DOI QR코드

DOI QR Code

The FE-SM/SONN for Recognition of the Car Skid Mark

자동차 스키드마크 인식을 위한 FE-SM/SONN

  • Koo, Gun-Seo (Dept. Internet Information & Processing, Soong Eui Women's College)
  • 구건서 (숭의여자대학 인터넷정보과)
  • Received : 2011.12.19
  • Accepted : 2012.01.16
  • Published : 2012.01.31

Abstract

In this paper, We proposes FE-SM/SONN for recognizing blurred and smeared skid mark image caused by sudden braking of a vehicle. In a blurred and smeared skid marks, tread pattern image is ambiguous. To improve recognition of such image, FE-SM/SONN reads skid marks utilizing Fuzzy Logic and distinguishing tread pattern SONN(Self Organization Neural Networks) recognizer. In order to substantiate this finding, 48 tire models and 144 skid marks were compared and overall recognition ratio was 89%. This study showed 13.51% improved recognition compared to existing back propagation recognizer, and 8.78% improvement than FE-MCBP. The expected effect of this research is achieving recognition of ambiguous images by extracting distinguishing features, and the finding concludes that even when tread pattern image is in grey scale, Fuzzy Logic enables the tread pattern recognizable.

본 논문은 차량이 급제동으로 인해 도로 위에 생성된 스키드마크와 같이 형태가 모호하게 나타난 영상을 인식하기 위해 FE-SM/SONN을 제안하였다. FE-SM/SONN은 타이어 트레드 패턴이 뭉개져서 나타나는 스키드마크 경우, 그 패턴이 모호한 영상으로 취득된다. 이를 인식하기 위해 퍼지 이론과 트레드 패턴의 특징을 이용한 자기 조직 신경망 인식기를 통해 스키드마크를 인식하는 방법이다. 이러한 실험을 위해 48개 타이어모델과 144개 스키드마크가 사용되었고, 전체 인식율은 89%이며, 비교 분석을 위해서는 기존 역전파 인식기에 비해 인식률 면에서 13.51%가 향상되었고, FE-MCBP에 비해 8.78% 향상을 보였다. 이 논문의 기대효과로는 모호한 영상의 특징을 추출하여 인식이 가능하였고, 트레드 패턴 영상이 그레이 영상으로 나타날 경우도 퍼지 이론에 의해 인식이 가능한 것으로 연구결과 나타났다.

Keywords

References

  1. W.S. Hong, "Investigation of Traffic Accident using Skid Mark", Journal of The institute Of Webcasting, Internet And Telecommunication", Vol. 10, No. 1,, pp.113-120, 2010.
  2. S.W. Hwang, Skid Mark, "Investigation of Traffic Accident, pp.1-10, 2010
  3. http://www.msgroup.org/forums/MTT/topic.asp?TOPIC_ID=956
  4. http://carsago.co.kr/jaryo/ja_gyo/g_su/g_su_1.htm
  5. Hankook Tire Korea, Hankook Tire Manual
  6. G. S. Koo, "Feature Extraction based FE-SONN for Signature Verification", Journal of The Korea Society of Computer and Information, Vol. 10, No. 6, pp. 94-101, 2005.
  7. C. Y. Ko, "Recognition of Handwritten Digits Based on Neural Network and Fuzzy Interface", Journal of The Korea Society of Computer and Information, Vol. 16, No. 10, pp. 64-69, 2011.
  8. J.H. Jang, "The Hangeul images Recognition and Restoration based on Neural Network and Memory Theory", Journal of The Korea Society of Computer and Information, Vol. 10, No. 4, pp. 17-27, 2005.
  9. G. K. Sierra, J. O. Bulla, M. A. Melgarejo, "An Embedded Type-2 Fuzzy Processor For The Inverted Pendulum Control Problem", IEEE Latin America Transaction, Vol. 9, No. 3, pp.240-246, June 2011. https://doi.org/10.1109/TLA.2011.5893768
  10. Madhi Khezi, MECHRAN JAHED, "A Neuro- Fuzzy Interface System for SEMG- based Identification of Hand Motion Commands", IEEE Transaction on Industrial Electronics, vol. 58, No. 3, pp. 240-246, June 2011.
  11. G. S. Koo, "The FE-MCBP for Recognition of the Titled New-Type Vehicle License Plate" Journal of The Korea Society of Computer and Information, Vol. 12 No. 5, pp. 73-81, 2007.
  12. G. S. Koo, "Edge Feature Extract CBIRS for Car Retrieval : CBIRS/EFI", Journal of The Korea Society of Computer and Information, Vol. 15 No. 11, pp. 75-82, 2010.