Pattern Spectrum Component Function and Warning Traffic Sign Recognition

패턴 스펙트럼 성분 함수와 주의 교통 표지 인식

  • 김회진 (동명전문대학 전자과) ;
  • 장강의 (중국 Hangzhou 대학교 전자공학과) ;
  • 최태영 (아주대학교 전자공학과)
  • Published : 1997.03.01

Abstract

In this paper, a pattern spectrum component function is introduced for an oriented shape analysis and its properties are discussed. It can represent directional information of shape more precisely than the conventional oriented pattern spectrum. An adaptive distance function between two pattern spectrum component functions is presented to recognize different shapes in noise. As a practical application, the pattern spectrum component function is applied to warning traffic sign recognitions utilizing the adaptive distance functions. Favorable results are obtained compared to the oriented pattern spectrum.

기존의 방향성 패턴 스펙트럼에 비하여 보다 정확하게 방향성 정보를 나타낼 수 있는 패턴 스펙트럼 성분 함수를 도입하고 그 성질을 분석하였다. 그리고 패턴 스펙트럼 성분 함수에 의하여 잡음성 형상을 식별하고자 적응 거리 함수를 제안하였다. 제안한 패턴 스펙트럼 성분 함수를 주의 교통 표지 식별에 실제 응용하여 적응 거리 함수에 의하여 평가한 본 결과 방향성 패턴 스펙트럼에 비하여 보다 만족할 만한 결과를 얻을 수 있었다.

Keywords

References

  1. Computer and Robot Vision R.M. Haralick;L.G. Shapiro
  2. Random Sets and Integral Geometry G. Matheron
  3. Image Analysis and Mathematical Morphology J. Serra
  4. Proc. of IEEE v.78 no.4 Morphological system for multidimensional signal processing P. Maragos;R.W. Schafer
  5. IEEE Trans. Acoust. Speech Signal Process v.34 no.5 Morphological skeleton representation and coding of binary image P.A. Maragos;R.W. Schafer
  6. IEEE Trans. Pattern Analysis Mach. Intell. v.11 no.7 Pattern spectrum and multiscale shape representation P. Maragos
  7. IEEE Trans.Signal Process v.39 no.6 Morphological representation of discrete and binary images J. Goutsias;D. Schonfeld
  8. Pattern Recognition v.25 no.6 Morphological shape representation I. Pitas;A.N. Venetsanopoulos
  9. Pattern Recognition v.26 no.7 A morphological signature transform for shape description S. Loncric;A.P. Dhawan
  10. Digital Image Processing Methods E.R. Dougherty
  11. Pattern Recognition v.25 no.9 M orphological shape description using geometric spectrum on multidimensional binary images F.Y. Shih;C.C. Pu
  12. Mathematical Morphology And Its Applicatoins To Image Processing Statistical pattern spectrum for binary pattern recognition C.S.Regazzoni;G.L.Foresti;A.N.Venetsanopoulos;J. Serra;P. Soille (eds.)
  13. Pattern Recognition v.28 no.9 Shape description and recognition using the high order morphological pattern spectrum X. Zhou;B. Yuan