Title :
On least-squares digital contour smoothing
Author :
Glasa, Jan ; Podhajecky, Gerard
Author_Institution :
Inst. of Informatics, Slovak Acad. of Sci., Bratislava, Slovakia
Abstract :
In this paper an efficient procedure for constrained least-squares smoothing for digital contours is presented. It is described by linear operators represented by circulant Toeplitz matrices with integer coefficients which can be efficiently implemented. They allow to calculate several characteristics and invariants related to the corresponding original digitized contour and also to estimate its derivatives and corresponding invariants. The use of the smoothed values and the values of estimated first derivatives obtained by the described procedure for the problem of length estimation of digitized contours is discussed.
Keywords :
Toeplitz matrices; edge detection; least squares approximations; smoothing methods; circulant Toeplitz matrix; integer coefficient; least-squares digital contour smoothing; linear operator; Constraint theory; Digital images; Image analysis; Informatics; Kernel; Polynomials; Shape; Signal processing; Smoothing methods; Testing;
Conference_Titel :
Image and Signal Processing and Analysis, 2003. ISPA 2003. Proceedings of the 3rd International Symposium on
Print_ISBN :
953-184-061-X
DOI :
10.1109/ISPA.2003.1296362