Poster + Paper
2 October 2023 Convolutional neuronal network to restore images encoded by a wavefront coding imaging system
José M. Reyes-Alfaro, Carina Toxqui-Quitl, María A. Espejel Rivera, Enrique González Amador, Alfonso Padilla-Vivanco
Author Affiliations +
Conference Poster
Abstract
Wavefront coding (WFC) technique has been used to extend the depth of field (DoF) in an imaging system. Its effectiveness lies in using a phase mask (PM), which allows the point spread function (PSF) to remain almost invariant in an axial range of the DoF. Subsequently, a restoration method of the acquired coded images is required. An optical-computational technique that uses Trefoil profile PM for encoding and a convolutional neural network (CNN) to restore is presented. Comparative results are done between the restored image using the Wiener filter and the one obtained with a CNN. An image quality evaluation is done in terms of spatial resolution and contrast.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
José M. Reyes-Alfaro, Carina Toxqui-Quitl, María A. Espejel Rivera, Enrique González Amador, and Alfonso Padilla-Vivanco "Convolutional neuronal network to restore images encoded by a wavefront coding imaging system", Proc. SPIE 12666, Current Developments in Lens Design and Optical Engineering XXIV, 126660J (2 October 2023); https://doi.org/10.1117/12.2677797
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Imaging systems

Wavefronts

Deblurring

Convolutional neural networks

Image deconvolution

Image quality

Signal filtering

Back to Top