A proposed method for computing convolution neural network algorithms
DOI:
https://doi.org/10.59992/IJCI.2026.v5n1p3Keywords:
Convolutional Neural Networks, Fast Convolution Algorithms, Infinite Length Input ProcessingAbstract
One kind of deep neural network that uses low-level input, such shapes and lines, to find more nuanced patterns is called a convolutional neural network (CNN). CNNs compare subsets of data using a kernel or filter through a variety of convolution processes. Training deep convolutional neural networks on large datasets requires days of GPU processing. Very low latency is necessary for self-driving automobiles to identify pedestrians. This study examines a suggested quick way to calculate the operations needed by convolutional neural network algorithms. In these situations, the speed at which convolution neural networks compute determines how well they perform. MATLAB is used to build this algorithm, and a single graphical user interface (GUI) window is used for all operations. When the approach is used instead of MATLAB's built-in functions, processing time is decreased. It saves 70% of the processing time when compared to the built-in features. This technique works on the widely recognized fast table method (FTM) principle.
References
1. Abtahi, T., Shea, C., Kulkarni, A., & Mohsenin, T. (2018). Accelerating convolutional neural network with fft on embedded hardware. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 26(9), 1737–1749.
2. Lavin, A., & Gray, S. (2016). Fast algorithms for convolutional neural networks. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 4013–4021). IEEE.
3. Alkadhim, S. A. S. (2020, July 9). Digital Convolution with Digital Signal Processing (DSP). Available at SSRN: https://ssrn.com/abstract=3647517.
4. Blahut, R. E. (2010). Fast algorithms for signal processing. Cambridge University Press.
5. Cariow, A., & Cariowa, G. (2020). Minimal Filtering Algorithms for Convolutional Neural Networks. arXiv: 2004.05607.
6. Cong, J., & Xiao, B. (2014). Minimizing computation in convolutional neural networks. In Artificial Neural Networks and Machine Learning–ICANN 2014 (pp. 281–290). Springer.
7. Di Curzio Lera, R., & de Carvalho Albertini, B. (2023). Hardware-efficient convolution algorithms for CNN accelerators: A brief review. Anais do Encontro Nacional de Inteligência Artificial e Computacional (ENIAC).
8. Gupta, S., Zhang, W., & Milthrope, J. (2015). Model accuracy and runtime tradeoff in distributed deep learning.
9. He, L., Zhao, Y., Gao, R., Du, Y., & Du, L. (2024). SFC: Achieve Accurate Fast Convolution under Low-precision Arithmetic. Proceedings of Machine Learning Research, (253).
10. Krishna, H. (2017). Digital Signal Processing Algorithms: Number Theory, Convolution, Fast Fourier Transforms, and Applications. Routledge.
11. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90.
12. Madisetti, V. (2010). The Digital Signal Processing Handbook (Vol. 2). CRC.
13. Nahar, A., & Khleaf, H. (2025). A FAST ALGORITHM FOR COMPUTING SHORT AND LONG –LENGTH LINEAR AND CIRCULAR DISCRETE CONVOLUTION. Kufa Journal of Engineering, 16(3), (2025) 485–498.
14. Proakis, J. G., & Manolakis, D. G. (2007). Digital Signal Processing: Principles, Algorithms, and Applications (4th ed.). Pearson Prentice Hall.
15. Rangayyan, R. M. (2015). Biomedical Signal Analysis: A Case-Study Approach. Wiley-IEEE Press.
16. Tan, L., & Jiang, J. (2018). Fundamentals of Digital Signal Processing: Using MATLAB, Lab VIEW and the Finite Element Method. Academic Press.
