Using NOAA-AVHRR estimates of land surface temperature for regional agrometeoro[lo]gical modelling

A.J.W. de Wit, H.L. Boogaard, C.A. van Diepen

    Research output: Contribution to journalArticleAcademicpeer-review

    9 Citations (Scopus)

    Abstract

    This paper presents a case study on the use of features derived from remote sensing data for mapping the highly fragmented semideciduous Atlantic forest in Brazil. Innovative aspects of this research include the evaluation of different feature sets in order to improve land cover mapping. The feature sets were defined based on expert knowledge and on data mining techniques to be input to traditional and machine learning algorithms for pattern recognition, viz. maximum likelihood, univariate decision trees, multivariate decision trees, and neural networks. The results showed that the maximum likelihood classification using temporal texture descriptors as extracted with wavelet transforms was most accurate to classify the semideciduous Atlantic forest. In this study, a special accuracy measure was used: the so-called class mapping accuracy. Maximum likelihood performed relatively well, with forest mapping accuracies ranging from 34.5 to 51.3%. In contrast, accuracies for neural networks ranged from 19.0 to 45.2%. Classification confusion occurred mainly with coffee and eucalyptus plantations. Univariate trees provided the most robust results for different feature sets, with accuracies ranging from 39.6 to 46.7%. Temporal information of vegetation indices was more important than image texture, terrain topography and raw spectral information for discriminating semideciduous Atlantic forest.
    Original languageEnglish
    Pages (from-to)187-204
    JournalInternational Journal of applied Earth Observation and Geoinformation
    Volume5
    Issue number3
    DOIs
    Publication statusPublished - 2004

    Fingerprint

    Advanced very high resolution radiometers (AVHRR)
    AVHRR
    land surface
    surface temperature
    Maximum likelihood
    Decision trees
    modeling
    texture
    Neural networks
    Image texture
    Coffee
    Temperature
    data mining
    pattern recognition
    coffee
    vegetation index
    Wavelet transforms
    Topography
    Learning algorithms
    wavelet

    Cite this

    @article{68324a009f7448d1b2e7aaaa77e7c6e6,
    title = "Using NOAA-AVHRR estimates of land surface temperature for regional agrometeoro[lo]gical modelling",
    abstract = "This paper presents a case study on the use of features derived from remote sensing data for mapping the highly fragmented semideciduous Atlantic forest in Brazil. Innovative aspects of this research include the evaluation of different feature sets in order to improve land cover mapping. The feature sets were defined based on expert knowledge and on data mining techniques to be input to traditional and machine learning algorithms for pattern recognition, viz. maximum likelihood, univariate decision trees, multivariate decision trees, and neural networks. The results showed that the maximum likelihood classification using temporal texture descriptors as extracted with wavelet transforms was most accurate to classify the semideciduous Atlantic forest. In this study, a special accuracy measure was used: the so-called class mapping accuracy. Maximum likelihood performed relatively well, with forest mapping accuracies ranging from 34.5 to 51.3{\%}. In contrast, accuracies for neural networks ranged from 19.0 to 45.2{\%}. Classification confusion occurred mainly with coffee and eucalyptus plantations. Univariate trees provided the most robust results for different feature sets, with accuracies ranging from 39.6 to 46.7{\%}. Temporal information of vegetation indices was more important than image texture, terrain topography and raw spectral information for discriminating semideciduous Atlantic forest.",
    author = "{de Wit}, A.J.W. and H.L. Boogaard and {van Diepen}, C.A.",
    year = "2004",
    doi = "10.1016/j.jag.2004.03.003",
    language = "English",
    volume = "5",
    pages = "187--204",
    journal = "International Journal of applied Earth Observation and Geoinformation",
    issn = "0303-2434",
    publisher = "Elsevier",
    number = "3",

    }

    Using NOAA-AVHRR estimates of land surface temperature for regional agrometeoro[lo]gical modelling. / de Wit, A.J.W.; Boogaard, H.L.; van Diepen, C.A.

    In: International Journal of applied Earth Observation and Geoinformation, Vol. 5, No. 3, 2004, p. 187-204.

    Research output: Contribution to journalArticleAcademicpeer-review

    TY - JOUR

    T1 - Using NOAA-AVHRR estimates of land surface temperature for regional agrometeoro[lo]gical modelling

    AU - de Wit, A.J.W.

    AU - Boogaard, H.L.

    AU - van Diepen, C.A.

    PY - 2004

    Y1 - 2004

    N2 - This paper presents a case study on the use of features derived from remote sensing data for mapping the highly fragmented semideciduous Atlantic forest in Brazil. Innovative aspects of this research include the evaluation of different feature sets in order to improve land cover mapping. The feature sets were defined based on expert knowledge and on data mining techniques to be input to traditional and machine learning algorithms for pattern recognition, viz. maximum likelihood, univariate decision trees, multivariate decision trees, and neural networks. The results showed that the maximum likelihood classification using temporal texture descriptors as extracted with wavelet transforms was most accurate to classify the semideciduous Atlantic forest. In this study, a special accuracy measure was used: the so-called class mapping accuracy. Maximum likelihood performed relatively well, with forest mapping accuracies ranging from 34.5 to 51.3%. In contrast, accuracies for neural networks ranged from 19.0 to 45.2%. Classification confusion occurred mainly with coffee and eucalyptus plantations. Univariate trees provided the most robust results for different feature sets, with accuracies ranging from 39.6 to 46.7%. Temporal information of vegetation indices was more important than image texture, terrain topography and raw spectral information for discriminating semideciduous Atlantic forest.

    AB - This paper presents a case study on the use of features derived from remote sensing data for mapping the highly fragmented semideciduous Atlantic forest in Brazil. Innovative aspects of this research include the evaluation of different feature sets in order to improve land cover mapping. The feature sets were defined based on expert knowledge and on data mining techniques to be input to traditional and machine learning algorithms for pattern recognition, viz. maximum likelihood, univariate decision trees, multivariate decision trees, and neural networks. The results showed that the maximum likelihood classification using temporal texture descriptors as extracted with wavelet transforms was most accurate to classify the semideciduous Atlantic forest. In this study, a special accuracy measure was used: the so-called class mapping accuracy. Maximum likelihood performed relatively well, with forest mapping accuracies ranging from 34.5 to 51.3%. In contrast, accuracies for neural networks ranged from 19.0 to 45.2%. Classification confusion occurred mainly with coffee and eucalyptus plantations. Univariate trees provided the most robust results for different feature sets, with accuracies ranging from 39.6 to 46.7%. Temporal information of vegetation indices was more important than image texture, terrain topography and raw spectral information for discriminating semideciduous Atlantic forest.

    U2 - 10.1016/j.jag.2004.03.003

    DO - 10.1016/j.jag.2004.03.003

    M3 - Article

    VL - 5

    SP - 187

    EP - 204

    JO - International Journal of applied Earth Observation and Geoinformation

    JF - International Journal of applied Earth Observation and Geoinformation

    SN - 0303-2434

    IS - 3

    ER -