A comparison of least-squares and Bayesian minimum risk edge parameter estimation

Nanno Mulder, Ali-Akbar Abkar

    Research output: Contribution to journalArticleAcademicpeer-review

    2 Citations (Scopus)

    Abstract

    The problem considered here is to compare two methods for finding a common boundary between two objects with two unknown geometric parameters, such as edge position and edge orientation. We compare two model-based approaches: the least squares and the minimum Bayesian risk method. An expression is derived for the expected error as a function of radiometry. This shows that the least squares method emphasises outliers. The outliers increase the error in parameter estimation. In an experiment with edge parameter estimation, the negative aspects of least squares are detectable.
    Original languageUndefined
    Pages (from-to)1397-1405
    JournalPattern recognition letters
    Volume1999
    Issue number20
    DOIs
    Publication statusPublished - 1999

    Keywords

    • METIS-112349
    • IR-73951
    • Parameter estimation
    • Edge detection
    • Remote sensing
    • Likelihood
    • Least squares

    Cite this

    Mulder, Nanno ; Abkar, Ali-Akbar. / A comparison of least-squares and Bayesian minimum risk edge parameter estimation. In: Pattern recognition letters. 1999 ; Vol. 1999, No. 20. pp. 1397-1405.
    @article{ced138c439e448c1ae9590e61a3f5259,
    title = "A comparison of least-squares and Bayesian minimum risk edge parameter estimation",
    abstract = "The problem considered here is to compare two methods for finding a common boundary between two objects with two unknown geometric parameters, such as edge position and edge orientation. We compare two model-based approaches: the least squares and the minimum Bayesian risk method. An expression is derived for the expected error as a function of radiometry. This shows that the least squares method emphasises outliers. The outliers increase the error in parameter estimation. In an experiment with edge parameter estimation, the negative aspects of least squares are detectable.",
    keywords = "METIS-112349, IR-73951, Parameter estimation, Edge detection, Remote sensing, Likelihood, Least squares",
    author = "Nanno Mulder and Ali-Akbar Abkar",
    year = "1999",
    doi = "10.1016/S0167-8655(99)00111-7",
    language = "Undefined",
    volume = "1999",
    pages = "1397--1405",
    journal = "Pattern recognition letters",
    issn = "0167-8655",
    publisher = "Elsevier",
    number = "20",

    }

    A comparison of least-squares and Bayesian minimum risk edge parameter estimation. / Mulder, Nanno; Abkar, Ali-Akbar.

    In: Pattern recognition letters, Vol. 1999, No. 20, 1999, p. 1397-1405.

    Research output: Contribution to journalArticleAcademicpeer-review

    TY - JOUR

    T1 - A comparison of least-squares and Bayesian minimum risk edge parameter estimation

    AU - Mulder, Nanno

    AU - Abkar, Ali-Akbar

    PY - 1999

    Y1 - 1999

    N2 - The problem considered here is to compare two methods for finding a common boundary between two objects with two unknown geometric parameters, such as edge position and edge orientation. We compare two model-based approaches: the least squares and the minimum Bayesian risk method. An expression is derived for the expected error as a function of radiometry. This shows that the least squares method emphasises outliers. The outliers increase the error in parameter estimation. In an experiment with edge parameter estimation, the negative aspects of least squares are detectable.

    AB - The problem considered here is to compare two methods for finding a common boundary between two objects with two unknown geometric parameters, such as edge position and edge orientation. We compare two model-based approaches: the least squares and the minimum Bayesian risk method. An expression is derived for the expected error as a function of radiometry. This shows that the least squares method emphasises outliers. The outliers increase the error in parameter estimation. In an experiment with edge parameter estimation, the negative aspects of least squares are detectable.

    KW - METIS-112349

    KW - IR-73951

    KW - Parameter estimation

    KW - Edge detection

    KW - Remote sensing

    KW - Likelihood

    KW - Least squares

    U2 - 10.1016/S0167-8655(99)00111-7

    DO - 10.1016/S0167-8655(99)00111-7

    M3 - Article

    VL - 1999

    SP - 1397

    EP - 1405

    JO - Pattern recognition letters

    JF - Pattern recognition letters

    SN - 0167-8655

    IS - 20

    ER -