Improved capsule network based on multipath feature

doi:10.11772/j.issn.1001-9081.2022030367

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (5): 1330-1335.DOI: 10.11772/j.issn.1001-9081.2022030367

Special Issue: 第九届中国数据挖掘会议(CCDM 2022)

• China Conference on Data Mining 2022 (CCDM 2022) • Previous Articles Next Articles

Improved capsule network based on multipath feature

Qinghai XU¹, Shifei DING¹^,²(), Tongfeng SUN¹^,², Jian ZHANG¹^,², Lili GUO¹^,²

^1.School of Computer Science and Technology，China University of Mining and Technology，Xuzhou Jiangsu 221116，China
^2.Engineering Research Center of Mine Digitization，Ministry of Education （China University of Mining and Technology），Xuzhou Jiangsu 221116，China

Received:2022-03-22 Revised:2022-04-20 Accepted:2022-04-28 Online:2023-05-08 Published:2023-05-10
Contact: Shifei DING
About author:XU Qinghai， born in 1998， M. S. candidate. His research interests include deep learning， neural network.
DING Shifei， born in 1963， Ph. D.， professor. His research interests include artificial intelligence， pattern recognition， machine learning， data mining.
SUN Tongfeng， born in 1977， Ph. D.， associate professor. His research interests include machine learning， neural network， deep learning.
ZHANG Jian， born in 1990， Ph. D.， lecturer. His research interests include machine learning， deep learning.
GUO Lili， born in 1990， Ph. D.， lecturer. Her research interests include deep learning， multimodal emotion recognition.
Supported by:
National Natural Science Foundation of China(61976216)

改进的基于多路径特征的胶囊网络

徐清海¹, 丁世飞¹^,²(), 孙统风¹^,², 张健¹^,², 郭丽丽¹^,²

^1.中国矿业大学计算机科学与技术学院, 江苏徐州 221116
^2.矿山数字化教育部工程技术研究中心(中国矿业大学), 江苏徐州 221116

通讯作者: 丁世飞
作者简介:徐清海（1998—），男，江西景德镇人，硕士研究生，主要研究方向：深度学习、神经网络
丁世飞（1963—），男，山东青岛人，教授，博士，CCF会员，主要研究方向：人工智能、模式识别、机器学习、数据挖掘 dingsf@cumt.edu.cn
孙统风（1977—），男，江苏徐州人，副教授，博士，CCF会员，主要研究方向：机器学习、神经网络、深度学习
张健（1990—），男，山东泰安人，讲师，博士，CCF会员，主要研究方向：机器学习、深度学习
郭丽丽（1990—），女，山东临沂人，讲师，博士，CCF会员，主要研究方向：深度学习、多模态情感识别。
基金资助:
国家自然科学基金资助项目(61976216)

Abstract

Abstract:

Concerning the problems of poor classification of Capsule Network （CapsNet） on complex datasets and large number of parameters in the routing process， a Capsule Network based on Multipath feature （MCNet） was proposed， including a novel capsule feature extractor and a novel capsule pooling method. By the capsule feature extractor， the features of different layers and locations were extracted in parallel from multiple paths， and then the features were encoded into capsule features containing more semantic information. In the capsule pooling method， the most active capsules at each position of the capsule feature map were selected， and the effective capsule features were represented by a small number of capsules. Comparisons were performed on four datasets （CIFAR-10， SVHN， Fashion-MNIST， MNIST） with models such as CapsNet. Experimental results show that MCNet has the classification accuracy of 79.27% on CIFAR-10 dataset and the number of trainable parameters of 6.25×10⁶； compared with CapsNet， MCNet has the classification accuracy improved by 8.7%， and the number of parameters reduced by 46.8%. MCNet can effectively improve the classification accuracy while reducing the number of trainable parameters.

Key words: Capsule Network (CapsNet), deep learning, dynamic routing, capsule pooling, deconvolutional reconstruction

摘要：

针对胶囊网络（CapsNet）在复杂数据集上的分类效果差，而且在路由过程中参数数量过大等问题，提出一种基于多路径特征的胶囊网络（MCNet），包含新的胶囊特征提取器和新的胶囊池化方法。该胶囊特征提取器从多个不同路径中并行地提取不同层次、不同位置的特征，然后将特征编码为包含更多语义信息的胶囊特征；胶囊池化方法则在胶囊特征图的每个位置选取最活跃的胶囊，用少量的胶囊表示有效的胶囊特征。在4个数据集（CIFAR-10、SVHN、Fashion-MNIST、MNIST）上与CapsNet等模型进行了对比。实验结果显示，MCNet在CIFAR-10数据集上的分类准确率为79.27%，可训练的参数数量为6.25×10⁶，与CapsNet相比，MCNet的分类准确率提升了8.7%，参数数量减少了46.8%。MCNet能够有效提升分类准确率，同时减少可训练的参数数量。

关键词: 胶囊网络, 深度学习, 动态路由, 胶囊池化, 反卷积重构

CLC Number:

TP183

Qinghai XU, Shifei DING, Tongfeng SUN, Jian ZHANG, Lili GUO. Improved capsule network based on multipath feature[J]. Journal of Computer Applications, 2023, 43(5): 1330-1335.

徐清海, 丁世飞, 孙统风, 张健, 郭丽丽. 改进的基于多路径特征的胶囊网络[J]. 《计算机应用》唯一官方网站, 2023, 43(5): 1330-1335.

Figures/Tables 8

References 29

1	KRIZHEVSKY A， SUTSKEVER I， HINTON G E. ImageNet classification with deep convolutional neural networks［C］// Proceedings of the 25th International Conference on Neural Information Processing Systems — Volume 1. Red Hook， NY： Curran Associates Inc.， 2012： 1097-1105.
2	HINTON G E， KRIZHEVSKY A， WANG S D. Transforming auto-encoders［C］// Proceedings of the 2011 International Conference on Artificial Neural Networks， LNTCS 6791. Berlin： Springer， 2011： 44-51.
3	SABOUR S， FROSST N， HINTON G E. Dynamic routing between capsules［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2017： 3859-3869.
4	HINTON G， SABOUR S， FROSST N. Matrix capsules with EM routing［EB/OL］. （2023-01-24）［2023-01-30］..
5	KOSIOREK A R， SABOUR S， TEH Y W， et al. Stacked capsule autoencoders［C/OL］// Proceedings of the 33rd Conference on Neural Information Processing System ［2022-02-10］..
6	XIANG H L， HUANG Y S， LEE C H， et al. 3-D Res-CapsNet convolutional neural network on automated breast ultrasound tumor diagnosis［J］. European Journal of Radiology， 2021， 138： No.109608. 10.1016/j.ejrad.2021.109608
7	AFSHAR P， NADERKHANI F， OIKONOMOU A， et al. MIXCAPS： a capsule network-based mixture of experts for lung nodule malignancy prediction［J］. Pattern Recognition， 2021， 116： No.107942. 10.1016/j.patcog.2021.107942
8	PANIGRAHI S， DAS J， SWARNKAR T. Capsule network based analysis of histopathological images of oral squamous cell carcinoma［J］. Journal of King Saud University - Computer and Information Sciences， 2022， 34（7）： 4546-4553. 10.1016/j.jksuci.2020.11.003
9	SUN G C， DING S F， SUN T F， et al. SA-CapsGAN： using Capsule Networks with embedded self-attention for Generative Adversarial Network［J］. Neurocomputing， 2021， 423： 399-406. 10.1016/j.neucom.2020.10.092
10	JAISWAL A， AbdALMAGEED W， WU Y， et al. CapsuleGAN： generative adversarial capsule network［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11131. Cham： Springer， 2019： 526-535.
11	JACOB I J. Performance evaluation of Caps-Net based multitask learning architecture for text classification［J］. Journal of Artificial Intelligence and Capsule Networks， 2020， 2（1）： 1-10. 10.36548/jaicn.2020.1.001
12	WU Y J， LI J， WU J， et al. Siamese capsule networks with global and local features for text classification［J］. Neurocomputing， 2020， 390： 88-98. 10.1016/j.neucom.2020.01.064
13	LALONDE R， XU Z Y， IRMAKCI I， et al. Capsules for biomedical image segmentation［J］. Medical Image Analysis， 2021， 68： No.101889. 10.1016/j.media.2020.101889
14	DUARTE K， RAWAT Y S， SHAH M. VideoCapsuleNet： a simplified network for action detection［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2018： 7621-7630.
15	HUA Q， WEI L L， DONG C R， et al. Improved variational inference with dynamic routing flow［J］. International Journal of Machine Learning and Cybernetics， 2020， 11（2）： 301-312. 10.1007/s13042-019-00974-x
16	WANG Y H， XIAO W D， TAN Z， et al. Caps-OWKG： a capsule network model for open-world knowledge graph［J］. International Journal of Machine Learning and Cybernetics， 2021， 12（6）： 1627-1637. 10.1007/s13042-020-01259-4
17	XIONG Y， SU G P， YE S W， et al. Deeper capsule network for complex data［C］// Proceedings of the 2019 International Joint Conference on Neural Networks. Piscataway： IEEE， 2019： 1-8. 10.1109/ijcnn.2019.8852020
18	XIANG C Q， ZHANG L， TANG Y， et al. MS-CapsNet： a novel multi-scale capsule network［J］. IEEE Signal Processing Letters， 2018， 25（12）： 1850-1854. 10.1109/lsp.2018.2873892
19	AMER M， MAUL T. Path capsule networks［J］. Neural Processing Letters， 2020， 52（1）： 545-559. 10.1007/s11063-020-10273-0
20	ROSARIO V M DO， BORIN E， BRETERNITZ M. The multi-lane capsule network［J］. IEEE Signal Processing Letters， 2019， 26（7）： 1006-1010. 10.1109/lsp.2019.2915661
21	TISSERA D， VITHANAGE K， WIJESINGHE R， et al. Feature-dependent cross-connections in multi-path neural networks［C］// Proceedings of the 25th International Conference on Pattern Recognition. Piscataway： IEEE， 2021： 4032-4039. 10.1109/icpr48806.2021.9413187
22	PHAYE S S R， SIKKA A， DHALL A， et al. Multi-level dense capsule networks［C］// Proceedings of the 2018 Asian Conference on Computer Vision， LNCS 11365. Cham： Springer， 2019： 577-592.
23	SUN G C， DING S F， SUN T F， et al. A novel dense capsule network based on dense capsule layers［J］. Applied Intelligence， 2022， 52（3）： 3066-3076. 10.1007/s10489-021-02630-w
24	REN Q， SHANG S H， HE L H. Adaptive routing between capsules［EB/OL］. ［2021-11-09］.. 10.1109/iccvw.2019.00247
25	GUO X F. CapsNet-Pytorch： Pytorch implementation for NIPS2017 paper ‘Dynamic Routing Between Capsules’［EB/OL］. ［2021-11-30］..
26	XI E， BING S， JIN Y. Capsule network performance on complex data［EB/OL］. ［2021-10-14］..
27	NAIR P， DOSHI R， KESELJ S. Pushing the limits of capsule networks［EB/OL］. ［2022-01-08］..
28	DELIÈGE A， CIOPPA A， VAN DROOGENBROECK M. HitNet： a neural network with capsules embedded in a Hit-or-Miss layer， extended with hybrid data augmentation and ghost capsules［EB/OL］. ［2021-11-30］.. 10.48550/arXiv.1806.06519
29	ZHAO Z， KLEINHANS A， SANDHU G， et al. Capsule networks with Max-Min normalization［EB/OL］. ［2022-02-18］..

模型	CIFAR-10	SVHN	Fashion-MNIST	MNIST
CapsNet	72.90	93.76	92.52	99.70
文献［26］模型	71.52	—	—	—
文献［27］模型	67.53	91.06	89.80	99.50
HitNet	73.30	94.50	92.30	99.68
文献［29］模型	75.92	—	92.07	99.55
MLCN2	75.18	—	92.63	—
MS-CapsNet	75.70	—	92.70	—
PathCapsNet	69.03	—	—	99.62
文献［24］模型	78.41	94.27	93.01	99.65
MCNet	79.27	94.70	93.17	99.63

模型	CIFAR-10	SVHN	Fashion-MNIST	MNIST
CapsNet	72.90	93.76	92.52	99.70
文献［26］模型	71.52	—	—	—
文献［27］模型	67.53	91.06	89.80	99.50
HitNet	73.30	94.50	92.30	99.68
文献［29］模型	75.92	—	92.07	99.55
MLCN2	75.18	—	92.63	—
MS-CapsNet	75.70	—	92.70	—
PathCapsNet	69.03	—	—	99.62
文献［24］模型	78.41	94.27	93.01	99.65
MCNet	79.27	94.70	93.17	99.63

模型	CIFAR-10		Fashion-MNIST
模型	准确率/%	参数量/10⁶	准确率/%	参数量/10⁶
CapsNet	72.90	11.75	92.52	8.22
MLCN2	75.18	14.25	92.63	10.63
MS-CapsNet	75.70	11.20	92.70	10.80
MCNet	79.27	6.25	93.17	5.88

模型	CIFAR-10		Fashion-MNIST
模型	准确率/%	参数量/10⁶	准确率/%	参数量/10⁶
CapsNet	72.90	11.75	92.52	8.22
MLCN2	75.18	14.25	92.63	10.63
MS-CapsNet	75.70	11.20	92.70	10.80
MCNet	79.27	6.25	93.17	5.88

[1]	Jing QIN, Zhiguang QIN, Fali LI, Yueheng PENG. Diagnosis of major depressive disorder based on probabilistic sparse self-attention neural network [J]. Journal of Computer Applications, 2024, 44(9): 2970-2974.
[2]	Xiyuan WANG, Zhancheng ZHANG, Shaokang XU, Baocheng ZHANG, Xiaoqing LUO, Fuyuan HU. Unsupervised cross-domain transfer network for 3D/2D registration in surgical navigation [J]. Journal of Computer Applications, 2024, 44(9): 2911-2918.
[3]	Yexin PAN, Zhe YANG. Optimization model for small object detection based on multi-level feature bidirectional fusion [J]. Journal of Computer Applications, 2024, 44(9): 2871-2877.
[4]	Shunyong LI, Shiyi LI, Rui XU, Xingwang ZHAO. Incomplete multi-view clustering algorithm based on self-attention fusion [J]. Journal of Computer Applications, 2024, 44(9): 2696-2703.
[5]	Yunchuan HUANG, Yongquan JIANG, Juntao HUANG, Yan YANG. Molecular toxicity prediction based on meta graph isomorphism network [J]. Journal of Computer Applications, 2024, 44(9): 2964-2969.
[6]	Yuhan LIU, Genlin JI, Hongping ZHANG. Video pedestrian anomaly detection method based on skeleton graph and mixed attention [J]. Journal of Computer Applications, 2024, 44(8): 2551-2557.
[7]	Yanjie GU, Yingjun ZHANG, Xiaoqian LIU, Wei ZHOU, Wei SUN. Traffic flow forecasting via spatial-temporal multi-graph fusion [J]. Journal of Computer Applications, 2024, 44(8): 2618-2625.
[8]	Qianhong SHI, Yan YANG, Yongquan JIANG, Xiaocao OUYANG, Wubo FAN, Qiang CHEN, Tao JIANG, Yuan LI. Multi-granularity abrupt change fitting network for air quality prediction [J]. Journal of Computer Applications, 2024, 44(8): 2643-2650.
[9]	Yiqun ZHAO, Zhiyu ZHANG, Xue DONG. Anisotropic travel time computation method based on dense residual connection physical information neural networks [J]. Journal of Computer Applications, 2024, 44(7): 2310-2318.
[10]	Song XU, Wenbo ZHANG, Yifan WANG. Lightweight video salient object detection network based on spatiotemporal information [J]. Journal of Computer Applications, 2024, 44(7): 2192-2199.
[11]	Xun SUN, Ruifeng FENG, Yanru CHEN. Monocular 3D object detection method integrating depth and instance segmentation [J]. Journal of Computer Applications, 2024, 44(7): 2208-2215.
[12]	Zheng WU, Zhiyou CHENG, Zhentian WANG, Chuanjian WANG, Sheng WANG, Hui XU. Deep learning-based classification of head movement amplitude during patient anaesthesia resuscitation [J]. Journal of Computer Applications, 2024, 44(7): 2258-2263.
[13]	Huanhuan LI, Tianqiang HUANG, Xuemei DING, Haifeng LUO, Liqing HUANG. Public traffic demand prediction based on multi-scale spatial-temporal graph convolutional network [J]. Journal of Computer Applications, 2024, 44(7): 2065-2072.
[14]	Zhi ZHANG, Xin LI, Naifu YE, Kaixi HU. DKP： defending against model stealing attacks based on dark knowledge protection [J]. Journal of Computer Applications, 2024, 44(7): 2080-2086.
[15]	Yajuan ZHAO, Fanjun MENG, Xingjian XU. Review of online education learner knowledge tracing [J]. Journal of Computer Applications, 2024, 44(6): 1683-1698.

Improved capsule network based on multipath feature

改进的基于多路径特征的胶囊网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 29

Related Articles 15

Recommended Articles

Metrics

模型	胶囊特征提取器	胶囊池化	准确率/%
a	—	—	71.34
b	—	√	72.06
c	√	—	79.34
d	√	√	79.27

模型	胶囊特征提取器	胶囊池化	准确率/%
a	—	—	71.34
b	—	√	72.06
c	√	—	79.34
d	√	√	79.27