A novel algorithm on adaptive image compressed sensing with sparsity fitting

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

When the image is compressed adaptively with compressed sensing theory, the determination of sampling rate and sparsity threshold are highly subjective. In order to solve the problem, an accurately adaptive sampling algorithm with sparsity fitting is proposed in this paper. This algorithm determines the minimum sampling rate under certain sparsity to meet the PSNR requirements by iteration, and an optimal objective function of sampling rate choices is obtained by fitting sparsity and sampling rate data with the method of least squares. The adaptive sampling algorithm is simulated based on TVAL3. Experimental results show that the PSNR values of reconstructed images are higher than that with the same fixed sampling rate algorithm, and the PSNR increment of clear texture distinction images can reach at least 3.5dB. Compared to the roughly adaptive compression method, when the average sampling rate is lower, the reconstructed image obtains a higher PSNR value.

Original languageEnglish
Title of host publicationProceedings of the 34th Chinese Control Conference, CCC 2015
EditorsQianchuan Zhao, Shirong Liu
PublisherIEEE Computer Society
Pages4552-4557
Number of pages6
ISBN (Electronic)9789881563897
DOIs
Publication statusPublished - 11 Sept 2015
Event34th Chinese Control Conference, CCC 2015 - Hangzhou, China
Duration: 28 Jul 201530 Jul 2015

Publication series

NameChinese Control Conference, CCC
Volume2015-September
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference34th Chinese Control Conference, CCC 2015
Country/TerritoryChina
CityHangzhou
Period28/07/1530/07/15

Keywords

  • accurately adaptive sampling
  • compressed sensing
  • data fitting
  • sparsity

Fingerprint

Dive into the research topics of 'A novel algorithm on adaptive image compressed sensing with sparsity fitting'. Together they form a unique fingerprint.

Cite this