Abstract
Recently, a new compressive sensing (CS) based spectral imaging system, namely, coded aperture snapshot spectral imager (CASSI) has been reported. CASSI captures multi-spectral image cubes with sub-Nyquist sampling rate and fast acquisition speed. However, the single shot imaging process with highly compressed measurements usually results in the loss of useful information and degraded imaging quality. Various image reconstruction or estimation algorithms have been proposed to solve this problem. However, the design of efficient coded aperture pattern remains as a challenging problem. In this paper, we propose the design of spectral aperture code patterns for CASSI admitting multi-shot measurements, which leads to improve imaging quality, as well as spectral band selectivity. Furthermore, results show that the reconstruction speed can be improved by dividing the large data cube into smaller groups and perform reconstruction algorithm on each group simultaneously. The proposed coding scheme is especially useful when a pre-determined subset of spectral bands is needed for reconstruction.
© 2010 Optical Society of America
PDF ArticleMore Like This
Ana Ramirez, Gonzalo R. Arce, and Brian M. Sadler
CM4B.6 Computational Optical Sensing and Imaging (COSI) 2012
Henry Arguello and Gonzalo Arce
ITuA4 Imaging Systems and Applications (IS) 2011
Nelson Diaz, Jorge Bacca, and Henry Arguello
JTu5A.4 3D Image Acquisition and Display: Technology, Perception and Applications (3D) 2017