Abstract
Capsule network is a novel network structure based on vector, which has excellent performance in image classification and attracted the attention of researchers. However, the capsule network has a large amount of computation and does not perform well in the task of classifying complex datasets. Therefore, we propose a capsule network based on deep routing and residual learning. The network reduces the number of capsules to reduce the computation during routing, adds local routing layers to extract deep hidden capsule information and constructs residual connections to assist parameters training. In addition, we propose a new capsule dropout method to enhance the robustness of the network and reduce the computation of the network. We conducted several experiments on multiple benchmark datasets (CIFAR-10, SVHN, Fashion-MNIST and MNIST). The experimental results show that the proposed method achieves competitive results on multiple benchmark datasets. For example, the accuracy of 81.61% is achieved on the CIFAR-10 dataset, and the number of training parameters is reduced by 72.09%.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00500-023-08018-x/MediaObjects/500_2023_8018_Fig8_HTML.png)
Similar content being viewed by others
Data availability
Enquiries about data availability should be directed to the authors.
References
Afshar P, Naderkhani F, Oikonomou A et al (2021) Mixcaps: a capsule network-based mixture of experts for lung nodule malignancy prediction. Pattern Recognit 116:107942
Amer M, Maul T (2020) Path capsule networks. Neural Process Lett 52:545–559
Cheng X, He J, He J et al (2019) Cv-CapsNet: complex-valued capsule network. IEEE Access 7:85492–85499
Choi J, Seo H, Im S (2019) Attention routing between capsules. In: proceedings of the IEEE international conference on computer vision workshops, Seoul, pp 1981–1989
Deliege A, Cioppa A, Droogenbroeck MV (2018) Hitnet: a neural network with capsules embedded in a hit-or-miss layer, extended with hybrid data augmentation and ghost capsules. ar**v preprint ar**v:1806.06519
Dhanaraj RK, Lalitha K, Anitha S et al (2021a) Hybrid and dynamic clustering based data aggregation and routing for wireless sensor networks. J Intell Fuzzy Syst 40(6):10751–10765
Dhanaraj RK, Lalitha K, Oana G et al (2021b) Black hole and sink hole attack detection in wireless body area networks. CMC-Comput Mater Con 68(2):1949–1965
Dhanaraj RK, Ramakrishnan V, Poongodi M et al (2021c) Random forest bagging and X-means clustered antipattern detection from SQL query log for accessing secure mobile data. Wirel Commun Mob Comput 2021:2730246
Duarte K, Rawat Y S, Shah M (2018) VideoCapsuleNet: a simplified network for action detection. In: Proceedings of the 32nd international conference on neural information processing systems, Montréal, pp 7621–7630
Guo X (2017) A Keras implementation of CapsNet in NIPS2017 paper "Dynamic routing between capsules". https://github.com/**fengGuo/CapsNet-Keras
He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, pp 770–778
Hinton GE, Krizhevsky A, Wang S D (2011) Transforming auto-encoders. In: International conference on artificial neural networks, Berlin, Heidelberg, pp 44–51
Hinton GE, Sabour S, Frosst N (2018) Matrix capsules with EM routing. In: International conference on learning representations, Vancouver, pp 1–15
Hua Q, Wei L, Dong C et al (2020) Improved variational inference with dynamic routing flow. Int J Mach Learn Cyb 11(2):301–312
Kosiorek AR, Sabour S, Teh YW et al (2019) Stacked capsule autoencoders. Adv Neural Inf Process Syst 32:1–11
Krishnamoorthi S, Jayapaul P, Dhanaraj RK et al (2021) Design of pseudo-random number generator from turbulence padded chaotic map. Nonlinear Dyn 104(2):1627–1643
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105
LaLonde R, Xu Z, Irmakci I et al (2021) Capsules for biomedical image segmentation. Med Image Anal 68:101889
Mazzia V, Salvetti F, Chiaberge M (2021) Efficient-CapsNet: capsule network with self-attention routing. ar**v preprint ar**v:2101.12491
Nair P, Doshi R, Keselj S (2021) Pushing the limits of capsule networks. ar**v preprint ar**v:2103.08074
Phaye SSR, Sikka A, Dhall A, Bathula DR (2018) Multi-level dense capsule networks. In: Asian conference on computer vision, Perth, pp 577–592
Rajasegaran J, Jayasundara V, Jayasekara S et al (2019) Deepcaps: going deeper with capsule networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Long Beach, pp 10725–10733
Ren Q, Shang S, He L (2019) Adaptive routing between capsules. ar**v preprint ar**v:1911.08119
Rosario VMD, Borin E, Breternitz M (2019) The multi-lane capsule network. IEEE Signal Process Lett 26(7):1006–1010
Sabour S, Frosst N, Hinton GE (2017) Dynamic routing between capsules. In: Proceedings of the 31st international conference on neural information processing systems, California, pp 3859–3869
Saravanakumar C, Senthilvel P, Thirupurasundari D et al (2021) Plant syndrome recognition by Gigapixel Image using convolutional neural network. In: Proceedings of the first international conference on advanced scientific innovation in science, Engineering and Technology (ICASISET), Chennai
Sathyamoorthy M, Kuppusamy S, Dhanaraj RK et al (2022) Improved K-means based Q learning algorithm for optimal clustering and node balancing in WSN. Wireless Pers Commun 122:2745–2766
Sun G, Ding S, Sun T et al (2021a) SA-CapsGAN: using capsule networks with embedded self-attention for generative adversarial network. Neurocomputing 423:399–406
Sun K, Xu H, Yuan L et al (2021b) A tiny deep capsule network. Int J Mach Learn Cyb 11:301–312
Sun G, Ding S, Sun T et al (2022) A novel dense capsule network based on dense capsule layers. Appl Intell 52(3):3066–3076
Vinothsaravanan R, Palanisamy C, Rajesh KD et al (2021) Bootstrap aggregative mean shift clustering for big data anti-pattern detection analytics in 5G/6G communication networks. Comput Electr Eng 95:107380
Wang Y, **ao W, Tan Z et al (2021) Caps-OWKG: a capsule network model for open-world knowledge graph. Int J Mach Learn Cyb 12(6):1627–1637
Wu Y, Li J, Wu J et al (2020) Siamese capsule networks with global and local features for text classification. Neurocomputing 390:88–98
** E, Bing S, ** Y (2017) Capsule network performance on complex data. ar**v preprint ar**v:1712.03480
**ang C, Zhang L, Tang Y et al (2018) MS-CapsNet: a novel multi-scale capsule network. IEEE Signal Process Lett 25(12):1850–1854
**ong Y, Su G, Ye S et al (2019) Deeper capsule network for complex data. In: International joint conference on neural networks, Rio, pp 1–8
Zhao Z, Kleinhans A, Sandhu G, et al (2019) Capsule networks with max-min normalization. ar**v preprint ar**v:1903.09662
Acknowledgements
This work is supported by the National Natural Science Foundations of China (no.61976216, no.62276265 and no.62206297).
Funding
The authors have not disclosed any funding.
Author information
Authors and Affiliations
Contributions
All authors contributed to the writing of the manuscript and final approval.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Informed consent
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5). Additional informed consent was obtained from all patients for which identifying information is included in this article.
Human and animal rights
This article does not contain any studies with human or animal subjects performed by any of the authors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, J., Xu, Q., Guo, L. et al. A novel capsule network based on deep routing and residual learning. Soft Comput 27, 7895–7906 (2023). https://doi.org/10.1007/s00500-023-08018-x
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-023-08018-x