Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing

Xingsong Hou; Chunli Wang; Xueming Qian

doi:10.1145/2808492.2808563

Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing

Xingsong Hou
, Chunli Wang
, Xueming Qian

Faculty of Electronic and Information Engineering

Xi'an Jiaotong University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In compressive imaging (CI), accurate coefficients recovery is possible when the transform coefficients for an image are sufficiently sparse. However, conventional transforms, such as discrete cosine transform (DCT) and discrete wavelet transform (DWT), can not acquire a sufficiently sparse representation. A large amount of small coefficients indeed bring interference to the the recovery of large ones. This paper aims to improve the recovery performance by accurately reconstructing the large coefficients as many as possible. Thus, a compressive imaging scheme based on coefficients cutting in directional lifting wavelet transform (DLWT) domain and low-complexity iterative Bayesian algorithm is proposed. The proposed scheme is an improved version of our previous work. In this paper, relations between the best-fitted cutoff ratio and sampling rate are further analyzed. Due to the efficient Bayesian recovery algorithm, the proposed method offers better recovery performance with much lower complexity than our previous work. Experimental results show that our method outperforms many state-of-the-art compressive imaging recovery methods.

Original language	English
Title of host publication	ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service
Editors	Ramesh Jain, Shuqiang Jiang, John Smith, Jitao Sang, Guohui Li, Tianzhu Zhang, Shuhui Wang
Publisher	Association for Computing Machinery
Pages	63-67
Number of pages	5
ISBN (Electronic)	9781450335287
DOIs	https://doi.org/10.1145/2808492.2808563
State	Published - 19 Aug 2015
Event	7th International Conference on Internet Multimedia Computing and Service, ICIMCS 2015 - Zhangjiajie, Hunan, China Duration: 19 Aug 2015 → 21 Aug 2015

Publication series

Name	ACM International Conference Proceeding Series
Volume	2015-August

Conference

Conference	7th International Conference on Internet Multimedia Computing and Service, ICIMCS 2015
Country/Territory	China
City	Zhangjiajie, Hunan
Period	19/08/15 → 21/08/15

Keywords

Approximate message passing
Coefficients cutting
Compressive imaging
DLWT
Tail folding

Access to Document

10.1145/2808492.2808563

Cite this

Hou, X., Wang, C., & Qian, X. (2015). Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing. In R. Jain, S. Jiang, J. Smith, J. Sang, G. Li, T. Zhang, & S. Wang (Eds.), ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service (pp. 63-67). (ACM International Conference Proceeding Series; Vol. 2015-August). Association for Computing Machinery. https://doi.org/10.1145/2808492.2808563

Hou, Xingsong ; Wang, Chunli ; Qian, Xueming. / Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing. ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service. editor / Ramesh Jain ; Shuqiang Jiang ; John Smith ; Jitao Sang ; Guohui Li ; Tianzhu Zhang ; Shuhui Wang. Association for Computing Machinery, 2015. pp. 63-67 (ACM International Conference Proceeding Series).

@inproceedings{57973898a8ea46319b8b6cd561e9ba8b,

title = "Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing",

abstract = "In compressive imaging (CI), accurate coefficients recovery is possible when the transform coefficients for an image are sufficiently sparse. However, conventional transforms, such as discrete cosine transform (DCT) and discrete wavelet transform (DWT), can not acquire a sufficiently sparse representation. A large amount of small coefficients indeed bring interference to the the recovery of large ones. This paper aims to improve the recovery performance by accurately reconstructing the large coefficients as many as possible. Thus, a compressive imaging scheme based on coefficients cutting in directional lifting wavelet transform (DLWT) domain and low-complexity iterative Bayesian algorithm is proposed. The proposed scheme is an improved version of our previous work. In this paper, relations between the best-fitted cutoff ratio and sampling rate are further analyzed. Due to the efficient Bayesian recovery algorithm, the proposed method offers better recovery performance with much lower complexity than our previous work. Experimental results show that our method outperforms many state-of-the-art compressive imaging recovery methods.",

keywords = "Approximate message passing, Coefficients cutting, Compressive imaging, DLWT, Tail folding",

author = "Xingsong Hou and Chunli Wang and Xueming Qian",

note = "Publisher Copyright: {\textcopyright} 2015 ACM.; 7th International Conference on Internet Multimedia Computing and Service, ICIMCS 2015 ; Conference date: 19-08-2015 Through 21-08-2015",

year = "2015",

month = aug,

day = "19",

doi = "10.1145/2808492.2808563",

language = "英语",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "63--67",

editor = "Ramesh Jain and Shuqiang Jiang and John Smith and Jitao Sang and Guohui Li and Tianzhu Zhang and Shuhui Wang",

booktitle = "ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service",

}

Hou, X, Wang, C & Qian, X 2015, Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing. in R Jain, S Jiang, J Smith, J Sang, G Li, T Zhang & S Wang (eds), ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service. ACM International Conference Proceeding Series, vol. 2015-August, Association for Computing Machinery, pp. 63-67, 7th International Conference on Internet Multimedia Computing and Service, ICIMCS 2015, Zhangjiajie, Hunan, China, 19/08/15. https://doi.org/10.1145/2808492.2808563

Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing. / Hou, Xingsong; Wang, Chunli; Qian, Xueming.
ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service. ed. / Ramesh Jain; Shuqiang Jiang; John Smith; Jitao Sang; Guohui Li; Tianzhu Zhang; Shuhui Wang. Association for Computing Machinery, 2015. p. 63-67 (ACM International Conference Proceeding Series; Vol. 2015-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing

AU - Hou, Xingsong

AU - Wang, Chunli

AU - Qian, Xueming

PY - 2015/8/19

Y1 - 2015/8/19

N2 - In compressive imaging (CI), accurate coefficients recovery is possible when the transform coefficients for an image are sufficiently sparse. However, conventional transforms, such as discrete cosine transform (DCT) and discrete wavelet transform (DWT), can not acquire a sufficiently sparse representation. A large amount of small coefficients indeed bring interference to the the recovery of large ones. This paper aims to improve the recovery performance by accurately reconstructing the large coefficients as many as possible. Thus, a compressive imaging scheme based on coefficients cutting in directional lifting wavelet transform (DLWT) domain and low-complexity iterative Bayesian algorithm is proposed. The proposed scheme is an improved version of our previous work. In this paper, relations between the best-fitted cutoff ratio and sampling rate are further analyzed. Due to the efficient Bayesian recovery algorithm, the proposed method offers better recovery performance with much lower complexity than our previous work. Experimental results show that our method outperforms many state-of-the-art compressive imaging recovery methods.

AB - In compressive imaging (CI), accurate coefficients recovery is possible when the transform coefficients for an image are sufficiently sparse. However, conventional transforms, such as discrete cosine transform (DCT) and discrete wavelet transform (DWT), can not acquire a sufficiently sparse representation. A large amount of small coefficients indeed bring interference to the the recovery of large ones. This paper aims to improve the recovery performance by accurately reconstructing the large coefficients as many as possible. Thus, a compressive imaging scheme based on coefficients cutting in directional lifting wavelet transform (DLWT) domain and low-complexity iterative Bayesian algorithm is proposed. The proposed scheme is an improved version of our previous work. In this paper, relations between the best-fitted cutoff ratio and sampling rate are further analyzed. Due to the efficient Bayesian recovery algorithm, the proposed method offers better recovery performance with much lower complexity than our previous work. Experimental results show that our method outperforms many state-of-the-art compressive imaging recovery methods.

KW - Approximate message passing

KW - Coefficients cutting

KW - Compressive imaging

KW - DLWT

KW - Tail folding

UR - https://www.scopus.com/pages/publications/84947572150

U2 - 10.1145/2808492.2808563

DO - 10.1145/2808492.2808563

M3 - 会议稿件

AN - SCOPUS:84947572150

T3 - ACM International Conference Proceeding Series

SP - 63

EP - 67

BT - ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service

A2 - Jain, Ramesh

A2 - Jiang, Shuqiang

A2 - Smith, John

A2 - Sang, Jitao

A2 - Li, Guohui

A2 - Zhang, Tianzhu

A2 - Wang, Shuhui

PB - Association for Computing Machinery

T2 - 7th International Conference on Internet Multimedia Computing and Service, ICIMCS 2015

Y2 - 19 August 2015 through 21 August 2015

ER -

Hou X, Wang C, Qian X. Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing. In Jain R, Jiang S, Smith J, Sang J, Li G, Zhang T, Wang S, editors, ICIMCS 2015 - Proceedings of the 7th International Conference on Internet Multimedia Computing and Service. Association for Computing Machinery. 2015. p. 63-67. (ACM International Conference Proceeding Series). doi: 10.1145/2808492.2808563

Compressive Imaging based on Coefficients Cutting in DLWT Domain and Approximate Message Passing

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this