计算机科学 ›› 2021, Vol. 48 ›› Issue (6A): 118-121.doi: 10.11896/jsjkx.200700122

• 图像处理&多媒体技术 • 上一篇    下一篇

基于区域激活策略的Tiny YOLOv3目标检测算法

余晗青, 杨贞, 殷志坚   

  1. 江西科技师范大学通信与电子学院 南昌330000
  • 出版日期:2021-06-10 发布日期:2021-06-17
  • 通讯作者: 杨贞(yangzhen@jxstnu.edu.cn)
  • 作者简介:yuhanqing_1996@163.com
  • 基金资助:
    国家自然科学基金(61866016);江西科技师范大学青年拔尖项目(2018QNBJRC002);江西省教育厅一般项目(GJJ190587);江西省自然科学基金面上项目(20202BABL202014);甲骨文信息处理教育部重点实验室开放课题(OIP2019E008)

Tiny YOLOv3 Target Detection Algorithm Based on Region Activation Strategy

YU Han-qing, YANG Zhen, YIN Zhi-jian   

  1. School of Communication and Electronics,Jiangxi Science and Technology Normal University,Nanchang 330000,China
  • Online:2021-06-10 Published:2021-06-17
  • About author:YU Han-qing,born in 1996,postgra-duate.Her main research interests include object detection and so on.
    YANG Zhen,born in 1985,Ph.D.His main research interests include pattern recognition and intelligent systems,machine learning and image processing.
  • Supported by:
    National Natural Science Foundation of China(61866016),Youth Top-notch Project of Jiangxi Science and Technology Normal University(2018QNBJRC002),General Project of Jiangxi Provincial Department of Education (GJJ190587),General Project of Natural Science Foundation of Jiangxi Province(20202BABL202014) and Oracle Information Processing Ministry of Education Key Laboratory Open Project Funding Project(OIP2019E008).

摘要: 针对Tiny YOLOv3模型检测精度低的问题,提出一种将分割信息引入深度卷积神经网络结构中的方法。模型训练期间,将目标真实的位置信息加入网络层中,并手动激活这些目标区域,激励的大小随着训练的进行逐渐减小直至降为零。测试结果表明,在VOC2007数据集上,改进后的Tiny YOLOv3模型的平均准确率提升至58.9%,并且在检测速度上与原模型保持一致,满足实时检测的需要。

关键词: Tiny YOLOv3, 分割信息, 深度卷积神经网络, 位置信息

Abstract: Aiming at the problem of low detection accuracy of Tiny YOLOv3 model,a method to introduce segmentation information into deep convolutional neural network structure is proposed.During the model training,the real position information of the target is added to the network layer,and these target areas are manually activated.The size of the excitation gradually decreases as the training proceeds until it drops to zero.The test results show that on the VOC2007 data set,the average accuracy of the improved Tiny YOLOv3 model is increased to 58.9%,and the detection speed is consistent with the original model to meet the needs of real-time detection.

Key words: Deep convolutional neural network, Location information, Segmentation information, Tiny YOLOv3

中图分类号: 

  • TP391
[1] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis & Machince Intelligence,2017,39(6):1137-1149.
[2] CHEN C,QI F.The development of convolutional neuralnet-work and its application in the field of computer vision[J].Computer Science,2019,46(3):69-79.
[3] 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.
[4] REDMON J,FARHADI A.YOLO9000:Better,Faster,Stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017.
[5] REDMON J,FARHADI A.YOLOv3:An Incremental Improvement[J].arXiv:1804.02767,2018.
[6] BOCHKOVSKIY A,WANG C Y,LIAO H Y M.YOLOv4:Op-timal Speed and Accuracy of Object Detection[J].arXiv:2004.10934,2020.
[7] LIU Z H.Structural model learning based on local feat-ures and its application in target detection and location[D].Shanghai:Shanghai Jiaotong University,2012.
[8] DAI J F,HE K M,SUN J.Instance-aware semantic segmentation via multi-task network cascades[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:3150-3158.
[9] HARIHARAN B,ARBELAEZ P,GIRSHICK R,et al.Simultaneous detection and segmentation[C]//European Conferenc on Computer Vision.Springer,2014:297-312.
[10] GIDARIS S,KOMODAKIS N.Object detection via a multi-region and semantic segmentation-aware cnn model[C]//Procee-dings of the IEEE International Conference on Computer Vision.2015:1134-1142.
[11] ZHANG Z S,QIAO S Y,XIE C H,et al.Single-shot object detection with enriched semantics[R].Technical report,Center for Brains,Minds and Machines (CBMM),2018.
[12] HE P,HUANG W L,HE T,et al.Single Shot Text Detector with Regional Attention[C]//2017 IEEE International Conference on Computer Vision(ICCV).IEEE,2017.
[13] DERAKHSHANI M M,MASOUDNIA S,SHAKER A H,et al.Assisted Excitation of Activations:A Learning Technique to Improve Object Detectors[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2020.
[14] HE T,ZHANG Z,ZHANG H,et al.Bag of Tricks for ImageClassification with Convolutional Neural Networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2019.
[15] 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.
[16] BENGIO Y S,LOURADOUR J,COLLOBERT R,et al.Curriculum learning[C]//Proceedings of the 26th Annual International Conference on Machine Learning.ACM,2009:41-48.
[1] 周芳泉, 成卫青.
基于全局增强图神经网络的序列推荐
Sequence Recommendation Based on Global Enhanced Graph Neural Network
计算机科学, 2022, 49(9): 55-63. https://doi.org/10.11896/jsjkx.210700085
[2] 孙洁琪, 李亚峰, 张文博, 刘鹏辉.
基于离散小波变换的双域特征融合深度卷积神经网络
Dual-field Feature Fusion Deep Convolutional Neural Network Based on Discrete Wavelet Transformation
计算机科学, 2022, 49(6A): 434-440. https://doi.org/10.11896/jsjkx.210900199
[3] 胡伏原, 万新军, 沈鸣飞, 徐江浪, 姚睿, 陶重犇.
深度卷积神经网络图像实例分割方法研究进展
Survey Progress on Image Instance Segmentation Methods of Deep Convolutional Neural Network
计算机科学, 2022, 49(5): 10-24. https://doi.org/10.11896/jsjkx.210200038
[4] 程松盛, 潘金山.
基于深度学习特征匹配的视频超分辨率方法
Video Super-resolution Method Based on Deep Learning Feature Warping
计算机科学, 2021, 48(7): 184-189. https://doi.org/10.11896/jsjkx.200800224
[5] 石先让, 宋廷伦, 唐得志, 戴振泳.
一种新颖的单目视觉深度学习算法:H_SFPN
Novel Deep Learning Algorithm for Monocular Vision:H_SFPN
计算机科学, 2021, 48(4): 130-137. https://doi.org/10.11896/jsjkx.200400090
[6] 许虞俊, 李晨.
基于YOLO优化的轻量级目标检测网络
Light-weight Object Detection Network Optimized Based on YOLO Family
计算机科学, 2021, 48(11A): 265-269. https://doi.org/10.11896/jsjkx.201000152
[7] 许华杰, 杨洋, 李桂兰.
基于注意力机制和深度卷积神经网络的材质识别方法
Material Recognition Method Based on Attention Mechanism and Deep Convolutional Neural Network
计算机科学, 2021, 48(10): 220-225. https://doi.org/10.11896/jsjkx.200800073
[8] 陈庆超, 王韬, 尹世庄, 冯文博.
多级字典存储的未知文本协议候选关键词链式合并方法
Chain Merging Method for Unknown Text Protocol Candidate Keyword Stored in Multi-levelDictionary
计算机科学, 2020, 47(12): 332-335. https://doi.org/10.11896/jsjkx.190900116
[9] 纪明轩, 宋玉蓉.
一种基于对数位置表示和自注意力的机器翻译新模型
New Machine Translation Model Based on Logarithmic Position Representation and Self-attention
计算机科学, 2020, 47(11A): 86-91. https://doi.org/10.11896/jsjkx.200200003
[10] 曲佳, 时增林, 叶阳东.
基于卷积特征的非平衡人群密度估计方法
Unbalanced Crowd Density Estimation Based on Convolutional Features
计算机科学, 2018, 45(8): 236-241. https://doi.org/10.11896/j.issn.1002-137X.2018.08.042
[11] 雷倩,郝存明,张伟平.
基于超分辨率和深度神经网络的车型识别
Vehicle Recognition Based on Super-resolution and Deep Neural Networks
计算机科学, 2018, 45(6A): 230-233.
[12] 陈湘涛,肖碧文.
基于位置信息的显露序列模式挖掘研究
Emerging Sequences Pattern Mining Based on Location Information
计算机科学, 2017, 44(7): 175-179. https://doi.org/10.11896/j.issn.1002-137X.2017.07.031
[13] 王青芸,程春玲.
基于位置信息的移动SNS数据动态划分复制算法
Mobile SNS Data Dynamic Partitioning and Replication Algorithm Based on Location Information
计算机科学, 2017, 44(3): 220-225. https://doi.org/10.11896/j.issn.1002-137X.2017.03.046
[14] 庞松超,罗长远,韩东东,庞涵滢.
一种新的航空自组网混合路由算法
Aeronautical Ad hoc Network Hybrid Routing Algorithm
计算机科学, 2016, 43(5): 56-61. https://doi.org/10.11896/j.issn.1002-137X.2016.05.010
[15] 席瑞,李玉军,侯孟书.
室内定位方法综述
Survey on Indoor Localization
计算机科学, 2016, 43(4): 1-6. https://doi.org/10.11896/j.issn.1002-137X.2016.04.001
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!