Compressive Acquisition of Color and Near-Infrared Images
We propose using a single silicon sensor and a modified Bayer CFA for joint acquisition of color and near-infrared (NIR) images. Silicon sensors, which are placed in most color cameras, are inherently sensitive to NIR. Hence, our proposed design is very similar to consumer color cameras in terms of hardware. The main contribution of this work is an algorithm that estimates full-resolution color and NIR images from subsampled and mixed sensor measurements. Our method results in high-quality RGB and NIR images.
Cite this item
Sadeghipoor Kermani, Z., Lu, Y., & Süsstrunk, S. (2014). Compressive Acquisition of Color and Near-Infrared Images. IEEE International Conference on Computational Photography, Santa Clara, California, USA. https://infoscience.epfl.ch/handle/20.500.14299/103514
Click Copy to copy the citation, or select the text manually.
Sadeghi_ICCP14.pdf
openaccess
19.65 MB
Adobe PDF
d911e1d68bed885fb732e513830d658f