面向无人机图像场景的小目标检测模型

doi:10.3778/j.issn.1002-8331.2401-0209

摘要/Abstract

摘要： 无人机因其飞行高度和拍摄角度的独特性，采集的遥感图像中存在大量小目标。小目标物体像素小、语义信息少，容易受背景信息干扰和出现聚集遮挡，是当前检测模型性能不佳的主要原因之一。提出一种面向小目标的无人机图像目标检测模型UAIDet（unmanned aerial vehicles images detector），从解决信息冲突和检测框回归难入手，提升模型的检测性能。其一，构建自适应的通道融合模块，在特征融合阶段动态学习通道权重以过滤不同尺度特征之间的信息冲突，抑制特征融合时的尺度不一致性，提高小目标物体的检测能力；其二，设计误差敏感定位损失函数，在小目标物体检测框的收敛阶段提出偏移敏感损失项以解决小目标对几何误差的敏感性，提高定位损失函数的鲁棒性，优化小目标物体的检测精度。在数据集Visdrone2022上对文章方法进行实验，mAP（means average precision）和AP50（average precision at IOU threshold 50%）分别达到了22.0%和37.1%，相较于基准模型分别提高3和4.7个百分点。TinyPerson数据集上的mAP和AP50为9.9%和29.1%，分别提高了4.29和4.2个百分点，证明UAIDet模型的有效性和鲁棒性。

关键词: 目标检测, 无人机图像, 小目标, 特征融合, 损失函数

Abstract: Due to the flying altitude and shooting angle of UAV, many small targets can be seen in the captured images. The small objects have so few pixels and semantic information that can easily be disturbed by complex background information. Meanwhile, aggregation often takes place. Accurate detection of small targets is the key to improve the performance of the detection model for UAV. In this paper, a detection model called UAIDet for small objects is proposed. An adaptive channel fusion module is developed. In the feature fusion stage, the channel weights are dynamically learned to filter out the information conflicts between the different features levels, to improve the detection ability for small targets. In addition, an offset-sensitive loss function is developed for the location. In the convergence phase of the small object bounding box, the offset-sensitive term solves the sensitivity for geometric error by using root function. The model UAIDet is tested in dataset Visdrone2022, the mAP and AP50 reach 22.0% and 37.1% respectively, which is 3 and 4.7 percentage points higher than that of the benchmark model. The experiment in TinyPerson dataset shows 9.9% mAP and 29.1% AP50, improve 4.29 and 4.2 percentage points separately. The results have verified the robustness and effectiveness of UAIDet model.

Key words: object detection, unmanned aerial vehicles images, small object, feature fusion, loss function

朱堃煌, 孙博, 毛国君. 面向无人机图像场景的小目标检测模型[J]. 计算机工程与应用, 2024, 60(15): 243-251.

ZHU Kunhuang, SUN Bo, MAO Guojun. Detection Model for Small Objects in UAV Image Scene[J]. Computer Engineering and Applications, 2024, 60(15): 243-251.

参考文献

[1] 罗旭东, 吴一全, 陈金林. 无人机航拍影像目标检测与语义分割的深度学习方法研究进展[J]. 航空学报, 2024, 45(6): 235-264.
LUO X D, WU Y Q, CHEN J L. Research progress on deep learning methods for object detection and semantic segmentation in UAV aerial images[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(6): 235-264.
[2] 叶回春, 陈森政, 郭安廷, 等. 香蕉枯萎病无人机多光谱影像数据集[J]. 中国科学数据 (中英文网络版), 2024, 9(2): 1-8.
YE H C, CHEN S Z, GUO A T, et al. A dataset of UAV multispectral images for a banana Fusarium wilt survey[J]. China Scientific Data, 2024, 9(2): 1-8.
[3] 杨佳乐. 基于无人机遥感影像的三维道路线形识别[D]. 桂林: 桂林理工大学, 2023.
YANG J L. Three-dimensional road linear identification based on UAV remote sensing image[D]. Guilin: Guilin University of Technology, 2023.
[4] EVERINGHAM M, ESLAMI S A, VAN GOOL L, et al. The pascal visual object classes challenge: a retrospective[J]. International Journal of Computer Vision, 2015, 111: 98-136.
[5] LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft coco: common objects in context[C]//Proceedings of 13th European Conference on Computer Vision, Zurich, Switzerland, September 6-12, 2014. [S.l.]: Springer, 2014: 740-755.
[6] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of 14th European Conference on Computer Vision (ECCV 2016), Amsterdam, The Netherlands, October 11-14, 2016. [S.l.]: Springer, 2016: 21-37.
[7] LIN T Y, DOLLáR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
[8] 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.
[9] 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.
[10] 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.
[11] 李安达, 吴瑞明, 李旭东. 改进YOLOv7的小目标检测算法研究[J]. 计算机工程与应用, 2024, 60(1): 122-134.
LI A D, WU R M, LI X D. Research on improving YOLOv7’s small target detection algorithm[J]. Computer Engineering and Applications, 2024, 60(1): 122-134.
[12] ZHANG G, LI Z, LI J, et al. CFNET: cascade fusion network for dense prediction[J]. arXiv:2302.06052, 2023.
[13] WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 7464-7475.
[14] 张光华, 李聪发, 李钢硬, 等. 基于改进YOLOv7-tiny的无人机航拍图像小目标检测算法[J/OL]. 工程科学与技术: 1-14(2024-04-19). https://doi.org/10.15961/j.jsuese.202300593.
ZHANG G H, LI C F, LI G Y, et al. Small target detection algorithm for UAV aerial images based on improved YOLOv7-tiny[J/OL]. Advanced Engineering Sciences: 1-14(2024-04-19). https://doi.org/10.15961/j.jsuese.202300593.
[15] TROCKMAN A, KOLTER J Z. Patches are all you need?[J]. arXiv:2201.09792, 2022.
[16] 赵鑫, 陈里里 , 杨维川, 等. DY-YOLOv5: 基于多重注意力机制的航拍图像目标检测[J]. 计算机工程与应用, 2024, 60(7): 183-191.
ZHAO X, CHEN L L, YANG W C, et al. DY-YOLOv5: target detection for aerial image based on multiple attention[J]. Computer Engineering and Applications, 2024, 60(7): 183-191.
[17] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 4700-4708.
[18] 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.
[19] 赵耘彻, 张文胜, 刘世伟. 基于改进YOLOv4的无人机航拍目标检测算法[J]. 电子测量技术, 2023, 46(8): 169-175.
ZHAO Y C, ZHANG W S, LIU S W. UAV aerial object detection algorithm based on improved YOLOv4[J]. Electronic Measurement Technology, 2023, 46(8): 169-175.
[20] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. Yolov4: optimal speed and accuracy of object detection[J]. arXiv:2004.10934, 2020.
[21] HOWARD A, SANDLER M, CHU G, et al. Searching for mobilenetv3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324.
[22] ARTHUR D, VASSILVITSKII S. K-means++ the advantages of careful seeding[C]//Proceedings of the Eighteenth Annual ACM-SIAM Symposium on Discrete Algorithms, 2007: 1027-1035.
[23] 薛珊, 卢涛, 吕琼莹, 等. 基于多尺度融合和轻量化网络的无人机目标检测算法[J]. 湖南大学学报 (自然科学版), 2023, 50(8): 82-93.
XUE S, LU T, LYU Q Y, et al. Drone target detection algorithm based on multi-scale fusion and lightweight network[J]. Journal of Hunan University (Natural Sciences), 2023, 50(8): 82-93.
[24] 白宇, 周艳媛, 安胜彪. 改进YOLOv5的无人机小目标检测方法研究[J]. 计算机工程与应用, 2024, 60(10): 276-284.
BAI Y, ZHOU Y Y, AN S B. Research on UAV small object detection method improved by YOLOv5[J]. Computer Engineering and Applications, 2024, 60(10): 276-284.
[25] LIU S, HUANG D, WANG Y. Learning spatial fusion for single-shot object detection[J]. arXiv:1911.09516, 2019.
[26] WOO S, PARK J, LEE J Y, et al. Cbam: convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018: 3-19.
[27] 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.
[28] CAO Y, HE Z, WANG L, et al. VisDrone-DET2021: the vision meets drone object detection challenge results[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 2847-2854.
[29] YU X, GONG Y, JIANG N, et al. Scale match for tiny person detection[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2020: 1257-1265.
[30] 刘展威, 陈慈发, 董方敏. 基于YOLOv5s的航拍小目标检测改进算法研究[J]. 无线电工程, 2023, 53(10): 2286-2294.
LIU Z W, CHEN C F, DONG F M. Improved aerial small object detection algorithm based on YOLOv5s[J]. Radio Engineering, 2023, 53(10): 2286-2294.
[31] LI Z, PENG C, YU G, et al. Light-head R-CNN: in defense of two-stage object detector[J]. arXiv:1711.07264, 2017.
[32] LAW H, DENG J. Cornernet: detecting objects as paired keypoints[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018: 734-750.
[33] 冒国韬, 邓天民, 于楠晶. 基于多尺度分割注意力的无人机航拍图像目标检测算法[J]. 航空学报, 2023, 44(5): 273-283.
MAO G T, DENG T M, YU N J. Object detection in UAV images based on multi-scale split attention[J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(5): 273-283.
[34] JIANG B, QU R, LI Y, et al. VC-YOLO: towards real-time object detection in aerial images[J]. Journal of Circuits, Systems and Computers, 2022, 31(8): 2250147.
[35] 陈卫彪, 贾小军, 朱响斌, 等. 基于DSM-YOLO v5的无人机航拍图像目标检测[J]. 计算机工程与应用, 2023, 59(18): 226-233.
CHEN W B, JIA X J, ZHU X B, et al. Target detection for UAV image based on DSM-YOLO v5[J]. Computer Engineering and Applications, 2023, 59(18): 226-233.
[36] GE Z, LIU S, WANG F, et al. YOLOX: exceeding YOLO series in 2021[J]. arXiv:2107.08430, 2021.
[37] 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.
[38] SUNKARA R, LUO T. No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects[M/OL]//AMINI M R, CANU S, FISCHER A, et al. Machine learning and knowledge discovery in databases. Cham: Springer Nature Switzerland, 2023: 443-459.
[39] CHEN H, LIU H, SUN T, et al. MC-YOLOv5: a multi-class small object detection algorithm[J]. Biomimetics, 2023, 8(4): 342.
[40] 李钟华, 林初俊, 朱恒亮, 等. 基于结构感知和全局上下文信息的小目标检测[J]. 计算机工程与应用, 2024, 60(9): 292-298.
LI Z H, LIN C J, ZHU H L, et al. Small object detection based on structure perception and global context information[J]. Computer Engineering and Applications, 2024, 60(9): 292-298.