한국정보통신학회논문지 (Journal of the Korea Institute of Information and Communication Engineering)

  • 제26권11호
  • /
  • Pages.1630-1636
  • /
  • 2022
  • /
  • 2234-4772(pISSN)
  • /
  • 2288-4165(eISSN)
한국정보통신학회 (The Korea Institute of Information and Commucation Engineering)
권호 표지
DOI QR코드

DOI QR Code

홉필드 네트워크와 퍼지 Max-Min 신경망을 이용한 손상된 교통 표지판 인식

Damaged Traffic Sign Recognition using Hopfield Networks and Fuzzy Max-Min Neural Network

  • Kim, Kwang Baek (Department of Artificial Intelligence, Silla University)
  • 투고 : 2022.11.08
  • 심사 : 2022.11.17
  • 발행 : 2022.11.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

현재 교통 표지판 인식 기법들은 다양한 날씨, 빛의 변화 등과 같은 외부환경 뿐만 아니라 교통 표지판이 일부 훼손된 경우에는 인식 성능이 저하되는 경우가 발생한다. 따라서 본 논문에서는 이러한 문제점을 개선하기 위하여 홉필드 네트워크와 퍼지 Max-Min 신경망을 이용하여 손상된 교통 표지판의 인식 성능을 개선하는 방법을 제안한다. 제안된 방법은 손상된 교통 표지판에서 특징들을 분석한 후, 그 특징들을 학습 패턴으로 구성하여 퍼지 Max-Min 신경망에 적용하여 1차적으로 교통 표지판의 특징을 분류한다. 1차적 분류된 특징이 있는 학습 영상들을 홉필드 네트워크에 적용하여 손상된 특징을 복원한다. 홉필드 네트워크를 적용하여 복원된 교통 표지판의 특징들을 다시 퍼지 Max-Min 신경망에 적용하여 최종적으로 손상된 교통 표지판을 분류하고 인식한다. 제안된 방법의 성능을 평가하기 위하여 손상된 정도가 다른 다양한 교통 표지판 8개를 적용하여 실험한 결과, 제안된 방법이 퍼지 Max-Min 신경망에 비해 평균적으로 38.76%의 분류 성능이 개선되었다.

The results of current method of traffic sign detection gets hindered by environmental conditions and the traffic sign's condition as well. Therefore, in this paper, we propose a method of improving detection performance of damaged traffic signs by utilizing Hopfield Network and Fuzzy Max-Min Neural Network. In this proposed method, the characteristics of damaged traffic signs are analyzed and those characteristics are configured as the training pattern to be used by Fuzzy Max-Min Neural Network to initially classify the characteristics of the traffic signs. The images with initial characteristics that has been classified are restored by using Hopfield Network. The images restored with Hopfield Network are classified by the Fuzzy Max-Min Neural Network onces again to finally classify and detect the damaged traffic signs. 8 traffic signs with varying degrees of damage are used to evaluate the performance of the proposed method which resulted with an average of 38.76% improvement on classification performance than the Fuzzy Max-Min Neural Network.

키워드

과제정보

Following are results of a study on the "Leaders in INdustry-university Cooperation 3.0" Project, supported by the Ministry of Education and National Research Foundation of Korea.

참고문헌

  1. B. S. Chu, "Impact of Visual Performance on Recognition of Road and Traffic Sign," Journal of Korean Society of Transportation, vol. 29 no. 1, pp. 48-56, Feb. 2011.
  2. G. W. Bang, D. W. Kang, and W. H. Cho, "Traffic Sign Recognition Using Color Information and Error Back Propagation Algorithm," The KIPS Transactions : Part D, vol. 14-D, no. 7, pp. 809-818, Dec. 2007.
  3. J. T. Oh, H. W. Kwak, and W. H. Kim, "Recognition of Traffic Signs using Wavelet Transform and Shape Information," Journal of the Institute of Electronics and Information Engineers, vol. 41, no. 5, pp. 125-134, Sep. 2004.
  4. S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 6, pp. 1137-1149, Jun. 2017.
  5. J. Dai, Y. Li, K. He, and J. Sun, "R-FCN: Object Detection via Region-based Fully Convolutional Networks," in Proceedings of the 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, vol. 29, pp. 379-387, 2016.
  6. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. -Y. Fu, and A. C. Berg, "SSD: Single Shot MultiBox Detector," in Proceedings of European Conference on Computer Vision(ECCV), Amsterdam, The Netherlands, pp. 21-37, 2016.
  7. P. Jiang, D. Ergu, F. Liu, Y. Cai, and B. Ma, "A Review of Yolo Algorithm Developments," Pocedia Computer Science, vol. 199, pp. 1066-1073, 2022. https://doi.org/10.1016/j.procs.2022.01.135
  8. N. Upasani and H. Om, "Optimized fuzzy min-max neural network: an efficient approach for supervised outlier detection," Neural Network World, vol. 28, no. 4, pp. 285-303, Jan. 2018. https://doi.org/10.14311/nnw.2018.28.017
  9. Y. L. Karpov, L. E. Karpov, Y. G. Smetanin,"Some Aspects of Associative Memory Construction Based on a Hopfield Network," Programming and Computer Software, vol. 46, pp. 305-311, Oct. 2020. https://doi.org/10.1134/S0361768820050023