改进SSD的机场场面多尺度目标检测算法

doi:10.3778/j.issn.1002-8331.2010-0110

计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (5): 264-270.DOI: 10.3778/j.issn.1002-8331.2010-0110

改进SSD的机场场面多尺度目标检测算法

黄国新，李炜，张比浩，梁斌斌，韩笑冬，宫江雷，武长青

1.四川大学视觉合成图形图像技术国防重点学科实验室，成都 610065
2.四川大学空天科学与工程学院，成都 610065
3.四川川大智胜软件股份有限公司，成都 610045
4.中国空间技术研究院通信与导航卫星总体部，北京 100094

出版日期:2022-03-01 发布日期:2022-03-01

Improved SSD-Based Multi-scale Object Detection Algorithm in Airport Surface

HUANG Guoxin, LI Wei, ZHANG Bihao, LIANG Binbin, HAN Xiaodong, GONG Jianglei, WU Changqing

1.National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, Chengdu 610065, China
2.School of Aeronautics and Astronautics, Sichuan University, Chengdu 610065, China
3.Wisesoft Co. , Ltd. , Chengdu 610045, China
4.General Department of Communications and Navigation Satellites, China Academy of Space Technology, Beijing 100094, China

Online:2022-03-01 Published:2022-03-01

摘要/Abstract

摘要： 针对现有基于深度学习的通用目标检测方法对机场场面环境目标尺度差别大，特别是小目标难以检测到的问题，提出了一个基于SSD算法并结合特征金字塔融合网络的多尺度目标检测算法。该算法采用了更深的ResNet-50作为骨干网络，并单独设计了六层额外特征层。使用特征金字塔网络进行特征融合，以获得更鲁棒的语义信息。使用Soft-NMS以解决存在的漏检情况，调整先验框的尺度比以更好地检测小目标。通过在机场场面数据集实验表明，该改进算法能够在推断速度为32?frame/s的情况下，取得86.31%的mAP，对比其他先进的检测器，达到领先水平。

关键词: 特征金字塔, 目标检测, 机场场面监视, Soft-NMS

Abstract: Aiming at the problem that the existing general object detection method based on deep learning is difficult to detect the airport scene environment object with large scale difference, especially small targets are difficult to detect. This paper proposes a multi-scale object detection algorithm based on SSD algorithm combined with feature pyramid fusion network. The algorithm first uses ResNet-50 as the backbone network, and separately designs six additional feature layers. Secondly, the feature pyramid network is used for feature fusion to obtain more robust semantic information. Finally, Soft-NMS is used to solve the missing detection situation, and the scale ratio of the prior frame is adjusted to better detect small objects. Experiments on the airport surface dataset show that the improved algorithm can achieve 86.31% mAP when the inferred speed is 32?frame/s, which is at the leading level compared with other advanced detectors.

Key words: feature pyramid, object detection, airport surface surveillance, Soft-NMS

黄国新, 李炜, 张比浩, 梁斌斌, 韩笑冬, 宫江雷, 武长青. 改进SSD的机场场面多尺度目标检测算法[J]. 计算机工程与应用, 2022, 58(5): 264-270.

HUANG Guoxin, LI Wei, ZHANG Bihao, LIANG Binbin, HAN Xiaodong, GONG Jianglei, WU Changqing. Improved SSD-Based Multi-scale Object Detection Algorithm in Airport Surface[J]. Computer Engineering and Applications, 2022, 58(5): 264-270.

参考文献

[1] BESADA J A，GARCIA J，PORTILLO J，et al.Airport surface surveillance based on video images[J].IEEE Transactions on Aerospace and Electronic Systems，2005，41（3）：1075-1082.
[2] QIN Z，WANG J，HUANG C.The monocular visual imaging technology model applied in the airport surface surveillance[C]//International Symposium on Photoelectronic Detection and Imaging 2013：Imaging Sensors and Applications，International Society for Optics and Photonics，2013.
[3] VADDI S，LU H，HAYASHI M.Computer vision based surveillance concept for airport ramp operations[C]//Digital Avionics Systems Conference，2013.
[4] SCHMIDT M，RUDOLPH M，PAPENFUSS A，et al.Remote airport traffic control center with augmented vision video panorama[C]//2009 IEEE/AIAA 28th Digital Avionics Systems Conference，2009.
[5] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[J].arXiv：1512.02325，2015.
[6] SIMONYAN K，ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[J].arXiv：1409.
1556，2014.
[7] EVERINGHAM M，VAN GOOL L，WILLIAMS C K，et al.The pascal visual object classes（voc） challenge[J].International Journal of Computer Vision，2010，88（2）：303-338.
[8] BODLA N，SINGH B，CHELLAPPA R，et al.Soft-NMS—improving object detection with one line of code[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：5561-5569.
[9] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
[10] LIN T，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.
[11] HE K，SUN J.Convolutional neural networks at constrained time cost[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2015：5353-5360.
[12] CHOLLET F.Xception：deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：1251-1258.
[13] ZHOU P，NI B，GENG C，et al.Scale-transferrable object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：528-537.
[14] WOO S，HWANG S，KWEON I S.StairNet：top-down semantic aggregation for accurate one shot detection[C]//2018 IEEE Winter Conference on Applications of Computer Vision（WACV），2018：1093-1102.
[15] KONG T，SUN F，YAO A，et al.Ron：reverse connection with objectness prior networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：5936-5944.
[16] NEUBECK A，VAN GOOL L.Efficient non-maximum suppression[C]//18th International Conference on Pattern Recognition（ICPR’06），2006：850-855.
[17] 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（1）：98-136.
[18] LIN T，MAIRE M，BELONGIE S，et al.Microsoft coco：common objects in context[C]//European Conference on Computer Vision，2014：740-755.
[19] PASZKE A，GROSS S，MASSA F，et al.Pytorch：an imperative style，high-performance deep learning library[C]//Advances in Neural Information Processing Systems，2019：8026-8037.
[20] REDMON J，FARHADI A.YOLOv3：an incremental improvement[J].arXiv：1804.02767，2018.
[21] CHEN K，WANG J，PANG J，et al.Mmdetection：open mmlab detection toolbox and benchmark[J].arXiv：1906. 07155，2019.

改进SSD的机场场面多尺度目标检测算法

Improved SSD-Based Multi-scale Object Detection Algorithm in Airport Surface

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王鑫鹏, 王晓强, 林浩, 李雷孝, 杨艳艳, 孟闯, 高静. 深度学习典型目标检测算法的改进综述[J]. 计算机工程与应用, 2022, 58(6): 42-57.
[2]	李彦辰, 张小俊, 张明路, 沈亮屹. 基于改进Efficientdet的自动驾驶场景目标检测[J]. 计算机工程与应用, 2022, 58(6): 183-191.
[3]	郭宇阳, 胡伟超, 戴帅, 陈艳艳. 面向路侧交通监控场景的轻量车辆检测模型[J]. 计算机工程与应用, 2022, 58(6): 192-199.
[4]	张振伟, 郝建国, 黄健, 潘崇煜. 小样本图像目标检测研究综述[J]. 计算机工程与应用, 2022, 58(5): 1-11.
[5]	贺宇哲, 何宁, 张人, 梁煜博, 刘晓晓. 面向深度学习目标检测模型训练不平衡研究[J]. 计算机工程与应用, 2022, 58(5): 172-178.
[6]	刘佳, 卞方舟, 陈大鹏, 李为斌. 基于UGF-Net的指尖检测模型[J]. 计算机工程与应用, 2022, 58(5): 225-231.
[7]	张馨月, 降爱莲. 融合特征增强和自注意力的SSD小目标检测算法[J]. 计算机工程与应用, 2022, 58(5): 247-255.
[8]	李文婧, 徐国伟, 孔维刚, 郭风祥, 宋庆增. 基于改进YOLOv4的植物叶茎交点目标检测研究[J]. 计算机工程与应用, 2022, 58(4): 221-228.
[9]	刘艳萍, 刘甜. 改进的Cascade RCNN行人检测算法研究[J]. 计算机工程与应用, 2022, 58(4): 229-236.
[10]	杨其晟, 李文宽, 杨晓峰, 岳琳茜, 李海芳. 改进YOLOv5的苹果花生长状态检测方法[J]. 计算机工程与应用, 2022, 58(4): 237-246.
[11]	关立文, 孙鑫磊, 杨佩. 基于关键点估计的抓取检测算法[J]. 计算机工程与应用, 2022, 58(4): 267-274.
[12]	胡春生, 闫小鹏, 魏红星, 李国利. 基于立体视觉的目标检测与轨迹预测研究综述[J]. 计算机工程与应用, 2022, 58(3): 50-65.
[13]	王胜春, 陈阳. 改进损失函数的地基云状目标检测算法[J]. 计算机工程与应用, 2022, 58(2): 169-175.
[14]	邢永鑫, 孙游东, 王天一. 基于改进SSD算法对奶牛的个体识别[J]. 计算机工程与应用, 2022, 58(2): 208-214.
[15]	袁国文, 张彩霞, 杨阳, 张文生, 白江波. 复杂场景下深度表示的SAR船舶目标检测算法[J]. 计算机工程与应用, 2022, 58(2): 289-294.