Single-pixel Computational Imaging: Application to Fluorescence and Diffuse Optical Imaging

Abstract

In this dissertation, we demonstrate the use of single-pixel imaging methods in two important imaging modalities in biomedical applications, time-resolved fluorescence imaging and diffuse optical imaging. In fluorescence imaging, we apply a novel method based on data fusion to acquire giga-voxel 4D hypercubes by measuring only a very small fraction of the dataset. The proposed systems combine conventional cameras with multispectral, time-resolved single-pixel cameras, and data fusion is used to merge the 2D and projections generated by each individual camera into a high-resolution 4D hypercube. In the area of diffuse optical imaging, novel systems based on single-pixel imaging techniques have been developed, allowing the use of simplified detection schemes using integrating spheres as bucket detectors.