Abstract
Image classification has become a major task in the area of computer vision technologies. Capsule network (CapsNets), a new technique that uses activity vectors and dynamic routing between capsules to address some of the constraints of convolutional neural networks(CNNs), was recently developed. This work compares the performance of the CapsNets algorithm to four well-known datasets in this study (FMNIST, CIFAR-10, Brain tumor MRI, and blood cell dataset). We used the datasets with varying numbers of instances and classes, such as blood cell images and everyday objects, to test classification accuracies and losses. The results reveal that even for simple architectures, training the CapsNets algorithm necessitates a significant amount of computational resources, with Sabour’s model as a baseline, generalizing well on small datasets and performing poorly on complex images. We suggest, however, that CapsNets appears to be a promising new technique for image classification, and that future experimentation with more resilient computational resources and modified CapsNet structures should yield superior results.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
All data generated or analysed during this study are included in this manuscript (and its supplementary information files).
Code availability
The code used for this study is a modification of the code at https://www.github.com/Xifeng- Guo/CapsNet-Keras.
References
Afriyie Y, Weyori BA, Opoku AA (2022) Classification of Blood Cells Using Optimized Capsule Networks. Neural Process Lett. https://doi.org/10.1007/s11063-022-10833-6
Ayidzoe MA, Yu Y, Mensah PK, Cai J, Kwabena A, Tashi N (2021a) Feature amplification capsule network for complex images. J Intell Fuzzy Syst 40(6):10955–10968. https://doi.org/10.3233/JIFS-202080
Ayidzoe MA, Yu Y, Kwabena P, Jingye M, Kwabena C, Yifan A (2021b) Gabor capsule network with preprocessing blocks for the recognition of complex images. Mach Vis Appl 123 https://doi.org/10.1007/s00138-021-01221-6
Carneiro G, Zheng Y, Xing F, Yang L (2017) Review of deep learning methods in mammography, cardiovascular, and microscopy image analysis. In Deep Learning and Convolutional Neural Networks for Medical Image Computing (pp. 11-32). Springer, Cham. https://doi.org/10.1007/978-3-319-42999-1_2
Chang S, Liu JIN (2020) Multi-Lane Capsule Network for Classifying Images With Complex Background. 8 https://doi.org/10.1109/ACCESS.2020.2990700
Chen Y, Jiang H, Li C, Jia X, Member S (2016) Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks. 54(10):6232–6251
Gunjan VK, Garcia Diaz V, Cardona M, Solanki VK, Sunitha KVN (eds) (2020) ICICCT 2019 – System Reliability, Quality Control, Safety, Maintenance and Management. https://doi.org/10.1007/978-981-13-8461-5; Retrieved from https://sci-hub.se/10.1007/978-981-13-8461-5
Hahn T, Pyeon M, Kim G (2019) Self-routing capsule networks. Advances in neural information processing systems, vol 32
Hajian-Tilaki K (2013) Receiver operating characteristic (ROC) curve analysis for medical diagnostic test evaluation. Casp J Intern Med 4(2):627–635
Hinton G, Sabour S, Frosst N (2018) Matrix capsules with em routing. pp. 1–15
Krizhevsky A, Hinton G (2009) Learning multiple layers of features from tiny images. Retrieved from http://www.cs.utoronto.ca/~kriz/learning-features-2009-TR.pdf
Kwabena Patrick M, Felix Adekoya A, Abra Mighty A, Edward BY (2019) Capsule Networks – A survey. J King Saud Univ - Comput Inf Sci (xxxx) https://doi.org/10.1016/j.jksuci.2019.09.014
Mensah P, Asubam B, Abra A (2020) Exploring the performance of LBP-capsule networks with K-Means routing on complex images. J King Saud Univ - Comput Inf Sci, (xxxx) https://doi.org/10.1016/j.jksuci.2020.10.006
Parvar MN (2021) https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset. Accessed 15 Mar 2022
Peer D, Stabinger S, Rodríguez-Sánchez A (2021) Limitation of capsule networks. Pattern Recognit Lett 144:68–74. https://doi.org/10.1016/j.patrec.2021.01.017
Rajasegaran J (nd) “DeepCaps: Going Deeper with Capsule Networks.”
Rawlinson D, Ahmed A, Kowadlo G (2018) Sparse Unsupervised Capsules Generalize Better. [Online]. Available: http://arxiv.org/abs/1804.06094
Sabour S, Frosst N, Hinton GE (2017) Dynamic routing between capsules. In Advances in Neural Information Processing Systems (NIPS), pp 3857–3867
Suganthan PN (2020) Journa lP. Expert Syst Appl 114161 https://doi.org/10.1016/j.eswa.2020.114161
Sun K, Yuan L, Xu H, Wen X (2020) Deep tensor capsule network. IEEE Access 8:96920–96933
Sun K, Wen X, Yuan L, Xu H (2021) Dense capsule networks with fewer parameters. Soft Comput 25(10):6927–6945. https://doi.org/10.1007/s00500-021-05774-6
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc. pp. 1–14
Tajbakhsh N et al (2016) Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? IEEE Trans Med Imaging 35(5):1299–1312. https://doi.org/10.1109/TMI.2016.2535302
Xiao H, Rasul K, Vollgraf R (2017) Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747. Retrieved from https://arxiv.org/pdf/1708.07747.pdf
Xi E, Bing S, Jin Y (2017) Capsule Network Performance on Complex Data. arXiv preprint arXiv:1712.03480
Author information
Authors and Affiliations
Contributions
All Authors contributed equally to the preparation of this study.
Corresponding author
Ethics declarations
Competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethics approval
Not applicable.
Consent to participate
Not applicable.
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
Afriyie, Y., Weyori, B.A. & Opoku, A.A. Comparative evaluation performances of capsule networks for complex image classification. J. of Data, Inf. and Manag. 4, 267–276 (2022). https://doi.org/10.1007/s42488-022-00079-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42488-022-00079-1