Abstract
Although complex-valued convolutional neural networks (CVCNNs) have shown an improvement over their real-valued counterparts (RVCNNs) when trained on real-valued images, in order to harness the full potential of CVCNNs, they should be used with fully complex-valued inputs. Because every image has one and only one Fourier transform, the problem of classifying real-valued images is equivalent to the problem of classifying their complex-valued Fourier transforms. Experiments done using the MNIST, SVHN, and CIFAR-10 datasets show an improved performance of CVCNNs trained on Fourier-transformed images over CVCNNs trained on real-valued images (for which the imaginary part of the input is considered to be zero), and over RVCNNs.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Arjovsky, M., Shah, A., Bengio, Y.: Unitary evolution recurrent neural networks. In: ICML 2016 (2016). https://arxiv.org/abs/1511.06464
Bengio, Y.: Learning deep architectures for AI. Found. Trends Mach. Learn. 2(1), 1–127 (2009)
Bengio, Y., LeCun, Y., Hinton, G.: Deep learning. Nature 521, 436–444 (2015)
Chaturvedi, A., Sharma, R., Wadekar, D., Bhandwalkar, A., Shitole, S.: Adaptive parametric estimator for complex valued images. In: International Conference on Technologies for Sustainable Development (ICTSD) (2015)
Eichel, P., Ives, R.: Compression of complex-valued SAR images. IEEE Trans. Image Process. 8(10), 1483–1487 (1999)
Gonzalez, R.C., Woods, R.E.: Digital Image Processing. Pearson Prentice Hall, Upper Saddle River (2008)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016)
Guberman, N.: On Complex Valued Convolutional Neural Networks. Master’s thesis, School of Computer Science and Engineering. The Hebrew University of Jerusalem (2016). https://arxiv.org/abs/1602.09046
Hansch, R., Hellwich, O.: Complex-valued convolutional neural networks for object detection in PolSAR data. In: European Conference on Synthetic Aperture Radar (EUSAR) (2010)
Hernandez-Garcia, L., Vazquez, A., Rowe, D.: Complex-valued analysis of arterial spin labeling based FMRI signals. Magn. Reson. Med. 62(6), 1597–1608 (2009)
Hui, Y., Smith, M.: MRI reconstruction from truncated data using a complex domain backpropagation neural network. In: Pacific Rim Conference on Communications, Computers, and Signal Processing (PACRIM) (1995)
Krizhevsky, A.: Learning multiple layers of features from tiny images. Technical report (2009)
LeCun, Y., Boser, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W., Jackel, L.D.: Handwritten digit recognition with a back-propagation network. In: Advances in Neural Information Processing Systems (NIPS) (1989)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Netzer, Y., Wang, T., Coates, A., Bissacco, A., Wu, B., Ng, A.Y.: Reading digits in natural images with unsupervised feature learning. In: NIPS Workshop on Deep Learning and Unsupervised Feature Learning (2011)
Popa, C.A.: Complex-valued convolutional neural networks for real-valued image classification. In: 2017 International Joint Conference on Neural Networks (IJCNN). IEEE May 2017
Samadi, S., Cetin, M., Masnadi-Shirazi, M.: Sparse signal representation for complex-valued imaging. In: Digital Signal Processing Workshop and Signal Processing Education Workshop (DSP/SPE), pp. 365–370 (2009)
Schmidhuber, J.: Deep learning in neural networks: an overview. Neural Netw. 61, 85–117 (2015)
Tygert, M., Bruna, J., Chintala, S., LeCun, Y., Piantino, S., Szlam, A.: A mathematical motivation for complex-valued convolutional networks. Neural Comput. 28(5), 815–825 (2016)
Virtue, P., Yu, S.X., Lustig, M.: Better than real: complex-valued neural nets for MRI fingerprinting. In: ICIP 2017 (2017). https://arxiv.org/abs/1707.00070
Wilmanski, M., Kreucher, C., Hero, A.: Complex input convolutional neural networks for wide angle SAR ATR. In: 2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE, DEC 2016
Yu, M.C., Lina, Q.H., Kuang, L.D., Gong, X.F., Cong, F., Calhoun, V.: ICA of full complex-valued fMRI data using phase information of spatial maps. J. Neurosci. Methods 249, 75–91 (2015)
Acknowledgement
This work was supported by research grant no. PCD-TC-2017-41 of the Polytechnic University Timişoara, Romania.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Popa, CA., Cernăzanu-Glăvan, C. (2018). Fourier Transform-Based Image Classification Using Complex-Valued Convolutional Neural Networks. In: Huang, T., Lv, J., Sun, C., Tuzikov, A. (eds) Advances in Neural Networks – ISNN 2018. ISNN 2018. Lecture Notes in Computer Science(), vol 10878. Springer, Cham. https://doi.org/10.1007/978-3-319-92537-0_35
Download citation
DOI: https://doi.org/10.1007/978-3-319-92537-0_35
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-92536-3
Online ISBN: 978-3-319-92537-0
eBook Packages: Computer ScienceComputer Science (R0)