EFFICIENT SPARSITY-BASED INVERSION FOR PHOTON-SIEVE SPECTRAL IMAGERS WITH TRANSFORM LEARNING

Date

2017-01-01

Author

Kamaci, Ulas
Akyon, Fatih C.
Alkanat, Tunc
Öktem, Sevinç Figen

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

134
views

0
downloads

We develop an efficient and adaptive sparse reconstruction approach for the recovery of spectral images from the measurements of a photon-sieve spectral imager (PSSI). PSSI is a computational imaging technique that enables higher resolution than conventional spectral imagers. Each measurement in PSSI is a superposition of the blurred spectral images; hence, the inverse problem can be viewed as a type of multi-frame deconvolution problem involving multiple objects. The transform learning-based approach reconstructs the spectral images from these superimposed measurements while simultaneously learning a sparsifying transform. This is performed using a block coordinate descent algorithm with efficient update steps. The performance is illustrated for a variety of measurement settings in solar spectral imaging. Compared to approaches with fixed sparsifying transforms, the approach is capable of efficiently reconstructing spectral images with improved reconstruction quality.

Subject Keywords

Hyperspectral imagery, Representations

URI

https://hdl.handle.net/11511/55113

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Accurate solutions of scattering problems involving low-contrast dielectric objects with surface integral equations Ergül, Özgür Salih (2007-11-16) We present the stabilization of the surface integral equationsfor accurate solutions of scattering problems involvinglow-contrast dielectric objects. Unlike volume formulations,conventional surface formulations fail to provide accurateresults for the scatteredfields when the contrast of theobject is small. Therefore, surface formulations are requiredto be stabilized by extracting the nonradiating parts of theequivalent currents. In addition to previous strategies forthe stabilization, we introduce a n...
Optimal Cavities to Enhance Free-space Matching in Solar Cells Karaosmanoglu, B.; Topcuoglu, U.; Guler, S.; Tuygar, E.; Ergül, Özgür Salih (2018-08-04) In this paper, we present design and numerical simulations of nano-cavities for free space matching in solar cells at optical frequencies. The cavity designs are inspired by the wellknown antenna geometries that are commonly used at radio and microwave frequencies. Using such designs that have not been explored before in the context of solar cells, the electromagnetic mismatch between solar-cell surfaces and vacuum is significantly reduced, in comparison to planar interfaces and those with the conventional ...
A Transformation Media Based Approach for Efficient Monte Carlo Analysis of Scattering From Rough Surfaces With Objects Ozgun, Ozlem; Kuzuoğlu, Mustafa (2013-03-01) This paper presents a computational model that utilizes transformation-based metamaterials to enhance the performance of numerical modeling methods for achieving the statistical characterization of two-dimensional electromagnetic scattering from objects on or above one-dimensional rough sea surfaces. Monte Carlo simulation of the rough surface scattering problem by means of differential equation-based finite methods (such as finite element or finite difference methods) usually places a heavy burden on compu...
Deep Learning-Based Joint Reconstruction and System Optimization for Single-Shot Compressive Spectral Imaging Gundogan, Utku; Öktem, Sevinç Figen (2022-01-01) We develop a joint reconstruction and system optimization method for snapshot spectral imaging with diffractive lenses. The method learns the diffractive lens design parameters jointly with a 3D deep prior in an unrolled reconstruction. Results illustrate the significance of jointly optimizing the prior and design parameters.
Automated Detection of Arbitrarily Shaped Buildings in Complex Environments From Monocular VHR Optical Satellite Imagery Ok, Ali Ozgun; Senaras, Caglar; Yuksel, Baris (2013-03-01) This paper introduces a new approach for the automated detection of buildings from monocular very high resolution (VHR) optical satellite images. First, we investigate the shadow evidence to focus on building regions. To do that, we propose a new fuzzy landscape generation approach to model the directional spatial relationship between buildings and their shadows. Once all landscapes are collected, a pruning process is developed to eliminate the landscapes that may occur due to non-building objects. The fina...

Citation Formats

U. Kamaci, F. C. Akyon, T. Alkanat, and S. F. Öktem, “EFFICIENT SPARSITY-BASED INVERSION FOR PHOTON-SIEVE SPECTRAL IMAGERS WITH TRANSFORM LEARNING,” 2017, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/55113.