改进YOLOv7+Bytetrack的小目标检测与追踪

doi:10.3778/j.issn.1002-8331.2311-0372

摘要/Abstract

摘要： 近年来，目标检测技术已经相当成熟，但小目标检测一直是目标检测领域的一大挑战。为了解决这一问题，设计一种名为MFF-YOLOv7的小目标检测算法，该算法旨在提高小目标检测的准确率。设计级联双向特征金字塔KBiFPN，以及联合提出的多级感受野特征聚合模块MFA，来聚合浅层特征并增强特征的信息表达能力。为了解决小目标漏检问题，设计了新的解耦头和新的注意力机制。新的解耦头对小目标的检测能力更强，新的注意力机制可以重点关注感兴趣的小目标区域。引入了一种新的损失函数ECIOU，旨在加快模型的收敛速度。为了验证模型的性能，分别在三个小目标数据集上进行了实验。实验结果表明，MFF-YOLOv7算法提高了检测精度。同时，使用多目标追踪Bytetrack算法在MOT17和VisDrone2019-MOT两个多目标追踪数据集上对新模型进行了验证，进一步证明了其有效性。此外，MFF-YOLOv7算法在动态视频追踪中表现出了良好的性能。

关键词: MFF-YOLOv7, 小目标检测, 多级感受野, 多目标追踪, Bytetrack

Abstract: In recent years, although target detection techniques have become quite mature, the detection of small targets has been a major challenge in the field of target detection. In order to solve this problem, a small target detection algorithm called MFF-YOLOv7 is developed. The algorithm aims to improve the accuracy of small target detection. Firstly, the designed cascaded bidirectional feature pyramid KBiFPN and the proposed multi-level sensory field feature aggregation module MFA are used to aggregate the shallow features and improve the information expression of the features to improve the accuracy of small target detection. To solve the problem of small target miss detection, a new decoupling head and a new attention mechanism are designed. This detection head is more capable of detecting small targets and can focus on the small target region of interest through the attention mechanism, thus reducing the leakage detection. Finally, a new loss function, ECIOU, is introduced to speed up the convergence of the model. To validate the performance of the model, experiments are performed on three small target datasets. The experimental results show that the MFF-YOLOv7 algorithm achieves higher detection accuracy. Meanwhile, the effectiveness of the new model is further demonstrated by the multi-target tracking Bytetrack algorithm on two multi-target tracking datasets, MOT17 and VisDrone 2019-MOT. MFF-YOLOv7 is also able to show good performance in dynamic video tracking.

Key words: MFF-YOLOv7, small target detection, multilevel sensory field, multi-target tracking, Bytetrack

聂源, 赖惠成, 高古学. 改进YOLOv7+Bytetrack的小目标检测与追踪[J]. 计算机工程与应用, 2024, 60(12): 189-202.

NIE Yuan, LAI Huicheng, GAO Guxue. Improved Small Target Detection and Tracking with YOLOv7+Bytetrack[J]. Computer Engineering and Applications, 2024, 60(12): 189-202.

参考文献

[1] ZOU Z, CHEN K, SHI Z, et al. Object detection in 20 years: a survey[J]. Proceedings of the IEEE, 2023, 111 (3): 257-276.
[2] VIOLA P, JONES M J. Robust real-time face detection[J]. International Journal of Computer Vision, 2004, 57: 137-154.
[3] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), 2005: 886-893.
[4] FELZENSZWALB P F, GIRSHICK R B, MCALLESTER D. Cascade object detection with deformable part models[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010: 2241-2248.
[5] 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.
[6] 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.
[7] 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.
[8] TAN M, PANG R, LE Q V. Efficientdet: scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10781-10790.
[9] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[J]. arXiv:2004.10934, 2020.
[10] YUN S, HAN D, OH S J, et al. CutMix: regularization strategy to train strong classifiers with localizable features[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 6023-6032.
[11] ZHU X, LYU S, WANG X, et al. TPH-YOLOv5: improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 2778-2788.
[12] WANG J, CHEN Y, DONG Z, et al. Improved YOLOv5 network for real-time multi-scale traffic sign detection[J]. Neural Computing and Applications, 2023, 35(10): 7853-7865.
[13] MUZAMMUL M, LI X. A survey on deep domain adaptation and tiny object detection challenges, techniques and datasets[J]. arXiv:2107.07927, 2021.
[14] LIU S, QI L, QIN H, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.
[15] 谌海云, 黄忠义, 王海川, 等. 基于改进Tracktor的行人多目标跟踪算法[J]. 计算机工程与应用, 2024, 60(8): 242-249.
SHEN H Y, HUANG Z Y, WANG H C, et al. Improved Tracktor-based pedestrian multi-objective tracking algorithm[J]. Computer Engineering and Applications, 2024, 60(8): 242-249.
[16] 赵珊, 郑爱玲, 刘子路, 等. 通道分离双注意力机制的目标检测算法[J]. 计算机科学与探索, 2023, 17(5): 1112-1125.
ZHAO S, ZHENG A L, LIU Z L, et al. Object detection algorithm based on channel separation dual attention mechanism[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(5): 1112-1125.
[17] 贾天豪, 彭力, 戴菲菲. 引入残差学习与多尺度特征增强的目标检测器[J]. 计算机科学与探索, 2023, 17(5): 1102-1111.
JIA T H, PENG L, DAI F F. Object detector with residual learning and multi-scale feature enhancement[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(5): 1102-1111.
[18] ZHANG Q L, YANG Y B. SA-Net: shuffle attention for deep convolutional neural networks[C]//2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021: 2235-2239.
[19] ZHANG X, ZHOU X, LIN M, et al. Shufflenet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6848-6856.
[20] GEVORGYAN Z. SIoU loss: more powerful learning for bounding box regression[J]. arXiv:2205.12740, 2022.
[21] ZHENG Z, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 12993-13000.
[22] 童小钟, 魏俊宇, 苏绍璟, 等. 融合注意力和多尺度特征的典型水面小目标检测[J]. 仪器仪表学报, 2023, 44(1): 212-222.
TONG X Z, WEI J Y, SU S J, et al. Typical small target detection on water surfaces fusing attention and multi-scale features[J]. Chinese Journal of Scientific Instrument, 2023, 44(1): 212-222.
[23] DU D, ZHU P, WEN L, et al. VisDrone-DET2019: the vision meets drone object detection in image challenge results[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, 2019.
[24] 吴明杰, 云利军, 陈载清, 等. 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(2): 191-199.
WU M J, YUN L J, CHEN Z Q. Improved YOLOv5s small object detection algorithm in UAV view[J]. Computer Engineering and Applications, 2024, 60(2): 191-199.
[25] WANG J, LIU W, ZHANG W, et al. LV-YOLOv5: a light-weight object detector of Vit on Drone-captured scenarios[C]//2022 16th IEEE International Conference on Signal Processing (ICSP), 2022: 178-183.
[26] LIU H, DUAN X, LOU H, et al. Improved GBS-YOLOv5 algorithm based on YOLOv5 applied to UAV intelligent traffic[J]. Scientific Reports, 2023, 13(1): 9577.
[27] GE Z, LIU S, WANG F, et al. Yolox: exceeding YOLO series in 2021[J]. arXiv:2107.08430, 2021.
[28] WANG X, HE N, HONG C, et al. YOLO-ERF: lightweight object detector for UAV aerial images[J]. Multimedia Systems, 2023, 29: 3329-3339.
[29] GUO J, LOU H, CHEN H, et al. A new detection algorithm for alien intrusion on highway[J]. Scientific Reports, 2023, 13(1): 10667.
[30] YANG Y, GAO X, WANG Y, et al. VAMYOLOX: an accurate and efficient object detection algorithm based on visual attention mechanism for UAV optical sensors[J]. IEEE Sensors Journal, 2022, 23(11): 11139-11155.
[31] LOU H, DUAN X, GUO J, et al. DC-YOLOv8: small-size object detection algorithm based on camera sensor[J]. Electronics, 2023, 12(10): 2323.
[32] TANG F, YANG F, TIAN X. Long-distance person detection based on YOLOv7[J]. Electronics, 2023, 12(6): 1502.
[33] LUITEN J, OSEP A, DENDORFER P, et al. Hota: a higher order metric for evaluating multi-object tracking[J]. International Journal of Computer Vision, 2021, 129: 548-578.