Convolutional Neural Networks based on Random Kernels in the Frequency Domain
- Authors
- Han, Y.; Derbel, B.; Hong, B.-W.
- Issue Date
- Jan-2021
- Publisher
- IEEE Computer Society
- Citation
- International Conference on Information Networking, v.2021-January, pp 671 - 673
- Pages
- 3
- Journal Title
- International Conference on Information Networking
- Volume
- 2021-January
- Start Page
- 671
- End Page
- 673
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/44036
- DOI
- 10.1109/ICOIN50884.2021.9333914
- ISSN
- 1976-7684
- Abstract
- Image classification in Fourier domain has been researched for many years via deep learning process, especially for spectral pooling methods and visualization. Point-wise multiplication of Fourier transformed image and kernel has solved high computational cost which is required for convolution operation through Convolutional Neural Network (CNNs) in spatial domain. However, there is still an open problem to deal with kernel method in the Fourier domain because larger images need bigger amount of computational cost by using the same size of kernel. In this work, we propose an efficient discrete Fourier transform-based CNNs using sparse random kernel. we expect the sparse random kernel contains critical low frequency and high frequency contents, but many zeros in kernel affect to the lower cost computation in the Frequency domain. The evaluation was conducted using the benchmark MNIST datasets using LeNet-5 (LeCun) which showed the possibility of this work, so we can expect that the implementation can be expanded for the future work. © 2021 IEEE.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Software > Department of Artificial Intelligence > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.