Reconstruction Improvements on Compressed Sensing
Abstract
This paper presents the design of Improvements on Reconstruction of Compressive Sensed images. The proposed techniques will improve the reconstruction time consumption. Those improvements use techniques including matrix simplification, multi-thread and GPU computations. Implementing those improvements achieve gains on time consumption, compared to the baseline. This paper also presents a novel scheme of buffering steamed image (video) to achieve optimum performance.
Downloads
References
Morgan Stanley, “The Mobile Internet Report,” Dec. 2009
E. Candès and Terence Tao, “Near optimal signal recovery from random projections: Universal encoding strategies?” IEEE Trans. on Information Theory, vol. 52, No. 12, pp. 5406 - 5425, December 2006
D. Donoho, “Compressed sensing,” IEEE Trans. on Information Theory, vol. 52, No. 4, pp. 1289 - 1306, April 2006
V. K Goyal, A.K. Fletcher and Sundeep Rangan, “Compressive Samplingand Lossy Compression,” IEEE SIGNAL PROCESSING MAGAZINE, pp. 48-56, MARCH 2008
J. E. Fowler, S. Mun, and E. W. Tramel, “Block-Based Compressed Sensing of Images and Video,” Foundations and Trends in Signal Processing, vol. 4, No. 4, pp. 297-416, March 2012
NVIDIA Corporation, “GPU_Programming_Guide”, Version 2.5, 2006
Jill Reese and Sarah Zaranek, “GPU Programming in MATLAB，The Mathworks technical paper, 2011
Copyright (c) 2017 Journal of Information Sciences and Computing Technologies
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
TRANSFER OF COPYRIGHT
JISCT is pleased to undertake the publication of your contribution to Journal of Information Sciences and Computing Technologies
The copyright to this article is transferred to JISCT(including without limitation, the right to publish the work in whole or in part in any and all forms of media, now or hereafter known) effective if and when the article is accepted for publication thus granting JISCT all rights for the work so that both parties may be protected from the consequences of unauthorized use.
