Journal of Information Processing Systems

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Transfer Learning-Based Feature Fusion Model for Classification of Maneuver Weapon Systems

  • Jinyong Hwang (Dept. of Artificial Intelligence and Big Data Engineering, Daegu Catholic University) ;
  • You-Rak Choi (Smart Structural Safety & Prognosis Research Division, Korea Atomic Energy Research Institute) ;
  • Tae-Jin Park (Smart Structural Safety & Prognosis Research Division, Korea Atomic Energy Research Institute) ;
  • Ji-Hoon Bae (Dept. of Artificial Intelligence and Big Data Engineering, Daegu Catholic University)
  • Received : 2022.11.22
  • Accepted : 2023.03.21
  • Published : 2023.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2022-00165225).

References

  1. S. K. Singh, J. Cha, T. W. Kim, and J. H. Park, "Machine learning based distributed big data analysis framework for next generation web in IoT," Computer Science and Information Systems, vol. 18, no. 2, pp. 597-618, 2021. https://doi.org/10.2298/CSIS200330012S
  2. B. Li, X. Zheng, W. Guo, Y. Wang, R. Mao, X. Cheng, et al., "Radiation pneumonitis prediction using multiomics fusion based on a novel machine learning pipeline," Human-centric Computing and Information Sciences, vol. 12, article no. 49, 2022. https://doi.org/10.22967/HCIS.2022.12.049
  3. E. Kristiani, Y. T. Tsan, P. Y. Liu, N. Y. Yen, and C. T. Yang, "Binary and multi-class assessment of face mask classification on edge AI using CNN and transfer learning," Human-centric Computing and Information Sciences, vol. 12, article no. 53, 2022. https://doi.org/10.22967/HCIS.2022.12.053
  4. H. Lee, J. Kim, J. Yu, Y. Jeong, and S. Kim, "Multi-class classification using transfer learning based convolutional neural network," Journal of Korean Institute of Intelligent Systems, vol. 28, no. 6, pp. 531-537, 2018. https://doi.org/10.5391/jkiis.2018.28.6.531
  5. Y. Kim, K. Yu, and J. Han, "Research for radar signal classification model using deep learning technique," Journal of the Korea Institute of Military Science and Technology, vol. 22, no. 2, pp. 170-178, 2019. https://doi.org/10.9766/KIMST.2019.22.2.170
  6. S. J. Hong, Y. G. Yi, J. W. Choi, J. Jo, and B. S. Seo, "Classification of Radar signals using machine learning techniques," Journal of IKEEE, vol. 22, no. 1, pp. 162-167, 2018. https://doi.org/10.7471/ikeee.2018.22.1.162
  7. H. Surrisyad and Wahyono, "A fast military object recognition using extreme learning approach on CNN," International Journal of Advanced Computer Science and Applications, vol. 11, no. 12, pp. 210-220, 2020. https://doi.org/10.14569/ijacsa.2020.0111227
  8. Stanford Vision Lab, "ImageNet overview," 2020 [Online]. Available: https://image-net.org/about.php.
  9. K. H. Kim and J. H. Bae, "Important parameter optimized flow-based transfer learning technique supporting heterogeneous teacher network based on deep learning," Journal of the Korean Institute of Information Technology, vol. 18, no. 3, pp. 21-29, 2019. http://doi.org/10.14801/jkiit.2020.18.3.21
  10. M. Jeong and J. Ma, "A study on the improvement of submarine detection based on mast images using an ensemble model of convolutional neural networks," Journal of the Korea Institute of Military Science and Technology, vol. 23, no. 2, pp. 115-124, 2020. https://doi.org/10.9766/KIMST.2020.23.2.115
  11. J. Y. Hwang, B. A. Choi, J. H. Lee, and J. H. Bae, "A study on transfer learning-based ensemble model for classification of maneuver weapon system," Journal of the Korean Institute of Information Technology, vol. 19, no. 12, pp. 1-10, 2021. http://doi.org/10.14801/jkiit.2021.19.12.1
  12. R. O. Lane, A. J. Wragge, W. J. Holmes, S. J. Bertram, and T. Lamont-Smith, "Object detection in EO/IR and SAR images using low-SWAP hardware," in Proceedings of 2021 Sensor Signal Processing for Defence Conference (SSPD), Edinburgh, UK, 2021, pp. 1-5. https://doi.org/10.1109/SSPD51364.2021.9541497
  13. Z. Yang, W. Yu, P. Liang, H. Guo, L. Xia, F. Zhang, Y. Ma, and J. Ma, "Deep transfer learning for military object recognition under small training set condition," Neural Computing and Applications, vol. 31, pp. 6469-6478, 2019. https://doi.org/10.1007/s00521-018-3468-3
  14. D. Legendre and J. Vankka, "Military vehicle recognition with different image machine learning techniques," in Information and Software Technologies. Cham, Switzerland: Springer, 2020, pp. 220-242. https://doi.org/10.1007/978-3-030-59506-7_19
  15. T. Hiippala, "Recognizing military vehicles in social media images using deep learning," in Proceedings of 2017 IEEE International Conference on Intelligence and Security Informatics (ISI), Beijing, China, 2017, pp. 60-65. https://doi.org/10.1109/ISI.2017.8004875
  16. Y. Ouyang, X. Wang, R. Hu, H. Xu, and F. Shao, "Military vehicle object detection based on hierarchical feature representation and refined localization," IEEE Access, vol. 10, pp. 99897-99908, 2022. https://doi.org/10.1109/ACCESS.2022.3207153
  17. S. Chung and M. G. Chung, "Pedestrian classification using CNN's deep features and transfer learning," Journal of Internet Computing and Services, vol. 20, no. 4, pp. 91-102, 2019. https://doi.org/10.7472/jksii.2019.20.4.91
  18. K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," in Proceedings of the 3rd International Conference on Learning Representations (ICLR), San Diego, CA, 2014, pp. 1-14.
  19. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. C. Chen, "MobileNetv2: inverted residuals and linear bottlenecks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, 2018, pp. 4510-4520. https://doi.org/10.1109/CVPR.2018.00474
  20. K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, 2016, pp. 770-778. https://doi.org/10.1109/CVPR.2016.90
  21. G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely connected convolutional networks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, 2017, pp. 2261-2269. https://doi.org/10.1109/CVPR.2017.243
  22. D. A. Clevert, T. Unterthiner, and S. Hochreiter, "Fast and accurate deep network learning by exponential linear units (ELUs)," in Proceedings of the 4th International Conference on Learning Representations, San Juan, Puerto Rico, 2016.
  23. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," The Journal of Machine Learning Research, vol. 15, no. 1, pp. 1929-1958, 2014.
  24. D. P. Kingma and J. Ba, "Adam: a method for stochastic optimization," in Proceedings of the 3rd International Conference on Learning Representations (ICLR), San Diego, CA, 2015.
  25. S. W. Park, J. C. Kim, and D. Y. Kim, "A study on classification performance analysis of convolutional neural network using ensemble learning algorithm," Journal of Korea Multimedia Society, vol. 22, no. 6, pp. 665-675, 2019. https://doi.org/10.9717/kmms.2019.22.6.665