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 Nov 19, 2024
Page view(s)
9
checked on Nov 19, 2024
Google ScholarTM
Check
Altmetric
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.