Reconstruction of Sparse Multi-dimensional Imaging Data for Time-efficient Imaging

点击翻译按钮获取中文摘要

Reconstruction of Sparse Multi-dimensional Imaging Data for Time-efficient Imaging In a continuation of my current NSERC-funded research program, I plan to further investigate new techniques for the reconstruction of images from sparse data. My investigations will focus on a general class of recently formulated methods, known as compressed sensing approaches. These approaches share some properties with image compression commonly employed in digital cameras to make pictures smaller. In compressed sensing, we assume that our final images can be formed form a subset of the needed data. Compressed sensing describes the process at which the limited acquired data is "uncompressed" to form high quality images. The overarching objective of my program over the next five years is to explore the reconstruction of multi-dimensional data, by extending our work from sparse sampling only along two or three spatial dimensions, to multi-dimensional sampling including the three spatial dimensions plus time. In accomplishing this goal, we will also enhance methods for improving 1) image reconstruction approaches and 2) the objective assessment of image reconstructed from compressed data. NSERC support will provide foundational support for this program of fundamental image science research and will principally support trainees engaged by this program.