Korean Journal of Artificial Intelligence (한국인공지능학회지)

Korea Artificial Intelligence Association (한국인공지능학회)
권호 표지
DOI QR코드

DOI QR Code

Efficient Large Dataset Construction using Image Smoothing and Image Size Reduction

  • Jaemin HWANG (Dept. of Computer Software, Korean Bible University) ;
  • Sac LEE (Dept. of Computer Software, Korean Bible University) ;
  • Hyunwoo LEE (Dept. of Computer Software, Korean Bible University) ;
  • Seyun PARK (Dept. of Computer Software, Korean Bible University) ;
  • Jiyoung LIM (Dept. of Computer Software, Korean Bible University)
  • Received : 2023.02.26
  • Accepted : 2023.03.04
  • Published : 2023.03.30
Download PDF
⟨ Previous Next ⟩

Abstract

With the continuous growth in the amount of data collected and analyzed, deep learning has become increasingly popular for extracting meaningful insights from various fields. However, hardware limitations pose a challenge for achieving meaningful results with limited data. To address this challenge, this paper proposes an algorithm that leverages the characteristics of convolutional neural networks (CNNs) to reduce the size of image datasets by 20% through smoothing and shrinking the size of images using color elements. The proposed algorithm reduces the learning time and, as a result, the computational load on hardware. The experiments conducted in this study show that the proposed method achieves effective learning with similar or slightly higher accuracy than the original dataset while reducing computational and time costs. This color-centric dataset construction method using image smoothing techniques can lead to more efficient learning on CNNs. This method can be applied in various applications, such as image classification and recognition, and can contribute to more efficient and cost-effective deep learning. This paper presents a promising approach to reducing the computational load and time costs associated with deep learning and provides meaningful results with limited data, enabling them to apply deep learning to a broader range of applications.

Keywords

Acknowledgement

This research was supported by UISP (University Innovation Support Project) of Korean Bible University in 2022.

References

  1. Al-Jarrah, O. Y., Yoo, P. D., Muhaidat, S., Karagiannidis, G. K., & Taha, K. (2015). Efficient Machine Learning for Big Data: A Review. Big Data Research, 2(3), 87-93. https://doi.org/10.1016/j.bdr.2015.04.001
  2. Bloice, M.D., Stocker, C., & Holzinger, A. (2017). Augmentor: An Image Augmentation Library for Machine Learning. The Journal of Open Source Software, 2(19), 432.
  3. Brownrigg, D. R. K. (1984). The weighted median filter. Communications of the ACM, 27(8), 807-818. https://doi.org/10.1145/358198.358222
  4. Ciregan, D., Meier, U., & Schmidhuber, J. (2012, June). Multicolumn deep neural networks for image classification. In 2012 IEEE conference on computer vision and pattern recognition. 3642-3649
  5. Deng, G., & Cahill, L. W. (1993, October). An adaptive Gaussian filter for noise reduction and edge detection. In 1993 IEEE conference record nuclear science symposium and medical imaging conference, 1615-1619.
  6. Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009, June). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, 248-255.
  7. Fei-Fei, L., Fergus, R., & Perona, P. (2004, June). Learning generative visual models from few training examples: An incremental bayesian approach tested on 101 object categories. In 2004 conference on computer vision and pattern recognition workshop, 178-178.
  8. Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F. A., & Brendel, W. (2018). ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. arXiv preprint arXiv:1811.12231.
  9. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778.
  10. Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... & Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.
  11. Kim, Kyung-A., & Chung, M.-A. (2022). Analysis of the Status of Artificial Medical Intelligence Technology Based on Big Data. Korean Journal of Artificial Intelligence, 10(2), 13-18.
  12. Krizhevsky, A. (2009). Learning Multiple Layers of Features from Tiny Images. Retrieved May 07, 2023 from https://www.semanticscholar.org/paper/Learning-MultipleLayers-of-Features-from-TinyKrizhevsky/5d90f06bb70a0a3dced62413346235c02b1aa086
  13. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90. https://doi.org/10.1145/3065386
  14. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradientbased learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324. https://doi.org/10.1109/5.726791
  15. LeCun, Y., Huang, F. J., & Bottou, L. (2004, June). Learning methods for generic object recognition with invariance to pose and lighting. In Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. 2, II-104.
  16. Li, Z., Zhou, F., Chen, F., & Li, H. (2017). Meta-SGD: Learning to Learn Quickly for Few-Shot Learning. ArXiv:1707.09835
  17. Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
  18. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition, 1-9.
  19. Thompson, N. C., Greenewald, K., Lee, K., & Manso, G. F. (2020). The computational limits of deep learning. arXiv preprint arXiv:2007.05558.