Authors: | Žunić, Joviša Acketa, Dragan |
Title: | Least squares fitting of digital polynomial segments | Journal: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Volume: | 1176 | First page: | 17 | Last page: | 23 | Conference: | 6th International Workshop on Discrete Geometry for Computer Imagery, DGCI 1996; Lyon; France; 13 November 1996 through 15 November 1996 | Issue Date: | 1-Jan-1996 | ISBN: | 978-3-540-62005-1 | ISSN: | 0302-9743 | DOI: | 10.1007/3-540-62005-2_2 | Abstract: | It is proved that digital polynomial segments and their least squares polynomial fits are in one-to-one correspondence. This enables an efficient representation of digital polynomial segments by n+3 parameters, under the condition that an upper bound, say n, for the degrees of the digitized polynomials is assumed. One of such representations is (x 1, m, an, an−1,…, a 0), where x 1 and m are the x-coordinate of the left endpoint and the number of digital points, respectively, while a n, a n−1,..., a 0 are the coefficients of the least squares polynomial fit Y=a nXn+an− 1Xn−1+ ...+a0, for a given digital polynomial segment. |
Keywords: | Coding | Computer vision | Digital polynomial segment | Image processing | Least squares fitting | Publisher: | Springer Link |
Show full item record
SCOPUSTM
Citations
1
checked on Dec 26, 2024
Page view(s)
10
checked on Dec 26, 2024
Google ScholarTM
Check
Altmetric
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.