Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Artículo científico |
| Lingua: | en |
| Pubblicazione: |
Instituto Politécnico Nacional
2014
|
| Soggetti: | |
| Accesso online: | https://www.redalyc.org/articulo.oa?id=402640463010 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
Sommario:
- Computing Polynomial Segmentation through Radial Surface Representation Leticia Flores-Pulido Gustavo Rodríguez-Gómez Oleg Starostenko Vicente Alarcón Alberto Portilla Computación segmentation data modeling polynomial function Subspace arrangement radial basis surface representation The Visual Information Retrieval (VIR) arearequires robust implementations achieved trough mathematicalrepresentations for images or data sets. The implementation ofa mathematical modeling goes from the corpus image selection,an appropriate descriptor method, a segmentation approach andthe similarity metric implementation whose are treated as VIRelements. The goal of this research is to found an appropriatemodeling to explain how its items can be represented to achievea better performance in VIR implementations. A direct methodis tested with a subspace arrangement approach. The GeneralPrincipal Component Analysis (GPCA) is modified inside itssegmentation process. Initially, a corpus data sample is tested,the descriptor of RGB colors is implemented to obtain a threedimensional description of image data. Then a selection ofradial basis function is achieved to improve the similarity metricimplemented. It is concluded that a better performance can beachieved applying powerful extraction methods in visual imageretrieval (VIR) designs based in a mathematical formulation. Theresults lead to design VIR systems with high level of performancebased in radial basis functions and polynomial segmentations forhandling data sets. 2014 artículo científico 1870-9044 https://www.redalyc.org/articulo.oa?id=402640463010 en http://www.redalyc.org/revista.oa?id=4026 Polibits application/pdf Instituto Politécnico Nacional Polibits (México) Vol.49