Training of a Neural Network for Image Superresolution Based on a Nonlinear Interpolative Vector Quantizer
Applied Optics, Vol. 39, Issue 20, pp. 3473-3485 (2000)
http://dx.doi.org/10.1364/AO.39.003473
Acrobat PDF (2442 KB)
Abstract
Superresolution is the process of extending the spectrum of a diffraction-limited image beyond the optical passband. We consider the neural-network approach to accomplish superresolution and present results on simulated gray-scale images degraded by diffraction blur and additive noise. Images are assumed to be sampled at the Nyquist rate, which requires spatial interpolation for avoiding aliasing, in addition to frequency-domain extrapolation. A novel, to our knowledge, use of vector quantization for the generation of training data sets is also presented. This is accomplished by training of a nonlinear vector quantizer, whose codebooks are subsequently used in the supervised training of the neural network with backpropagation.
© 2000 Optical Society of America
[Optical Society of America ]
OCIS Codes
(100.3010) Image processing : Image reconstruction techniques
(100.3020) Image processing : Image reconstruction-restoration
(100.6640) Image processing : Superresolution
(200.4260) Optics in computing : Neural networks
Citation
Carlos A. Dávila and B. R. Hunt, "Training of a Neural Network for Image Superresolution Based on a Nonlinear Interpolative Vector Quantizer," Appl. Opt. 39, 3473-3485 (2000)
http://www.opticsinfobase.org/ao/abstract.cfm?URI=ao-39-20-3473
You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Log in to access OSA Member Subscription
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Log in to access OSA Member Subscription
You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Log in to access OSA Member Subscription
Related Journal Articles 
- Superresolution of binary images with a nonlinear interpolative neural network (AO)
- Iterative image reconstruction using prior knowledge (JOSAA)
- Image reconstruction from limited Fourier data (JOSAA)
- Dealiased spectral images from aliased Fizeau Fourier transform spectroscopy measurements (JOSAA)
- Multiframe superresolution of binary images (AO)
Related Conference Papers
« Previous Article | Next Article »
OSA is a member of 