J].光学精密工程,2025,33(19):3135-3149.
WU Xiru,HAO Jiaqi,ZHAO Yibo,et al.Autonomous vehicle detection in complex weather based on YOLO-CGT[J].Optics and Precision Engineering,2025,33(19):3135-3149.
伍锡如,郝家琦,赵一波等.复杂天气条件下基于YOLO-CGT的自动驾驶车辆检测[J].光学精密工程,2025,33(19):3135-3149. DOI: 10.37188/OPE.20253319.3135. CSTR: 32169.14.OPE.20253319.3135.
WU Xiru,HAO Jiaqi,ZHAO Yibo,et al.Autonomous vehicle detection in complex weather based on YOLO-CGT[J].Optics and Precision Engineering,2025,33(19):3135-3149. DOI: 10.37188/OPE.20253319.3135. CSTR: 32169.14.OPE.20253319.3135.
针对复杂天气条件下检测车辆目标时,因存在目标模糊及遮挡造成车辆检测精度明显下降的现象,提出一种改进YOLOv8的车辆检测算法YOLO-CGT。该算法面向车载摄像头图像输入场景,通过在YOLOv8结构中引入多项改进,显著提升了在复杂环境下的检测稳定性。其中,设计多尺度残差聚合模块替换原有主干网络结构中的C2f结构,用于增强原始信息的利用并减少网络深度带来的梯度消失问题;引入空间聚合模块,融合全局信息提取和局部信息感知;设计轻量级动态检测头,保证检测精度和
效率的平衡;引入内最小点距离交并比(Inner-Minimum Points Distance Intersection over Union,Inner-MPDIoU)度量替换传统IoU,以减少目标框重叠问题。在复杂天气条件下的车辆数据集上进行训练和验证后,实验结果显示,该方法的平均检测精度达到81.4%,提升了6.3%,模型参数量为3.259
<math id="M1"><mo>×</mo></math>
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=92634679&type=
https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=92634664&type=
2.20133328
2.37066650
10
6
,计算量为9.7GFLOPs,在精度显著提升的同时保证了模型的轻量部署能力。该研究方法为自动驾驶系统的安全稳定运行提供了有力保障。
To mitigate the pronounced decline in vehicle detection performance caused by object blur and occlusion under adverse weather conditions, an enhanced YOLOv8-based vehicle detection algorithm, designated YOLO-CGT, is proposed. Tailored for vehicle-mounted camera imagery, the algorithm incorporates multiple enhancements to the YOLOv8 architecture to substantially improve detection robustness in challenging environments. Specifically, a multi-scale residual aggregation module replaces the original C2f module in the backbone network to increase exploitation of raw feature information and to alleviate gradient vanishing associated with greater network depth. A spatial aggregation module is incorporated to integrate global information extraction with local feature perception. Moreover, a lightweight dynamic detection head is developed to balance detection accuracy and computational efficiency. The conventional IoU metric is supplanted by the Inner-Minimum Points Distance Intersection over Union (Inner-MPDIoU) to reduce bounding-box overlap issues. Trained and validated on a vehicle dataset captured under complex weather conditions, the proposed method attains an average detection accuracy of 81.4%-an imp
rovement of 6.3%-with 3.259×10
6
model parameters and a computational cost of 9.7 GFLOPs, demonstrating suitability for lightweight deployment while delivering substantial accuracy gains. These results provide a robust foundation for the safe and reliable operation of autonomous driving systems.
宗长富 , 代昌华 , 张东 . 智能汽车的人机共驾技术研究现状和发展趋势 [J]. 中国公路学报 , 2021 , 34 ( 6 ): 214 - 237 .
ZONG CH F , DAI CH H , ZHANG D . Human-machine interaction technology of intelligent vehicles: current development trends and future directions [J]. China Journal of Highway and Transport , 2021 , 34 ( 6 ): 214 - 237 . (in Chinese)
范丽丽 , 赵宏伟 , 赵浩宇 , 等 . 基于深度卷积神经网络的目标检测研究综述 [J]. 光学 精密工程 , 2020 , 28 ( 5 ): 1152 - 1164 .
FAN L L , ZHAO H W , ZHAO H Y , et al . Survey of target detection based on deep convolutional neural networks [J]. Optics and Precision Engineering , 2020 , 28 ( 5 ): 1152 - 1164 . (in Chinese)
王晶 , 黄智鑫 , 徐京邦 , 等 . 面向地下环境机器人的多模态目标检测方法 [J]. 机器人 , 2025 , 47 ( 4 ): 537 - 547 .
WANG J , HUANG ZH X , XU J B , et al . A multimodal object detection method for robots in underground environments [J]. Robot , 2025 , 47 ( 4 ): 537 - 547 . (in Chinese)
WANG Z G , ZHAN J , DUAN C G , et al . A review of vehicle detection techniques for intelligent vehicles [J]. IEEE Transactions on Neural Networks and Learning Systems , 2023 , 34 ( 8 ): 3811 - 3831 . doi: 10.1109/tnnls.2021.3128968 http://dx.doi.org/10.1109/tnnls.2021.3128968
王鑫 , 李鹏程 , 崔海华 , 等 . 面向机翼线缆支架的装配符合性视觉检测 [J]. 光学 精密工程 , 2025 , 33 ( 7 ): 1130 - 1140 . doi: 10.37188/ope.20253307.1130 http://dx.doi.org/10.37188/ope.20253307.1130
WANG X , LI P CH , CUI H H , et al . Visual inspection for assembly conformity of wing cable brackets [J]. Optics and Precision Engineering , 2025 , 33 ( 7 ): 1130 - 1140 . (in Chinese) . doi: 10.37188/ope.20253307.1130 http://dx.doi.org/10.37188/ope.20253307.1130
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 . doi: 10.1109/tpami.2016.2577031 http://dx.doi.org/10.1109/tpami.2016.2577031
BHARATI P , PRAMANIK A . Deep learning techniques: R-CNN to mask R-CNN: a survey [C]. Computational Intelligence in Pattern Recognition. Singapore : Springer , 2020 : 657 - 668 . doi: 10.1007/978-981-13-9042-5_56 http://dx.doi.org/10.1007/978-981-13-9042-5_56
CAI Z W , VASCONCELOS N . Cascade R-CNN: high quality object detection and instance segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2021 , 43 ( 5 ): 1483 - 1498 . doi: 10.1109/tpami.2019.2956516 http://dx.doi.org/10.1109/tpami.2019.2956516
JI H , GAO Z , MEI T C , et al . Improved faster R-CNN with multiscale feature fusion and homography augmentation for vehicle detection in remote sensing images [J]. IEEE Geoscience and Remote Sensing Letters , 2019 , 16 ( 11 ): 1761 - 1765 . doi: 10.1109/lgrs.2019.2909541 http://dx.doi.org/10.1109/lgrs.2019.2909541
石杰 , 周亚丽 , 张奇志 . 基于改进Mask RCNN和Kinect的服务机器人物品识别系统 [J]. 仪器仪表学报 , 2019 , 40 ( 4 ): 216 - 228 .
SHI J , ZHOU Y L , ZHANG Q ZH . Service robot item recognition system based on improved Mask RCNN and Kinect [J]. Chinese Journal of Scientific Instrument , 2019 , 40 ( 4 ): 216 - 228 . (in Chinese)
李松江 , 吴宁 , 王鹏 , 等 . 基于改进Cascade RCNN的车辆目标检测方法 [J]. 计算机工程与应用 , 2021 , 57 ( 5 ): 123 - 130 .
LI S J , WU N , WANG P , et al . Vehicle target detection method based on improved cascade RCNN [J]. Computer Engineering and Applications , 2021 , 57 ( 5 ): 123 - 130 . (in Chinese)
蒋弘毅 , 王永娟 , 康锦煜 . 目标检测模型及其优化方法综述 [J]. 自动化学报 , 2021 , 47 ( 6 ): 1232 - 1255 . doi: 10.16383/j.aas.c190756 http://dx.doi.org/10.16383/j.aas.c190756
JIANG H Y , WANG Y J , KANG J Y . A survey of object detection models and its optimization methods [J]. Acta Automatica Sinica , 2021 , 47 ( 6 ): 1232 - 1255 . (in Chinese) . doi: 10.16383/j.aas.c190756 http://dx.doi.org/10.16383/j.aas.c190756
米增 , 连哲 . 面向通用目标检测的YOLO方法研究综述 [J]. 计算机工程与应用 , 2024 , 60 ( 21 ): 38 - 54 .
MI Z , LIAN ZH . Review of YOLO methods for universal object detection [J]. Computer Engineering and Applications , 2024 , 60 ( 21 ): 38 - 54 . (in Chinese)
陈家栋 , 雷斌 . 基于改进SSD轻量化的交通路口目标检测 [J]. 传感器与微系统 , 2022 , 41 ( 10 ): 117 - 121, 126 .
CHEN J D , LEI B . Traffic intersection target detection based on improved SSD lightweight [J]. Transducer and Microsystem Technologies , 2022 , 41 ( 10 ): 117 - 121, 126 . (in Chinese)
王佳琪 , 张淇 , 黄巍 . 不同光照下多模态注意力融合的车辆检测 [J]. 计算机工程与应用 , 2024 , 60 ( 16 ): 116 - 123 .
WANG J Q , ZHANG Q , HUANG W . Vehicle detection of multi-modal attention fusion under different illumination [J]. Computer Engineering and Applications , 2024 , 60 ( 16 ): 116 - 123 . (in Chinese)
LIU L C , SONG X Y , SONG H S , et al . Yolo-3DMM for simultaneous multiple object detection and tracking in traffic scenarios [J]. IEEE Transactions on Intelligent Transportation Systems , 2024 , 25 ( 8 ): 9467 - 9481 . doi: 10.1109/tits.2024.3360875 http://dx.doi.org/10.1109/tits.2024.3360875
胡丹丹 , 张忠婷 . 基于改进YOLOv5s的面向自动驾驶场景的道路目标检测算法 [J]. 智能系统学报 , 2024 , 19 ( 3 ): 653 - 660 .
HU D D , ZHANG ZH T . Road target detection algorithm for autonomous driving scenarios based on improved YOLOv5s [J]. CAAI Transactions on Intelligent Systems , 2024 , 19 ( 3 ): 653 - 660 . (in Chinese)
郭翠霞 , 徐永涛 , 邹章煌 , 等 . 基于可见和红外双模态融合的轻量化行人车辆检测算法 [J]. 光子学报 , 2025 , 54 ( 6 ): 153 - 166 .
GUO C X , XU Y T , ZOU ZH H , et al . Lightweight pedestrian vehicle detection algorithm based on visible and infrared bimodal fusion [J]. Acta Photonica Sinica , 2025 , 54 ( 6 ): 153 - 166 . (in Chinese)
赵海丽 , 许修常 , 潘宇航 . 基于改进YOLOv7-tiny的车辆目标检测算法 [J]. 兵工学报 , 2025 , 46 ( 4 ): 103 - 113 .
ZHAO H L , XU X CH , PAN Y H . Vehicle target detection algorithm based on improved YOLOv7-tiny [J]. Acta Armamentarii , 2025 , 46 ( 4 ): 103 - 113 . (in Chinese)
寇发荣 , 吕庚毅 , 谢伟华 , 等 . 基于改进YOLOv8的煤矿井下无人运输车辆目标检测研究 [J/OL]. 煤炭学报 , 2025 : 1 - 13 . ( 2025-05-20 ). https://link.cnki.net/doi/10.13225/j.cnki.jccs.2025. 0194 https://link.cnki.net/doi/10.13225/j.cnki.jccs.2025.0194 .
KOU F R , LÜ G Y , XIE W H , et al . Research on target detection of unmanned vehicles in coal mines based on improved YOLOv8 [J/OL]. Journal of China Coal Society , 2025 : 1 - 13 . ( 2025-05-20 ). https://link.cnki.net/doi/10.13225/j.cnki.jccs.2025.0194. https://link.cnki.net/doi/10.13225/j.cnki.jccs.2025.0194. (in Chinese)
江旺玉 , 王乐 , 姚叶鹏 , 等 . 多尺度特征聚合扩散和边缘信息增强的小目标检测算法 [J]. 计算机工程与应用 , 2025 , 61 ( 7 ): 105 - 116 .
JIANG W Y , WANG L , YAO Y P , et al . Multi-scale feature aggregation diffusion and edge information enhancement small object detection algorithm [J]. Computer Engineering and Applications , 2025 , 61 ( 7 ): 105 - 116 . (in Chinese)
杜宏 , 顾宸瑜 , 张孝峥 , 等 . 基于YOLOv10-vehicle算法的复杂天气情况下车辆目标检测方法 [J/OL]. 吉林大学学报(工学版) , 2025 : 1 - 10 . ( 2025-01-16 ). https://link.cnki.net/doi/10.13229/j.cnki.jdxbgxb.20250016 https://link.cnki.net/doi/10.13229/j.cnki.jdxbgxb.20250016 .
DU H , GU CH Y , ZHANG X ZH , et al . Vehicle target detection method based on the YOLOv10-vehicle algorithm under complex weather conditions [J/OL]. Journal of Jilin University (Engineering and Technology Edition) , 2025 : 1 - 10 . ( 2025-01-16 ). https://link.cnki.net/doi/10.13229/j.cnki.jdxbgxb.20250016. https://link.cnki.net/doi/10.13229/j.cnki.jdxbgxb.20250016. (in Chinese)
卜祥涛 , 宋亚芳 , 王晓宇 , 等 . 面向多模式图像的改进暗通道先验去雾增强 [J]. 光学 精密工程 , 2025 , 33 ( 13 ): 2124 - 2135 .
BU X T , SONG Y F , WANG X Y , et al . Improved dark channel prior dehazing enhancement for multi-modal images [J]. Optics and Precision Engineering , 2025 , 33 ( 13 ): 2124 - 2135 . (in Chinese)
ZHAO Y A , LV W Y , XU S L , et al . DETRs beat YOLOs on real-time object detection [C]. 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 16-22,2024 , Seattle, WA, USA. IEEE , 2024 : 16965 - 16974 . doi: 10.1109/cvpr52733.2024.01605 http://dx.doi.org/10.1109/cvpr52733.2024.01605
LIN T Y , GOYAL P , GIRSHICK R , et al . Focal loss for dense object detection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2020 , 42 ( 2 ): 318 - 327 . doi: 10.1109/tpami.2018.2858826 http://dx.doi.org/10.1109/tpami.2018.2858826
LIN T Y , DOLLÁR P , GIRSHICK R , et al . Feature pyramid networks for object detection [C]. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 21-26,2017 , Honolulu, HI, USA. IEEE , 2017 : 936 - 944 . doi: 10.1109/cvpr.2017.106 http://dx.doi.org/10.1109/cvpr.2017.106
PIAO Y R , JIANG Y Y , ZHANG M , et al . PANet: patch-aware network for light field salient object detection [J]. IEEE Transactions on Cybernetics , 2023 , 53 ( 1 ): 379 - 391 . doi: 10.1109/tcyb.2021.3095512 http://dx.doi.org/10.1109/tcyb.2021.3095512
TANG F L , XU Z X , HUANG Q M , et al . DuAT : Dual - aggregation Transformer Network for Medical Image Segmentation [M]. Pattern Recognition and Computer Vision . Singapore : Springer Nature Singapore , 2023 : 343 - 356 . doi: 10.1007/978-981-99-8469-5_27 http://dx.doi.org/10.1007/978-981-99-8469-5_27
ZHU X Z , HU H , LIN S , et al . Deformable ConvNets V2: more deformable, better results [C]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 15-20,2019 , Long Beach, CA, USA. IEEE , 2019 : 9300 - 9308 . doi: 10.1109/cvpr.2019.00953 http://dx.doi.org/10.1109/cvpr.2019.00953
0
浏览量
3
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621
