Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator

Xu Huang, A Madoc, Andrew Cheetham

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    2 Citations (Scopus)

    Abstract

    Images are in many cases degraded even before they are encoded. The major noise sources, in terms of distributions, are Gaussian noise, Poisson noise and impulse noise. Noise acquired by images during transmission would be Gaussian in distribution, while images such as emission and transmission tomography images, X-ray films, and photographs taken by satellites are usually contaminated by quantum noise, which is Poisson distributed. Poisson shot noise is a natural generalization of a compound Poisson process when the summands are stochastic processes starting at the points of the underlying Poisson process. Unlike additive Gaussian noise, Poisson noise is signal-dependent and consequently separating signal from noise is more difficult. In our previous papers we discussed a wavelet-based maximum likelihood for Bayesian estimator that recovers the signal component of wavelet coefficients in original images using an alpha-stable signal prior distribution. In this paper, it is demonstrated that the method can be extended to multi-noise sources comprising Gaussian, Poisson, and impulse noise. Results of varying the parameters of the Bayesian estimators of the model are presented after an investigation of α-stable simulations for a maximum likelihood estimator. As an example, a colour image is processed and presented to illustrate the effectiveness of this method.
    Original languageEnglish
    Title of host publicationIEEE International Symposium on Circuits and Systems, Conference Proceedings
    EditorsNobuo Fijii
    Place of PublicationJapan
    PublisherIEEE
    Pages2699-2702
    Number of pages4
    ISBN (Print)0-7803-8834-8
    DOIs
    Publication statusPublished - 2005
    EventIEEE International Symposium on Circuits and Systems (ISCAS) - Kobe, Japan
    Duration: 23 May 200526 May 2005

    Conference

    ConferenceIEEE International Symposium on Circuits and Systems (ISCAS)
    CountryJapan
    CityKobe
    Period23/05/0526/05/05

    Fingerprint

    estimators
    random noise
    poisson process
    impulses
    stochastic processes
    shot noise
    photographs
    tomography
    color
    coefficients

    Cite this

    Huang, X., Madoc, A., & Cheetham, A. (2005). Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator. In N. Fijii (Ed.), IEEE International Symposium on Circuits and Systems, Conference Proceedings (pp. 2699-2702). Japan: IEEE. https://doi.org/10.1109/ISCAS.2005.1465183
    Huang, Xu ; Madoc, A ; Cheetham, Andrew. / Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator. IEEE International Symposium on Circuits and Systems, Conference Proceedings. editor / Nobuo Fijii. Japan : IEEE, 2005. pp. 2699-2702
    @inproceedings{bfff2ef1bc854997ab15787f981fd5fc,
    title = "Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator",
    abstract = "Images are in many cases degraded even before they are encoded. The major noise sources, in terms of distributions, are Gaussian noise, Poisson noise and impulse noise. Noise acquired by images during transmission would be Gaussian in distribution, while images such as emission and transmission tomography images, X-ray films, and photographs taken by satellites are usually contaminated by quantum noise, which is Poisson distributed. Poisson shot noise is a natural generalization of a compound Poisson process when the summands are stochastic processes starting at the points of the underlying Poisson process. Unlike additive Gaussian noise, Poisson noise is signal-dependent and consequently separating signal from noise is more difficult. In our previous papers we discussed a wavelet-based maximum likelihood for Bayesian estimator that recovers the signal component of wavelet coefficients in original images using an alpha-stable signal prior distribution. In this paper, it is demonstrated that the method can be extended to multi-noise sources comprising Gaussian, Poisson, and impulse noise. Results of varying the parameters of the Bayesian estimators of the model are presented after an investigation of α-stable simulations for a maximum likelihood estimator. As an example, a colour image is processed and presented to illustrate the effectiveness of this method.",
    author = "Xu Huang and A Madoc and Andrew Cheetham",
    year = "2005",
    doi = "10.1109/ISCAS.2005.1465183",
    language = "English",
    isbn = "0-7803-8834-8",
    pages = "2699--2702",
    editor = "Nobuo Fijii",
    booktitle = "IEEE International Symposium on Circuits and Systems, Conference Proceedings",
    publisher = "IEEE",

    }

    Huang, X, Madoc, A & Cheetham, A 2005, Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator. in N Fijii (ed.), IEEE International Symposium on Circuits and Systems, Conference Proceedings. IEEE, Japan, pp. 2699-2702, IEEE International Symposium on Circuits and Systems (ISCAS), Kobe, Japan, 23/05/05. https://doi.org/10.1109/ISCAS.2005.1465183

    Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator. / Huang, Xu; Madoc, A; Cheetham, Andrew.

    IEEE International Symposium on Circuits and Systems, Conference Proceedings. ed. / Nobuo Fijii. Japan : IEEE, 2005. p. 2699-2702.

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    TY - GEN

    T1 - Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator

    AU - Huang, Xu

    AU - Madoc, A

    AU - Cheetham, Andrew

    PY - 2005

    Y1 - 2005

    N2 - Images are in many cases degraded even before they are encoded. The major noise sources, in terms of distributions, are Gaussian noise, Poisson noise and impulse noise. Noise acquired by images during transmission would be Gaussian in distribution, while images such as emission and transmission tomography images, X-ray films, and photographs taken by satellites are usually contaminated by quantum noise, which is Poisson distributed. Poisson shot noise is a natural generalization of a compound Poisson process when the summands are stochastic processes starting at the points of the underlying Poisson process. Unlike additive Gaussian noise, Poisson noise is signal-dependent and consequently separating signal from noise is more difficult. In our previous papers we discussed a wavelet-based maximum likelihood for Bayesian estimator that recovers the signal component of wavelet coefficients in original images using an alpha-stable signal prior distribution. In this paper, it is demonstrated that the method can be extended to multi-noise sources comprising Gaussian, Poisson, and impulse noise. Results of varying the parameters of the Bayesian estimators of the model are presented after an investigation of α-stable simulations for a maximum likelihood estimator. As an example, a colour image is processed and presented to illustrate the effectiveness of this method.

    AB - Images are in many cases degraded even before they are encoded. The major noise sources, in terms of distributions, are Gaussian noise, Poisson noise and impulse noise. Noise acquired by images during transmission would be Gaussian in distribution, while images such as emission and transmission tomography images, X-ray films, and photographs taken by satellites are usually contaminated by quantum noise, which is Poisson distributed. Poisson shot noise is a natural generalization of a compound Poisson process when the summands are stochastic processes starting at the points of the underlying Poisson process. Unlike additive Gaussian noise, Poisson noise is signal-dependent and consequently separating signal from noise is more difficult. In our previous papers we discussed a wavelet-based maximum likelihood for Bayesian estimator that recovers the signal component of wavelet coefficients in original images using an alpha-stable signal prior distribution. In this paper, it is demonstrated that the method can be extended to multi-noise sources comprising Gaussian, Poisson, and impulse noise. Results of varying the parameters of the Bayesian estimators of the model are presented after an investigation of α-stable simulations for a maximum likelihood estimator. As an example, a colour image is processed and presented to illustrate the effectiveness of this method.

    U2 - 10.1109/ISCAS.2005.1465183

    DO - 10.1109/ISCAS.2005.1465183

    M3 - Conference contribution

    SN - 0-7803-8834-8

    SP - 2699

    EP - 2702

    BT - IEEE International Symposium on Circuits and Systems, Conference Proceedings

    A2 - Fijii, Nobuo

    PB - IEEE

    CY - Japan

    ER -

    Huang X, Madoc A, Cheetham A. Image Multi-Noise Removal by Wavelet-Based Bayesian Estimator. In Fijii N, editor, IEEE International Symposium on Circuits and Systems, Conference Proceedings. Japan: IEEE. 2005. p. 2699-2702 https://doi.org/10.1109/ISCAS.2005.1465183