Journal of the Korea Institute of Military Science and Technology (한국군사과학기술학회지)

The Korea Institute of Military Science and Technology (한국군사과학기술학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Maritime Object Image Classification Using a Pruning-Based Lightweight Deep-Learning Model

가지치기 기반 경량 딥러닝 모델을 활용한 해상객체 이미지 분류에 관한 연구

  • Younghoon Han (Korea Combat Training Center, ROK Army Training & Doctrine Command) ;
  • Chunju Lee (Department of National Defense Science, Korea National Defense University) ;
  • Jaegoo Kang (Korea Combat Training Center, ROK Army Training & Doctrine Command)
  • Received : 2024.01.29
  • Accepted : 2024.04.01
  • Published : 2024.06.05
Download PDF
⟨ Previous Next ⟩

Abstract

Deep learning models require high computing power due to a substantial amount of computation. It is difficult to use them in devices with limited computing environments, such as coastal surveillance equipments. In this study, a lightweight model is constructed by analyzing the weight changes of the convolutional layers during the training process based on MobileNet and then pruning the layers that affects the model less. The performance comparison results show that the lightweight model maintains performance while reducing computational load, parameters, model size, and data processing speed. As a result of this study, an effective pruning method for constructing lightweight deep learning models and the possibility of using equipment resources efficiently through lightweight models in limited computing environments such as coastal surveillance equipments are presented.

Keywords

References

  1. Young A Lee, "Disclosure of North Korean small wooden boat's 'entry into Samcheok Port', government joint investigation results," citizen daily, http://www.siminilbo.co.kr/news/articleView.html?idxno=643486, 2019.
  2. Moon Kyung Kim, "'Taean smuggling' surveillance equipment was filmed 13 times... the military who could not see and misjudged," YTN, https://www.ytn.co.kr/_ln/0101_202006051848395193,
  3. Howard, Andrew G. et al., "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
  4. LeCun, et. al., "Deep learning," nature, 521.7553, pp 436-444, 2015. https://doi.org/10.1038/nature14539
  5. Lee K. H., Y. H. Kim, "Trends in Lightweighting Deep Learning Models," Journal of the Society for Information Science, pp. 19-20,
  6. Li, Hao et al., "Pruning filters for efficient convnets," 5th International Conference on Learning Representations(ICLR), Toulon, France, 2017.
  7. Dong-hwi Kim, Su-jin Lee and Sang-hyo Park, "A Study on Lightweight and Optimizing with Generative Adversarial Network Based Video Super-resolution Model," Korean Society of Broadcasting and Media Engineering Summer Conference, pp. 461-463, 2022.
  8. Pavlo Molchanov et al., "Importance Estimation for Neural Network Pruning," Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), pp. 11264-11272, 2019.
  9. Sara Elkerdawy et al., "To Filter Prune, or to Layer Prune, That Is The Question," Proceedings of the Asian Conference on Computer Vision(ACCV), 2020.