A semi nonnegative matrix factorization technique for pattern generalization in single-pixel imaging
Résumé
A single-pixel camera is a computational imaging device that only requires a single point detector to capture the image of a scene. It measures the inner products of the scene and some spatial light modulator patterns, which are to be processed to recover the scene. No matter the strategy used for image recovery, the spatial light modulator patterns have to be positive. In addition, the dark current measured in the absence of modulation must be rejected. So far, both experimental issues have been addressed empirically. In this paper, we solve them from a general perspective. Indeed, we propose to seek positive patterns that are linear combinations of the desired patterns (with negative values) and the linear combinations are chosen to reject the dark current. We refer to the problem of finding the positive patterns and the linear combinations as pattern generalization. To the best of our knowledge, this is the first time that this problem is introduced. In addition, we show that pattern generalization can be solved using a semi nonnegative matrix factorization algorithm. Results obtained from simulations demonstrate that our approach performs close or better than conventional methods while using fewer measurements.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...