A hybrid MLP-PNN architecture for fast image superresolution
Entity
UAM. Departamento de Ingeniería InformáticaPublisher
Springer Berlin HeidelbergDate
2003Citation
10.1007/3-540-44989-2_50
Artificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003: Joint International Conference ICANN/ICONIP 2003 Istanbul, Turkey, June 26–29, 2003 Proceedings. Lecture Notes in Computer Science, Volumen 2714. Springer 2003. 417-424
ISSN
0302-9743 (print); 1611-3349 (online)ISBN
978-3-540-40408-8 (print); 978-3-540-44989-8 (online)DOI
10.1007/3-540-44989-2_50Funded by
This work has been partially supported by TIC2001-0572-C02-02 grantEditor's Version
http://dx.doi.org/10.1007/3-540-44989-2_50Subjects
Computation by Abstract Devices; Computer Communication Networks; Special Purpose and Application-Based Systems; Database Management; Computer Imaging, Vision, Pattern Recognition and Graphics; InformáticaNote
Proceedings of Joint International Conference ICANN/ICONIP 2003 Istanbul, Turkey, June 26–29, 2003The final publication is available at Springer via http://dx.doi.org/10.1007/3-540-44989-2_50
Rights
© Springer-Verlag Berlin Heidelberg 2003Abstract
Image superresolution methods process an input image sequence of a scene to obtain a still image with increased resolution. Classical approaches to this problem involve complex iterative minimization procedures, typically with high computational costs. In this paper is proposed a novel algorithm for superresolution that enables a substantial decrease in computer load. First, a probabilistic neural network architecture is used to perform a scattered-point interpolation of the image sequence data. The network kernel function is optimally determined for this problem by a multi-layer perceptron trained on synthetic data. Network parameters dependence on sequence noise level is quantitatively analyzed. This super-sampled image is spatially filtered to correct finite pixel size effects, to yield the final high-resolution estimate. Results on a real outdoor sequence are presented, showing the quality of the proposed method.
Files in this item
Google Scholar:Miravet, Carlos
-
Rodríguez Ortiz, Francisco Borja
This item appears in the following Collection(s)
Related items
Showing items related by title, author, creator and subject.