The Optimal Resolution for Circle Analysis with the Minimum Error

최소 오차 원 해석을 위한 최적 해상도에 관한 연구

  • Published : 2000.05.01

Abstract

In this paper, an algorithm for determining the optimal resolution has been described for measuring the actual length of circular objects. As the resolution gets higher, the measurement error in general becomes smaller because of the reduced distance per pixel. However, the higher resolution makes circular objects enlarged, which may produce an ill-conditioned system. That is, a small error in the boundary positions may result in a large error in the analysis of the circular objects. Taking this fact into account, a new measure is proposed to determine the optimal resolution. The actual errors have been calculated with various resolutions and the resolution with the minimum error has been decided as the optimal resolution. The analysis using various circles with different sizes indicates that the minimum measurement error is obtained when the whole circle appears in the screen as large as possible, regardless of the size of circle. The experimental results using real images have verified the validity of our analysis.

Keywords

References

  1. 이일환, 박희재, '비젼을 이용한 기어 형상 측정 시스템 개발,' 한국정밀공학회 추계학술대회, pp. 485-489, 1996
  2. 이성천, 김민환, '산업용 비전시스템에서의 빠른 원 인식 방법,' 한국정보과학회 논문지, 제 24 권, 제 12 호, pp. 1419-1427, 1997
  3. A.J.Tabatabai and O.R. Mitchell, 'Edge location to subpixel values in digital imagery,' IEEE Trans. Pattern Anal. Machine Intell., Vol. PAMI-6, No. 2, pp. 188-201, Mar. 1984
  4. A. Huertas and G. Medioni, 'Detection of intensity changes with subpixel accuracy using Laplacian-Gaussian masks,' IEEE Trans. Pattern Anal. Machine Intell., Vol. PAMI-8, No. 5, pp. 188-201, Sept. 1986
  5. D. H. Ballard and C. M. Brown, Computer Vision, Prentice-Hall, pp. 143-145, 1982
  6. M.Sonka, V.Hlavac and R.Boyle, Image Processing, Analysis and Machine Vision, Chapman & Hall Computing, pp. 113-122, 1993