基于功能保持的特征金字塔目标检测网络

doi:10.16451/j.cnki.issn1003-6059.202006004

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2523 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对特征金字塔网络在多尺度与小目标检测上的问题,提出基于功能保持的特征金字塔目标检测网络.首先在主网络中选择特征图构建特征金字塔,针对不同尺度的特征图,通过功能保持融合模块自上而下地进行低损失的特征融合.功能保持融合模块有效保留高层的强语义信息,增强底层特征图对小目标的表示能力.再利用网络两个阶段的特征描述目标,提升检测精度.最后,充分利用上下文信息进一步增强对多尺度目标的判别能力.在PASCAL VOC公共数据集上的实验表明,文中网络检测效果较优.同时,通过检测效果图可看出,文中网络在目标遮挡、模糊等情况下的检测效果也较优.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	徐成琪
	洪学海

关键词 ：功能保持, 特征融合, 二阶段检测, 特征金字塔网络

Abstract：To solve the problem of feature pyramid network in multi-scale and small object detection, a feature pyramid object detection network based on function maintenance is proposed. Firstly, feature maps are selected in the backbone convolutional architecture to build feature pyramid. For these feature maps of different scales, feature fusion with low loss is carried out from top to bottom using function maintenance fusion module. The strong high-level semantic information is maintained more effectively, and the representation ability for small object of low-level feature maps is greatly enhanced. The detection precision is improved by two-stage features of the proposed network to describe the objects. Finally, context information is fully utilized to further enhance the ability to distinguish multi-scale object. Experiments on PASCAL VOC public dataset show that the detection result of the proposed network is satisfactory. Moreover, the proposed network achieves better results in the case of object occlusion and blur as well.

Key words： Function Maintenance Feature Fusion Two-Stage Detection Feature Pyramid Network

收稿日期: 2020-01-08

ZTFLH:

TP 391.41

基金资助:国家重点研发计划项目(No.2016YFC1401706)、江西省重点研发计划项目(No.S2019ZPYFB0553)资助

通讯作者: 洪学海,博士,研究员.主要研究方向为高性能计算、大数据与云计算、人工智能.E-mail:hxh@ict.ac.cn.

作者简介: 徐成琪,硕士研究生,主要研究方向为机器学习、计算机视觉.E-mail:1462004508@qq.com.

引用本文:

徐成琪, 洪学海. 基于功能保持的特征金字塔目标检测网络[J]. 模式识别与人工智能, 2020, 33(6): 507-517. XU Chengqi, HONG Xuehai. Feature Pyramid Object Detection Network Based on Function Maintenance. , 2020, 33(6): 507-517.

链接本文:

http://manu46.magtech.com.cn/Jweb_prai/CN/10.16451/j.cnki.issn1003-6059.202006004 或 http://manu46.magtech.com.cn/Jweb_prai/CN/Y2020/V33/I6/507

[1] REDMON J, DIVVALA S, GIRSHICK R, et al. You Only Look Once: Unified, Real-Time Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 779-788.
[2] REDMON J, FARHADI A. YOLO9000: Better, Faster, Stronger // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 6517-6525.
[3] REDMON J, FARHADI A. Yolov3: An Incremental Improvement[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1804.02767.pdf.
[4] LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot Multibox Detector // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 21-37.
[5] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal Loss for Dense Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 318-327.
[6] ZHANG S F, WEN L Y, BIAN X, et al. Single-Shot Refinement Neural Network for Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 4203-4212.
[7] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2014: 580-587.
[8] HE K M, ZHANG X Y, REN S Q, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2014: 346-361.
[9] GIRSHICK R. Fast R-CNN // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 1440-1448.
[10] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1506.01497.pdf.
[11] DAI J F, LI Y, HE K M, et al. R-FCN: Object Detection via Region-Based Fully Convolutional Networks[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1605.06409.pdf.
[12] LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature Pyramid Networks for Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 936-944.
[13] TIAN Z, SHEN C H, CHEN H, et al. FCOS: Fully Convolutional One-Stage Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1904.01355.pdf.
[14] ZHU C C, HE Y H, SAVVIDES M. Feature Selective Anchor-Free Module for Single-Shot Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1903.00621.pdf.
[15] CAI Z W, VASCONCELOS N. Cascade R-CNN: Delving into High Quality Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 6154-6162.
[16] GHIASI G, LIN T Y, LE Q V. NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 7029-7038.
[17] CHEN Y T, HAN C X, WANG N Y, et al. Revisiting Feature Alignment for One-Stage Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1908.01570.pdf.
[18] PENG J R, SUN M, ZHANG Z X, et al. Efficient Neural Architecture Transformation Search in Channel-Level for Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1909.02293.pdf.
[19] LI Y H, CHEN Y T, WANG N Y, et al. Scale-Aware Trident Networks for Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1901.01892.pdf.
[20] LIU S, QI L, QIN H F, et al. Path Aggregation Network for Instance Segmentation // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 8759-8768.
[21] MA W C, WU Y W, WANG Z B, et al. MDCN: Multi-scale, Deep Inception Convolutional Neural Networks for Efficient Object Detection // Proc of the 24th International Conference on Pattern Recognition. Washington, USA: IEEE, 2018: 2510-2515.
[22] SINGH B, DAVIS L S. An Analysis of Scale Invariance in Object Detection-SNIP // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 3578-3587.
[23] SINGH B, NAJIBI M, DAVIS L S. SNIPER: Efficient Multi-scale Training[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1805.09300.pdf.
[24] KONG T, YAO A B, CHEN Y R, et al. Hypernet: Towards Accurate Region Proposal Generation and Joint Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 845-853.
[25] BELL S, ZITNICK C L, BALA K, et al. Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 2874-2883.
[26] CAI Z W, FAN Q F, FERIS R S, et al. A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 354-370.
[27] HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 770-778.
[28] DENG J, DONG W, SOCHER R, et al. ImageNet: A Large-Scale Hierarchical Image Database // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2009: 248-255.
[29] SHRIVASTAVA A, GUPTA A, GIRSHICK R. Training Region-Based Object Detectors with Online Hard Example Mining // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 761-769.
[30] KONG T, SUN F C, YAO A B, et al. RON: Reverse Connection with Objectness Prior Networks for Object Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 5936-5944.
[31] FU C Y, LIU W, RANGA A, et al. DSSD: Deconvolutional Single Shot Detector[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1701.06659.pdf.
[32] SHEN Z Q, LIU Z, LI J G, et al. DSOD: Learning Deeply Supervised Object Detectors from Scratch // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2017: 1937-1945.
[33] YOO J, SEO G, CHUNG I, et al. Mixture-Model-Based Bounding Box Density Estimation for Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1911.12721.pdf.
[34] LIM J S, ASTRID M, YOON H J, et al. Small Object Detection Using Context and Attention[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1912.06319.pdf.
[35] MA W C, WU Y W, CEN F, et al. MDFN: Multi-scale Deep Feature Learning Network for Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1912.04514.pdf.
[36] GIDARIS S, KOMODAKIS N. Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 1134-1142.
[37] ZHOU W, LI Y Y. Feature Fusion Detector for Semantic Cognition of Remote Sensing[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1909.13047.pdf.
[38] YOO J H, KUM D, CHOI J W. ScarfNet: Multi-scale Features with Deeply Fused and Redistributed Semantics for Enhanced Object Detection[C/OL]. [2020-01-03]. https://arxiv.org/pdf/1908.00328.pdf.