The Journal of Korean Institute of Information Technology (한국정보기술학회논문지)

Korean Institute of Information Technology (한국정보기술학회)
권호 표지
DOI QR코드

DOI QR Code

Design of Convolutional Neural Network Structure for the Identification of Warhead and Debris in the Separation Phase

기두부와 단 분리 시 조각의 식별을 위한 합성곱 신경망 구조 설계

  • Received : 2018.04.20
  • Accepted : 2018.06.21
  • Published : 2018.06.30
⟨ Previous Next ⟩

Abstract

In this paper, we designed CNN(Convolutional Neural Network) structure to identify warhead and debris in boosting part separation phase. Through simulation, we determined variables of each layer constituting the CNN and designed CNN structure. Simulation were performed to classify four types of warhead with coning motion and six types of debris with tumbling motion through the CNN designed by the proposed method. Then we compared the performance of CNN with the well-known VGGNet. Simulation results show that the CNN structure optimized by the convolution filter, pooling method, and pooling size determined using the proposed method has equal classification performance or better classification performance than VGGNet for all SNR. In addition, the training time was improved approximately 22 times.

본 논문에서는 기두부와 단 분리 시 조각의 식별을 위한 합성곱 신경망(CNN, Convolutional Neural Network) 구조를 설계하였다. 시뮬레이션을 통하여, 합성곱 신경망을 구성하는 각 계층의 변수들을 결정하고 구조를 설계하였다. 이와 같은 방법으로 설계 된 합성곱 신경망을 통해 원추운동(coning)을 갖는 4종류의 기두부와 텀블링운동(tumbling)을 갖는 6종류의 단 분리 시 조각을 분류하는 시뮬레이션을 수행하고, 그 성능을 기존에 잘 알려진 VGGNet(Visual Geometry Group Net)과 비교하였다. 시뮬레이션 결과, 제안된 기법에 의해 선정 된 컨벌루션 필터, 풀링 기법, 풀링 크기를 통해 최적화된 합성곱 신경망 구조는 모든 SNR에 대해서 VGGNet과 동일하거나 우수한 구분 성능을 보였으며, 학습시간은 약 22배 향상되었음을 확인하였다.

Keywords

Acknowledgement

Supported by : 국방과학연구소, 한국연구재단

References

  1. A. R. Persico, C. V. Ilioudis, C. Clemente, and J. Soraghan, "Novel Approach for Ballistic Targets Classification from HRRP frame", Proc. IEEE Conf. Sensor Signal Processing for Defence (SSPD), pp, 1-5, Dec. 2017.
  2. P. Lei, K. Li, and Y. Liu, "Feature extraction and target recognition of missile targets based on micro-motion", 2012 IEEE 11th Int. Conf. Signal Process., pp. 1914-1919, Oct. 2012.
  3. Y. Q. Zhuang, "Accurate statistical modeling method for dynamic RCS", Progress in Electromagnetics Research Symposium Proceedings, pp. 25-28, Aug. 2014.
  4. P. Peng, C. Tong, J. Sun, D. Li, and W. Tong, "Dynamic RCS feature of ballistic missile for detection and classification in the boost phase", in Proc. Cross Strait Quad-Regional Radio Sci. and Wirel. Technol. Conf. (CSQRWC), pp. 427-430, Jul. 2013.
  5. S. H. Park, J. H. Jeong, C. K. Ryoo, and K. Y. Choi, "Detection and Classification of a Ballistic Missile in Ascent Phase", 11th International Conference on Control, Automation and Systems, pp. 1235-1238, Oct. 2011.
  6. O. Russakovsky et al., "ImageNet large scale visual recognition challenge", International Journal of Computer Vision, Vol. 115, No. 3, pp. 211-252, Dec. 2015. https://doi.org/10.1007/s11263-015-0816-y
  7. G. Hinton et al., "Deep Neural Networks for Acoustic Modeling in Speech Recognition", IEEE Signal Processing Magazine, Vol. 29, No. 6, pp. 82-97, Nov. 2012. https://doi.org/10.1109/MSP.2012.2205597
  8. A. Graves, A. Mohamed, and G. Hinton. "Speech recognition with deep recurrent neural networks," IEEE Conf. Acoustics Speech and Signal Processing (ICASSP), pp. 6645-6659, May 2013.
  9. A. Krizhevsky, I. Sutskever, and G. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks", Proc. Neural Information and Processing Systems, pp. 1106-1114, Dec. 2012.
  10. C. Szegedy et al., "Going Deeper with Convolutions", Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 1-9, Jun. 2015.
  11. Simonyan, K. and Zisserman, A. "Very Deep Convolutional Networks For Large-Scale Image Recognition", In Proc. International Conference on Learning Representations, pp. 1-14, Apr. 2015.
  12. K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning For Image Recognition", Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), pp. 770-778, Jun. 2016.
  13. Y. LeCun, Y. Bengio, and G. Hinton, Deep learning. Nature, Vol. 521, No. 7553, pp. 436-444, May 2015. https://doi.org/10.1038/nature14539
  14. Phill Kim, "Deep Learning for Beginners with MATLAB Examples", GLS, pp. 143-174, Dec. 2016.
  15. I. O. Choi, K. B. Kang, S. H Kim, J. H. Jung, K. T. Kim, and S. H. Park, "Performance Analysis of the Discrimination between Space-targets Using the Maneuvering", Journal of KIIT, Vol. 15, No. 11, pp. 55-63, Nov. 2017.
  16. C. de Boor, "A Practical Guide to Splines", New York:Springer-Verlag, 1978.
  17. D. C. Kwon and B. Y. Kang, "CPU and GPU Performance Analysis for Convolution Neural Network", Journal of KIIT, Vol. 15, No. 8, pp. 11-18, Aug. 2017.

Cited by

  1. Classification of Warhead and Debris using CFAR and Convolutional Neural Networks vol.17, pp.6, 2018, https://doi.org/10.14801/jkiit.2019.17.6.85