改进YOLOv5s的无人机视角下小目标检测算法

doi:10.3778/j.issn.1002-8331.2304-0150

摘要/Abstract

摘要： 无人机视角的小目标图像具有目标分布密集、类别不均衡以及特征不明显的特点，导致目标检测任务中出现漏检、误检的问题。针对这些问题，提出一种改进YOLOv5s小目标检测方法，以达到提高目标检测准确率与精确度的目的。重新聚类锚框，更精确地锁定检测区域。更改骨干网络结构，在空间金字塔池化层增加卷积，保证充分获取检测目标特征。同时，将网络结构中的C3模块替换成融合通道注意力机制的轻量级SEC2f模块，以提升网络对于小目标检测的局部特征捕获能力。融合解耦检测头，结合自适应锚框计算，有效提取目标区域的特征。在相同参数、相同环境条件下，在DOTA数据集上和VisDrone数据集上检测精度分别提升6.1%、5.2%，表明改进方法在小目标检测任务上的有效性；在公开数据集voc2007+2012上做通用性对比实验，结果表明改进算法具有通用性。

关键词: YOLOv5s, 聚类算法, SEC2f模块, 空间金字塔池化, 解耦检测头

Abstract: The small target image from the UAV perspective has the characteristics of dense target distribution, unbalanced category and inconspicuous features, which leads to the problem of missed detection and false detection in the target detection task. To solve these problems, an improved YOLOv5s small target detection method is proposed to improve the accuracy and accuracy of target detection. First, it reclusters the anchor box to lock the detection area more accurately. Secondly, the backbone network structure is changed and convolution is added to the spatial pyramid pool layer to ensure that the detection target features are fully obtained. At the same time, the C3 module in the network structure is replaced with a lightweight SEC2f module that fuses the channel attention mechanism to improve the local feature acquisition ability of the network for small target detection. Finally, the features of the target area are extracted effectively by combining the decoupled detection head with the adaptive anchor frame calculation. Under the same parameters and environmental conditions, the detection accuracy on DOTA data set and VisDrone data set is improved by 6.1% and 5.2%, respectively, indicating the effectiveness of the improved method on small target detection tasks. The comparison experiment on voc2007+2012 public data set shows the universality of the improved algorithm.

Key words: YOLOv5s, clustering algorithm, SEC2f module, spatial pyramid pool, decoupling detection head

刘涛, 高一萌, 柴蕊, 李政通. 改进YOLOv5s的无人机视角下小目标检测算法[J]. 计算机工程与应用, 2024, 60(1): 110-121.

LIU Tao, GAO Yimeng, CHAI Rui, LI Zhengtong. Improved YOLOv5s UAV View Small Target Detection Algorithm[J]. Computer Engineering and Applications, 2024, 60(1): 110-121.

参考文献

[1] 曹亚明, 肖奇, 杨震. 仿真图像作为模板的遥感影像小目标检测方法[J]. 计算机工程与应用, 2022, 58(17): 111-119.
CAO Y M, XIAO Q, YANG Z. Remote sensing image small target detection method using simulation image as template[J]. Computer Engineering and Applications, 2022, 58(17): 111-119.
[2] 王斌, 李靖, 赵康, 等. 面向火焰快速检测的轻量化深度网络研究[J]. 计算机工程与应用, 2022, 58(17): 256-262.
WANG B, LI J, ZHAO K, et al. Research on lightweight depth network for rapid flame detection[J]. Computer Engineering and Applications, 2022, 58(17): 256-262.
[3] 何丽, 张红艳, 房婉琳. 融合多尺度边界特征的显著实例分割[J]. 计算机科学与探索, 2022, 16(8): 1865-1876.
HE L, ZHANG H Y, FANG W L. Salient instance segmentation via multiscale boundary characteristic network[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(8): 1865-1876.
[4] 任宁, 付岩, 吴艳霞, 等. 深度学习应用于目标检测中失衡问题研究综述[J]. 计算机科学与探索, 2022, 16(9): 1933-1953.
REN N, FU Y, WU Y X, et al. Review of research on imbalance problem in deep learning applied to object detection[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(9): 1933-1953.
[5] 余颖舜, 李成瑶, 潘长开. 行车障碍识别算法的精度提升方法[J]. 汽车与新动力, 2022, 5(6): 48-50.
YU Y S, LI C Y, PAN C K. The accuracy improvement method of driving obstacle recognition algorithm[J]. Automobile and New Powertrain, 2022, 5(6): 48-50.
[6] 王波, 冯旭鹏, 刘利军, 等. 基于改进YOLO算法的肺部CT图像中结节检测研究[J]. 北京生物医学工程, 2020, 39(6): 615-621.
WANG B, FENG X P, LIU L J, et al. Study on nodule detection in lung CT images based on improved YOLO algorithm[J]. Beijing Biomedical Engineering, 2020, 39(6): 615-621.
[7] 宋晓茹, 刘康, 高嵩, 等. 复杂战场环境下改进YOLOv5军事目标识别算法研究[J/OL]. 兵工学报: 1-15[2023-03-23]. http://kns.cnki.net/kcms/detail/11.2176.TJ.20221226.1804.
003.html.
SONG X R, LIU K, GAO S, et al. Research on improved YOLOv5 military target recognition algorithm in complex battlefield environment[J/OL]. Acta Armamentarii: 1-15[2023-03?23].http://kns.cnki.net/kcms/detail/11.2176.TJ.20221226.
1804.003.html.
[8] 茅智慧, 朱佳利, 吴鑫, 等. 基于 YOLO 的自动驾驶目标检测研究综述[J]. 计算机工程与应用, 2022, 58(15): 68-77.
MAO Z H, ZHU J L, WU X, et al. Review of YOLO based target detection for autonomous driving[J]. Computer Engineering and Applications, 2022, 58(15): 68-77.
[9] 董刚, 谢维成, 黄小龙, 等. 深度学习小目标检测算法综述[J]. 计算机工程与应用, 2023, 59(11): 16-27. .
DONG G, XIE W C, HUANG X L, et al. Review of small object detection algorithms based on deep learning[J]. Computer Engineering and Applications, 2023, 59(11): 16-27.
[10] 刘谱, 张兴会, 张志利, 等. 从RCNN到YOLO的目标检测综述[C]//第十六届全国信号和智能信息处理与应用学术会议论文集, 2022: 16-23.
LIU P, ZHANG X H, ZHANG Z L, et al. Review of target detection ranging from RCNN to YOLO[C]//Proceedings of the 16th National Conference on Signal and Intelligent Information Processing and Application, 2022: 16-23.
[11] 张艳, 张明路, 吕晓玲, 等. 深度学习小目标检测算法研究综述[J]. 计算机工程与应用, 2022, 58(15): 1-17.
ZHANG Y, ZHANG M L, LYU X L, et al. Review of research on small target detection based on deep learning[J]. Computer Engineering and Applications, 2022, 58(15): 1-17.
[12] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
[13] GIRSHICK R. Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440-1448.
[14] REN S, HE K, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems, 2015.
[15] LYU Z, JIN H, ZHEN T, et al. Small object recognition algorithm of grain pests based on SSD feature fusion[J]. IEEE Access, 2021, 9: 43202-43213.
[16] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision, 2017: 2980-2988.
[17] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real?time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779-788.
[18] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7263-7271.
[19] REDMON J, FARHADI A. Yolov3: an incremental improvement[J]. arXiv:1804.02767, 2018.
[20] BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4: optimal speed and accuracy of object detection[J]. arXiv:2004. 10934, 2020.
[21] ZHAN W, SUN C, WANG M, et al. An improved Yolov5 real-time detection method for small objects captured by UAV[J]. Soft Computing, 2022, 26(1): 361-373.
[22] SONG Z, ZHANG Y, LIU Y, et al. MSFYOLO: feature fusion based detection for small objects[J]. IEEE Latin America Transactions, 2022, 20(5): 823-830.
[23] CHEN C, ZHANG Y, LV Q, et al. RRnet: a hybrid detector for object detection in drone-captured images[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2019: 100-108.
[24] ZHU P, WEN L, DU D, et al. Detection and tracking meet drones challenge[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 44(11): 7380-7399.
[25] 周华平, 郭伟. 改进YOLOv5网络在遥感图像目标检测中的应用[J]. 遥感信息, 2022, 37(5): 23-30.
ZHOU H P, GUO W. Improved YOLOv5 network in application of remote sensing image object detection[J]. Remote Sensing Information, 2022, 37(5): 23-30.
[26] 李惠惠, 范军芳, 陈启丽. 改进YOLOv5的遥感图像目标检测[J]. 弹箭与制导学报, 2022, 42(4): 17-23.
LI H H, FAN J F, CHEN Q L. Improved YOLOv5 remote sensing image target detection[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2022, 42(4): 17-23.
[27] XI D, QIN Y, WANG S. YDRSNet: an integrated Yolov5-Deeplabv3+real-time segmentation network for gear pitting measurement[J]. Journal of Intelligent Manufacturing, 2021, 34(4): 1585-1599.
[28] GONG H, MU T, LI Q, et al. Swin-transformer-enabled yolov5 with attention mechanism for small object detection on satellite images[J]. Remote Sensing, 2022, 14: 2861.
[29] LIU Y, SUN P, WERGELES N M, et al. A survey and performance evaluation of deep learning methods for small object detection[J]. Expert Syst Appl, 2021, 172: 114602.
[30] CHEN G, WANG H, CHEN K, et al. A survey of the four pillars for small object detection: multiscale representation, contextual information, super-resolution, and region proposal[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2022, 52: 936-953.
[31] 张徐, 朱正为, 郭玉英, 等. 基于cosSTR-YOLOv7的多尺度遥感小目标检测[J/OL]. 电光与控制: 1-9[2023-07-03]. http://kns.cnki.net/kcms/detail/41.1227.tn.20230615.1017.
002.html.
ZHANG X, ZHU Z W, GUO Y Y, et al. Multi-scale remote sensing small target detection based on cosSTR-YOLOv7[J/OL]. Electronics Optics & Control: 1-9 [2023-07-03]. http://kns.cnki.net/kcms/detail/41.1227.tn.20230615.1017.
002.html.
[32] GE Z, LIU S, WANG F, et al. YOLOX: exceeding YOLO series in 2021[J]. arXiv:2107.08430, 2021.
[33] HE K, ZHANG X, REN S, 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.
[34] WANG C Y, BOCHKOVSKIY A, LIAO H Y. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
[35] XIA G S, BAI X, DING J, et al. DOTA: a large-scale dataset for object detection in aerial images[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
[36] CHANDIO A, GUI G, KUMAR T, et al. Precise single-stage detector[J]. arXiv:2210.04252, 2022.