首页>>文献首页>>文献正文

Image Sparse Representation Based on Ensemble Learning in Compressed Sensing

DonghaiBao 添加于 2018/4/6 13:53:08  171次阅读 | 0次推荐 | 0个评论

Abstract: A novel framework in image sparse representation based on ensemble learning is proposed in this paper. Due to the random extraction of training patches and the variation of single optimal processing result, the proposed scheme develops classical dictionary learning algorithms in compressed sensing with ensemble learning theory to improve the performance of sparse representation. The analysis of computational complexity and proof of performance enhancement under the new framework are given in this paper. Simulations that use real images and classical dictionary learning algorithms illustrate the advantages of the proposed strategy in term of peak signal-to-noise ratio. Simulation results show that only a few number of learners can improve the performance dramatically and the robustness of new framework is shown with different testing images.

作 者:Donghai Bao ; Qingpei Wang ; Jiajun Ding ; Sheng Li ; Xiongxiong He
期刊名称:
期卷页: 第卷 第期 页
学科领域:信息科学 » 电子学与信息系统 » 信号理论与信号处理
添加人是否为作者:
原文链接:http://ieeexplore.ieee.org/document/8242440/
DOI: 10.1109/ICSPCC.2017.8242440
ISBN: 978-1-5386-3142-3
关键词:
备 注:
文献笔记

   
导出选项:

评论 (0 个评论)

Page 1 of 1
<<<[1]>>>

举报 | Archiver | 科学网( 京ICP备07017567 )

GMT+8, 2011-2-15 11:31

Powered by ScienceNet.cn

Copyright © 2007- 科学时报社