2018/07/29 by Laura Rebollo‐Neira, Rebollo-Neira, Laura, Daniel Whitehouse +1
Computer Science · Engineering · #Advanced Vision and Imaging #FOS: Electrical engineering #Image Enhancement Techniques #Image and Signal Denoising Methods #Image and Video Processing (eess.IV) #Sparse and Compressive Sensing Techniques #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1807.11116
openalex publication_date 2018/07/29 · openalex created_date 2022/08/04 · openalex updated_date 2026/07/28
Sparse representation of 3D images is considered within the context of data\nreduction. The goal is to produce high quality approximations of 3D images\nusing fewer elementary components than the number of intensity points in the 3D\narray. This is achieved by means of a highly redundant dictionary and a\ndedicated pursuit strategy especially designed for low memory requirements. The\nbenefit of the proposed framework is illustrated in the first instance by\ndemonstrating the gain in dimensionality reduction obtained when approximating\ntrue color images as very thin 3D arrays, instead of performing an independent\nchannel by channel approximation. The full power of the approach is further\nexemplified by producing high quality approximations of hyper-spectral images\nwith a reduction of up to 371 times the number of data points in the\nrepresentation.\n