A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region

Omoleomo Omo-Irabor, Kehinde O. K. Oduyemi

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

    Abstract

    The landscape of the Niger Delta region of Nigeria is undergoing rapid changes as a result of natural and anthropogenic activities. This necessitates the development of a rapid, cost effective and efficient land cover (LC) classification technique to monitor the biophysical dynamics in the region. Due to the intricate land cover patterns prominent in the study area and the irregularly indistinguishable relationship between land cover and spectral signals, this paper introduces a combined use of unsupervised and supervised image classification for detecting LC classes. The technique utilizes the spectral recognition of the unsupervised classification in the performance mode and the selection of sampling sites from a principal component analyzed image of the supervised classification in the training mode. The unsupervised and supervised classification algorithms used are the generalized form of Heckbert quantization and Maximum Likelihood (ML) respectively. With the continuous conflict over the impact of oil activities in the area, this work provides an initial basis of monitoring LC change, which is an important factor to consider in the design of an environmental decision-making framework. Landsat TM and ETM+ images of 1987 and 2002 were used to test the hybrid classification technique. The overall result indicates the ability to separate more LC classes. Furthermore, the approach provides a means of improving on the deficiencies of the unsupervised and supervised classification methods.
    Original languageEnglish
    Title of host publicationProceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.
    Number of pages7
    Publication statusPublished - 2007
    Event5th International Symposium Spatial Data Quality 2007: modelling qualities in space and time - Enschede, Netherlands
    Duration: 13 Jun 200715 Jun 2007
    http://www.isprs.org/proceedings/XXXVI/2-C43/

    Conference

    Conference5th International Symposium Spatial Data Quality 2007
    Abbreviated titleISPRS
    CountryNetherlands
    CityEnschede
    Period13/06/0715/06/07
    Internet address

    Fingerprint

    image classification
    land cover
    unsupervised classification
    Landsat thematic mapper
    analysis
    human activity
    decision making
    oil
    sampling
    monitoring
    cost

    Cite this

    Omo-Irabor, O., & Oduyemi, K. O. K. (2007). A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region. In Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.
    Omo-Irabor, Omoleomo ; Oduyemi, Kehinde O. K. / A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region. Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.. 2007.
    @inproceedings{b1fe829e4bc04e268e6ea0948692d6ca,
    title = "A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region",
    abstract = "The landscape of the Niger Delta region of Nigeria is undergoing rapid changes as a result of natural and anthropogenic activities. This necessitates the development of a rapid, cost effective and efficient land cover (LC) classification technique to monitor the biophysical dynamics in the region. Due to the intricate land cover patterns prominent in the study area and the irregularly indistinguishable relationship between land cover and spectral signals, this paper introduces a combined use of unsupervised and supervised image classification for detecting LC classes. The technique utilizes the spectral recognition of the unsupervised classification in the performance mode and the selection of sampling sites from a principal component analyzed image of the supervised classification in the training mode. The unsupervised and supervised classification algorithms used are the generalized form of Heckbert quantization and Maximum Likelihood (ML) respectively. With the continuous conflict over the impact of oil activities in the area, this work provides an initial basis of monitoring LC change, which is an important factor to consider in the design of an environmental decision-making framework. Landsat TM and ETM+ images of 1987 and 2002 were used to test the hybrid classification technique. The overall result indicates the ability to separate more LC classes. Furthermore, the approach provides a means of improving on the deficiencies of the unsupervised and supervised classification methods.",
    author = "Omoleomo Omo-Irabor and Oduyemi, {Kehinde O. K.}",
    year = "2007",
    language = "English",
    booktitle = "Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.",

    }

    Omo-Irabor, O & Oduyemi, KOK 2007, A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region. in Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.. 5th International Symposium Spatial Data Quality 2007, Enschede, Netherlands, 13/06/07.

    A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region. / Omo-Irabor, Omoleomo; Oduyemi, Kehinde O. K.

    Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.. 2007.

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

    TY - GEN

    T1 - A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region

    AU - Omo-Irabor, Omoleomo

    AU - Oduyemi, Kehinde O. K.

    PY - 2007

    Y1 - 2007

    N2 - The landscape of the Niger Delta region of Nigeria is undergoing rapid changes as a result of natural and anthropogenic activities. This necessitates the development of a rapid, cost effective and efficient land cover (LC) classification technique to monitor the biophysical dynamics in the region. Due to the intricate land cover patterns prominent in the study area and the irregularly indistinguishable relationship between land cover and spectral signals, this paper introduces a combined use of unsupervised and supervised image classification for detecting LC classes. The technique utilizes the spectral recognition of the unsupervised classification in the performance mode and the selection of sampling sites from a principal component analyzed image of the supervised classification in the training mode. The unsupervised and supervised classification algorithms used are the generalized form of Heckbert quantization and Maximum Likelihood (ML) respectively. With the continuous conflict over the impact of oil activities in the area, this work provides an initial basis of monitoring LC change, which is an important factor to consider in the design of an environmental decision-making framework. Landsat TM and ETM+ images of 1987 and 2002 were used to test the hybrid classification technique. The overall result indicates the ability to separate more LC classes. Furthermore, the approach provides a means of improving on the deficiencies of the unsupervised and supervised classification methods.

    AB - The landscape of the Niger Delta region of Nigeria is undergoing rapid changes as a result of natural and anthropogenic activities. This necessitates the development of a rapid, cost effective and efficient land cover (LC) classification technique to monitor the biophysical dynamics in the region. Due to the intricate land cover patterns prominent in the study area and the irregularly indistinguishable relationship between land cover and spectral signals, this paper introduces a combined use of unsupervised and supervised image classification for detecting LC classes. The technique utilizes the spectral recognition of the unsupervised classification in the performance mode and the selection of sampling sites from a principal component analyzed image of the supervised classification in the training mode. The unsupervised and supervised classification algorithms used are the generalized form of Heckbert quantization and Maximum Likelihood (ML) respectively. With the continuous conflict over the impact of oil activities in the area, this work provides an initial basis of monitoring LC change, which is an important factor to consider in the design of an environmental decision-making framework. Landsat TM and ETM+ images of 1987 and 2002 were used to test the hybrid classification technique. The overall result indicates the ability to separate more LC classes. Furthermore, the approach provides a means of improving on the deficiencies of the unsupervised and supervised classification methods.

    M3 - Conference contribution

    BT - Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.

    ER -

    Omo-Irabor O, Oduyemi KOK. A hybrid image classification approach for the systematic analysis of land cover (LC) changes in the Niger Delta region. In Proceedings of the 5th International Symposium on Spatial Data Quality (ISSDQ), 13-15 June 2007, Enschede, Netherlands.. 2007