Name
Abstract | ||
|---|---|---|
Compressive spectral imaging has demonstrated to be a reliable way to capture multispectral data using far few measurements than traditional scanning techniques. These systems capture coded and multiplexed projections of the underlying scene, and then exploits the theory of compressive sensing to solve an optimization algorithm which attains an estimation of the 3D spatio-spectral data cube. Up to date typical compressive imaging systems rely on black-and-white photomasks to perform the coding process. These photomasks block or permit the transmission of a complete pixel of the input data cube regardless of the wavelength, thus performing a coarse coding. In this paper, we report on the optical design, fabrication and implementation of a colored coded aperture compressive spectral imager, which introduces patterned color filter arrays in the coding step. The use of colored coded apertures provides higher flexibility in the coding process, due to their wavelength dependency. This in turn alleviates the inverse problem solution, entailing higher image quality in the reconstructions. Random distributions of the patterned color filter arrays are tested to evaluate the performance, flexibility and quality of the attained data cube estimations. |
Year | Venue | Keywords |
|---|---|---|
2015 | 2015 IEEE GLOBAL CONFERENCE ON SIGNAL AND INFORMATION PROCESSING (GLOBALSIP) | Color filter array, colored coded apertures, compressive sensing, hyperspectral imaging, optical imaging |
Field | DocType | Citations |
Aperture,Computer vision,Spectral imaging,Coded aperture,Computer science,Image quality,Pixel,Color gel,Artificial intelligence,Data cube,Compressed sensing | Conference | 0 |
PageRank | References | Authors |
0.34 | 5 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hoover F. Rueda | 1 | 8 | 1.50 |
| Henry Arguello | 2 | 32 | 5.39 |
| Gonzalo R. Arce | 3 | 1061 | 134.94 |