当前位置>主页 > 期刊在线 > 物联网 >

物联网21年13期

基于数据增强和多层反向传播网络的行人重识别研究
罗锋
(中国人民解放军空军工程大学航空机务士官学校,河南 信阳 464000)

摘  要:行人重识别在模式识别中占据很大的比重,这项技术的目的是识别出不同摄像机在不同时间且处于不同环境下的行人是否为相同身份。为了更好地表达行人特征,提出了一种数据增强和多层反向传播网络的方法(DAML)。文章认为样本图像类型的稀少和深度网络传播过程中数据丢失是导致识别率低的重要因素。我们希望增加图像样本数量,让深度网络中的每一层都进行回传,以提高识别率。在 Market-1501、CUHK03 和 DukeMTMC-reID 等主流数据集上,我们的方法取得了较好的效果。


关键词:行人重识别;数据增强;多层反向传播;特征



DOI:10.19850/j.cnki.2096-4706.2021.13.044


中图分类号:TP391.4                                     文献标识码:A                                  文章编号:2096-4706(2021)13-0170-03


Pedestrian Re-Identification Research Based On Data Enhancement and Multiple Layer Back Propagation Network

LUO Feng

(Aviation Maintenance Sergeant School, PLA Air Force Engineering University, Xinyang 464000, China)

Abstract: Pedestrian re-identification occupies a large proportion in pattern recognition. The purpose of this technology is to identify whether pedestrians under different cameras at different times and in different environments have the same identity. In order to express the characteristics of pedestrians better, a data augmentation and multiple layer back propagation network method (DAML) is proposed. It is considered that the scarcity of sample image types and data loss in the process of deep network propagation are important factors leading to low recognition rate. We hope to increase the number of image samples and let each layer in the deep network pass back to improve the recognition rate. Our method has achieved good effects on mainstream data sets such as Market-1501, CUHK03 and DukeMTMC-reID.

Keywords: pedestrian re-identification; data enhancement; multiple layer back propagation; feature


参考文献:

[1] PEREZ L,WANG J. The Effectiveness of Data Augmentation in Image Classification using DeepLearning [J/OL].arXiv:1712.04621 [cs.CV].(2017-12-13).https://arxiv.org/abs/1712.04621.

[2] ZHONG Z,ZHENG L,KANG G L,et al. Random Erasing Data Augmentation [J/OL].arXiv:1708.04896 [cs.CV].(2017-11-16). https://arxiv.org/abs/1708.04896v2.

[3] ZHONG Z,ZHENG L,ZENG Z D,et al. Camera Style Adaptation for Person Re-identification [J/OL].arXiv:1711.10295 [cs. CV].(2017-11-28).https://arxiv.org/abs/1711.10295v2.

[4] WANG G S,YUAN Y F,CHEN X,et al.Learning Discriminative Features with Multiple Granularities for Person ReIdentification [J/OL].arXiv:1804.01438 [cs.CV].(2018-04-04). https://arxiv.org/abs/1804.01438

[5] KALAYEH M M,Basaran E,Muhittin G,et al. GokmenHuman Semantic Parsing for Person Re-identification [J/ OL].arXiv:1804.00216 [cs.CV].(2018-05-31).https://arxiv.org/ abs/1804.00216.

[6] JIA D,DONG W,SOCHER R,et al.ImageNet:A largescale hierarchical image database [C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.Miami:IEEE,2009: 248-255.

[7] HUANG G,LIU Z,MAATEN L V D,et al.Densely Connected Convolutional Networks [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE,2017:2261-2269.

[8] LI W,ZHAO R,XIAO T,et al. DeepReID:Deep Filter Pairing Neural Network for Person Re-identification [C]//2014 IEEE Conference on Computer Vision and Pattern Recognition.Columbus: IEEE,2014:152-159.

[9] ZHENG L,SHEN L Y,TIAN L,et al. Scalable Person Reidentification:A Benchmark [C]//2015 IEEE International Conference on Computer Vision (ICCV).Santiago:IEEE,2015:1116-1124.

[10] ZHENG Z D,ZHENG L,YANG Y. Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in Vitro [C]//2017 IEEE International Conference on Computer Vision (ICCV).Venice:2017,3774-3782.


作者简介:罗锋(1993—),男,汉族,河南光山人,助教, 硕士,研究方向:人工智能、模式识别。