Advanced search
1 file | 1.81 MB

An assessment of geometric activity features for per-pixel classification of urban man-made objects using very high resolution satellite imagery

Author
Organization
Abstract
In this paper, we propose the use of Geometric Activity (GA) features for detecting man-made objects in urban areas using VHR satellite imagery. These features describe the geometric context of a pixel without the necessity of segmentation and can be integrated as extra bands in a per-pixel classification. Two main types of GA features were investigated: ridge features based on the well-known facet model and morphological features obtained by applying closing transforms with structuring elements of different size and shape. Our findings show a substantial increase in classification accuracy for the man-made object classes "roads and buildings with dark roof" after inclusion of GA features. Next to GA features, the use of object-based features derived from eCognition(R), containing both geometric and textural information, was also investigated for per-pixel classification. Accuracies obtained with object-based features are comparable to the accuracies obtained with GA features. The inclusion of both GA features and object-based features further improves the overall accuracy. GA features and object-based features thus contain complementary information.
Keywords
MACHINE LEARNING ALGORITHMS, LAND-COVER CLASSIFICATION, AUTOMATIC ROAD EXTRACTION, REMOTE-SENSING IMAGES, MULTISPECTRAL IMAGERY, SPATIAL-RESOLUTION, TEXTURE, MULTISCALE, MODEL

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.81 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Chan, Jonathan Cheung-Wai, Rik Bellens, Frank Canters, and Sidharta Gautama. 2009. “An Assessment of Geometric Activity Features for Per-pixel Classification of Urban Man-made Objects Using Very High Resolution Satellite Imagery.” Photogrammetric Engineering and Remote Sensing 75 (4): 397–411.
APA
Chan, J. C.-W., Bellens, R., Canters, F., & Gautama, S. (2009). An assessment of geometric activity features for per-pixel classification of urban man-made objects using very high resolution satellite imagery. PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING, 75(4), 397–411.
Vancouver
1.
Chan JC-W, Bellens R, Canters F, Gautama S. An assessment of geometric activity features for per-pixel classification of urban man-made objects using very high resolution satellite imagery. PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING. American Society for Photogrammetry and remote sensing; 2009;75(4):397–411.
MLA
Chan, Jonathan Cheung-Wai, Rik Bellens, Frank Canters, et al. “An Assessment of Geometric Activity Features for Per-pixel Classification of Urban Man-made Objects Using Very High Resolution Satellite Imagery.” PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING 75.4 (2009): 397–411. Print.
@article{539526,
  abstract     = {In this paper, we propose the use of Geometric Activity (GA) features for detecting man-made objects in urban areas using VHR satellite imagery. These features describe the geometric context of a pixel without the necessity of segmentation and can be integrated as extra bands in a per-pixel classification. Two main types of GA features were investigated: ridge features based on the well-known facet model and morphological features obtained by applying closing transforms with structuring elements of different size and shape. Our findings show a substantial increase in classification accuracy for the man-made object classes {\textacutedbl}roads and buildings with dark roof{\textacutedbl} after inclusion of GA features. Next to GA features, the use of object-based features derived from eCognition(R), containing both geometric and textural information, was also investigated for per-pixel classification. Accuracies obtained with object-based features are comparable to the accuracies obtained with GA features. The inclusion of both GA features and object-based features further improves the overall accuracy. GA features and object-based features thus contain complementary information.},
  author       = {Chan, Jonathan Cheung-Wai and Bellens, Rik and Canters, Frank and Gautama, Sidharta},
  issn         = {0099-1112},
  journal      = {PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING},
  keyword      = {MACHINE LEARNING ALGORITHMS,LAND-COVER CLASSIFICATION,AUTOMATIC ROAD EXTRACTION,REMOTE-SENSING IMAGES,MULTISPECTRAL IMAGERY,SPATIAL-RESOLUTION,TEXTURE,MULTISCALE,MODEL},
  language     = {eng},
  number       = {4},
  pages        = {397--411},
  publisher    = {American Society for Photogrammetry and remote sensing},
  title        = {An assessment of geometric activity features for per-pixel classification of urban man-made objects using very high resolution satellite imagery},
  volume       = {75},
  year         = {2009},
}

Web of Science
Times cited: