Designing of overcomplete dictionaries based on DCT and DWT

Abdul Qayyum, Aamir Saeed Malik, Mohamad Naufal, Mohamad Saad, Moona Mazher, Faris Abdullah, Tuan Ab Rashid Bin Tuan Abdullah

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Sparse representation is very active area in computer vision and image analysis. It has many applications in de-noising, stereo vision, image painting, image restoration, image de-blurring and many. For sparse modeling, there is need to design an appropriate dictionary. However, there are many dictionaries used for sparse modeling and were reported in literature. In this paper, we implemented the fixed dictionaries and adaptive dictionaries i.e., Method of Optimal Direction (MOD) and KSVD. Both adaptive are used for training the noisy images and computing the error and recovered the number of atoms using adaptive or small patches of images. The result showed that our proposed dictionaries performed much better for atom recovery in noisy patches of the images. The dictionary based on discrete wavelet transform (DWT) basis function with KSVD produced accurate result as compared to all other dictionaries. However, for fast convergence of RMSE value to minimum, DWT with KSVD and MOD dictionaries showed higher convergence rate as compared to discrete cosine transform (DCT) with KSVD and MOD. The computation complexity increased little using the DWT dictionary as compared to DCT dictionary.

Original languageEnglish
Title of host publicationISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences
Subtitle of host publicationBy the Student for the Student
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages134-139
Number of pages6
ISBN (Electronic)9781467378154
DOIs
Publication statusPublished - 17 Mar 2016
EventIEEE Student Symposium in Biomedical Engineering and Sciences, ISSBES 2015 - UiTM, Shah Alam, Malaysia
Duration: 04 Nov 2015 → …

Publication series

NameISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student

Other

OtherIEEE Student Symposium in Biomedical Engineering and Sciences, ISSBES 2015
CountryMalaysia
CityUiTM, Shah Alam
Period04/11/15 → …

Fingerprint

Wavelet Analysis
Discrete cosine transforms
Discrete wavelet transforms
Glossaries
Atoms
Paintings
Stereo vision
Painting
Image reconstruction
Image analysis
Computer vision
Recovery

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications
  • Signal Processing

Cite this

Qayyum, A., Malik, A. S., Naufal, M., Saad, M., Mazher, M., Abdullah, F., & Abdullah, T. A. R. B. T. (2016). Designing of overcomplete dictionaries based on DCT and DWT. In ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student (pp. 134-139). [7435883] (ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSBES.2015.7435883
Qayyum, Abdul ; Malik, Aamir Saeed ; Naufal, Mohamad ; Saad, Mohamad ; Mazher, Moona ; Abdullah, Faris ; Abdullah, Tuan Ab Rashid Bin Tuan. / Designing of overcomplete dictionaries based on DCT and DWT. ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 134-139 (ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student).
@inproceedings{e2c8239763a949228644e63df10e4ac6,
title = "Designing of overcomplete dictionaries based on DCT and DWT",
abstract = "Sparse representation is very active area in computer vision and image analysis. It has many applications in de-noising, stereo vision, image painting, image restoration, image de-blurring and many. For sparse modeling, there is need to design an appropriate dictionary. However, there are many dictionaries used for sparse modeling and were reported in literature. In this paper, we implemented the fixed dictionaries and adaptive dictionaries i.e., Method of Optimal Direction (MOD) and KSVD. Both adaptive are used for training the noisy images and computing the error and recovered the number of atoms using adaptive or small patches of images. The result showed that our proposed dictionaries performed much better for atom recovery in noisy patches of the images. The dictionary based on discrete wavelet transform (DWT) basis function with KSVD produced accurate result as compared to all other dictionaries. However, for fast convergence of RMSE value to minimum, DWT with KSVD and MOD dictionaries showed higher convergence rate as compared to discrete cosine transform (DCT) with KSVD and MOD. The computation complexity increased little using the DWT dictionary as compared to DCT dictionary.",
author = "Abdul Qayyum and Malik, {Aamir Saeed} and Mohamad Naufal and Mohamad Saad and Moona Mazher and Faris Abdullah and Abdullah, {Tuan Ab Rashid Bin Tuan}",
year = "2016",
month = "3",
day = "17",
doi = "10.1109/ISSBES.2015.7435883",
language = "English",
series = "ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "134--139",
booktitle = "ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences",
address = "United States",

}

Qayyum, A, Malik, AS, Naufal, M, Saad, M, Mazher, M, Abdullah, F & Abdullah, TARBT 2016, Designing of overcomplete dictionaries based on DCT and DWT. in ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student., 7435883, ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student, Institute of Electrical and Electronics Engineers Inc., pp. 134-139, IEEE Student Symposium in Biomedical Engineering and Sciences, ISSBES 2015, UiTM, Shah Alam, Malaysia, 04/11/15. https://doi.org/10.1109/ISSBES.2015.7435883

Designing of overcomplete dictionaries based on DCT and DWT. / Qayyum, Abdul; Malik, Aamir Saeed; Naufal, Mohamad; Saad, Mohamad; Mazher, Moona; Abdullah, Faris; Abdullah, Tuan Ab Rashid Bin Tuan.

ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student. Institute of Electrical and Electronics Engineers Inc., 2016. p. 134-139 7435883 (ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Designing of overcomplete dictionaries based on DCT and DWT

AU - Qayyum, Abdul

AU - Malik, Aamir Saeed

AU - Naufal, Mohamad

AU - Saad, Mohamad

AU - Mazher, Moona

AU - Abdullah, Faris

AU - Abdullah, Tuan Ab Rashid Bin Tuan

PY - 2016/3/17

Y1 - 2016/3/17

N2 - Sparse representation is very active area in computer vision and image analysis. It has many applications in de-noising, stereo vision, image painting, image restoration, image de-blurring and many. For sparse modeling, there is need to design an appropriate dictionary. However, there are many dictionaries used for sparse modeling and were reported in literature. In this paper, we implemented the fixed dictionaries and adaptive dictionaries i.e., Method of Optimal Direction (MOD) and KSVD. Both adaptive are used for training the noisy images and computing the error and recovered the number of atoms using adaptive or small patches of images. The result showed that our proposed dictionaries performed much better for atom recovery in noisy patches of the images. The dictionary based on discrete wavelet transform (DWT) basis function with KSVD produced accurate result as compared to all other dictionaries. However, for fast convergence of RMSE value to minimum, DWT with KSVD and MOD dictionaries showed higher convergence rate as compared to discrete cosine transform (DCT) with KSVD and MOD. The computation complexity increased little using the DWT dictionary as compared to DCT dictionary.

AB - Sparse representation is very active area in computer vision and image analysis. It has many applications in de-noising, stereo vision, image painting, image restoration, image de-blurring and many. For sparse modeling, there is need to design an appropriate dictionary. However, there are many dictionaries used for sparse modeling and were reported in literature. In this paper, we implemented the fixed dictionaries and adaptive dictionaries i.e., Method of Optimal Direction (MOD) and KSVD. Both adaptive are used for training the noisy images and computing the error and recovered the number of atoms using adaptive or small patches of images. The result showed that our proposed dictionaries performed much better for atom recovery in noisy patches of the images. The dictionary based on discrete wavelet transform (DWT) basis function with KSVD produced accurate result as compared to all other dictionaries. However, for fast convergence of RMSE value to minimum, DWT with KSVD and MOD dictionaries showed higher convergence rate as compared to discrete cosine transform (DCT) with KSVD and MOD. The computation complexity increased little using the DWT dictionary as compared to DCT dictionary.

UR - http://www.scopus.com/inward/record.url?scp=84965115437&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84965115437&partnerID=8YFLogxK

U2 - 10.1109/ISSBES.2015.7435883

DO - 10.1109/ISSBES.2015.7435883

M3 - Conference contribution

AN - SCOPUS:84965115437

T3 - ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student

SP - 134

EP - 139

BT - ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Qayyum A, Malik AS, Naufal M, Saad M, Mazher M, Abdullah F et al. Designing of overcomplete dictionaries based on DCT and DWT. In ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student. Institute of Electrical and Electronics Engineers Inc. 2016. p. 134-139. 7435883. (ISSBES 2015 - IEEE Student Symposium in Biomedical Engineering and Sciences: By the Student for the Student). https://doi.org/10.1109/ISSBES.2015.7435883