Facebook Instagram Twitter RSS Feed PodBean Back to top on side

On the use of high breakdown point estimators in the image

In: Tatra Mountains Mathematical Publications, vol. 17, no. 3
Stefan Müller
Detaily:
Rok, strany: 1999, 283 - 293
O článku:
For eliminating noise from an image, Meer et al. (1990, 1991) [P. Meer, D. Mintz, A. Rosenfeld: Least median of squares based robust analysis of image structure, in: Proceedings of Image Understanding Workshop, DARPA, 1990, pp. 231–254], [P. Meer, D. Mintz, A. Rosenfeld, D. Y. Kim: Robust regression methods in computer vision: A review, Internat. J. Comput. Vision 6} (1991), 59–70] have proposed to use high breakdown point estimators and in particular to use the least median of squares (LMS) estimators for linear regression. But. the highest breakdown point are attained by others estimator as the Cauchy estimator and some least trimmed squares the (LTS) Therefore, in this paper the behaviour of the Cauchy estimator and the estimator in image analysis is studied and compared with that of the LMS estimator and the least squares (LS) estimator. For that, test images with $0\%, 14\%, 30\%, 44\%$ and $49\%$ noise are used. It turns out that the LTS estimator can eliminate a high amount of noise and preserves discontinuities while the Cauchy estimator behaves similarly to the nonrobust LS estimator.
Ako citovať:
ISO 690:
Müller, S. 1999. On the use of high breakdown point estimators in the image. In Tatra Mountains Mathematical Publications, vol. 17, no.3, pp. 283-293. 1210-3195.

APA:
Müller, S. (1999). On the use of high breakdown point estimators in the image. Tatra Mountains Mathematical Publications, 17(3), 283-293. 1210-3195.