The Experts below are selected from a list of 66 Experts worldwide ranked by ideXlab platform

K. Vinod Kumar - One of the best experts on this subject based on the ideXlab platform.

  • Segment optimization and data-driven thresholding for knowledge-based landslide detection by object-based image analysis
    IEEE Transactions on Geoscience and Remote Sensing, 2011
    Co-Authors: Tapas Ranjan Martha, Cees J. Van Westen, Victor Jetten, Norman Kerle, K. Vinod Kumar
    Abstract:

    To detect landslides by object-based image analysis using criteria based on shape, color, texture, and, in particular, contextual information and process knowledge, candidate segments must be delineated properly. This has proved challenging in the past, since segments are mainly created using spectral and size criteria that are not consistent for landslides. This paper presents an approach to select objectively parameters for a region growing segmentation technique to outline landslides as individual segments and also addresses the scale dependence of landslides and false positives occurring in a natural landscape. Multiple scale parameters were determined using a plateau objective function derived from the spatial autocorrelation and intrasegment variance analysis, allowing for differently sized features to be identified. While a high-resolution Resourcesat-1 Linear Imaging and Self Scanning Sensor IV (5.8 m) multispectral image was used to create segments for landslide recognition, terrain curvature derived from a digital terrain model based on Cartosat-1 (2.5 m) data was used to create segments for subsequent landslide classification. Here, optimal segments were used in a knowledge-based classification approach with the thresholds of diagnostic parameters derived from If-means cluster analysis, to detect landslides of five different types, with an overall recognition accuracy of 76.9%. The approach, when tested in a geomorphologically dissimilar area, recognized landslides with an overall accuracy of 77.7%, without modification to the methodology. The multiscale classification-based segment optimization procedure was also able to reduce the Error of Commission significantly in comparison to a single-optimal-scale approach.

L.m. Marafa - One of the best experts on this subject based on the ideXlab platform.

Thomas Blaschke - One of the best experts on this subject based on the ideXlab platform.

  • Detection of gully-affected areas by applying object-based image analysis (OBIA) in the region of Taroudannt, Morocco
    Remote Sensing, 2014
    Co-Authors: Sebastian D'oleire-oltmanns, Irene Marzolff, Dirk Tiede, Thomas Blaschke
    Abstract:

    This study aims at the detection of gully-affected areas by applying object-based image analysis in the region of Taroudannt, Morocco, which is highly affected by gully erosion while simultaneously representing a major region of agro-industry with a high demand of arable land. As high-resolution optical satellite data are readily available from various sensors and with a much better temporal resolution than 3D terrain data, an area-wide mapping approach to extract gully-affected areas using only optical satellite imagery was developed. The methodology additionally incorporates expert knowledge and freely-available vector data in a cyclic object-based image analysis approach. This connects the two fields of geomorphology and remote sensing. The classification results show the successful implementation of the developed approach and allow conclusions on the current distribution of gullies. The results of the classification were checked against manually delineated reference data incorporating expert knowledge based on several field campaigns in the area, resulting in an overall classification accuracy of 62%. The Error of omission accounts for 38% and the Error of Commission for 16%, respectively. Additionally, a manual assessment was carried out to assess the quality of the applied classification algorithm. The limited Error of omission contributes with 23% to the overall Error of omission and the limited Error of Commission contributes with 98% to the overall Error of Commission. This assessment improves the results and confirms the high quality of the developed approach for area-wide mapping of gully-affected areas in larger regions. In the field of landform mapping, the overall quality of the classification results is often assessed with more than one method to incorporate all aspects adequately.

Tapas Ranjan Martha - One of the best experts on this subject based on the ideXlab platform.

  • Segment optimization and data-driven thresholding for knowledge-based landslide detection by object-based image analysis
    IEEE Transactions on Geoscience and Remote Sensing, 2011
    Co-Authors: Tapas Ranjan Martha, Cees J. Van Westen, Victor Jetten, Norman Kerle, K. Vinod Kumar
    Abstract:

    To detect landslides by object-based image analysis using criteria based on shape, color, texture, and, in particular, contextual information and process knowledge, candidate segments must be delineated properly. This has proved challenging in the past, since segments are mainly created using spectral and size criteria that are not consistent for landslides. This paper presents an approach to select objectively parameters for a region growing segmentation technique to outline landslides as individual segments and also addresses the scale dependence of landslides and false positives occurring in a natural landscape. Multiple scale parameters were determined using a plateau objective function derived from the spatial autocorrelation and intrasegment variance analysis, allowing for differently sized features to be identified. While a high-resolution Resourcesat-1 Linear Imaging and Self Scanning Sensor IV (5.8 m) multispectral image was used to create segments for landslide recognition, terrain curvature derived from a digital terrain model based on Cartosat-1 (2.5 m) data was used to create segments for subsequent landslide classification. Here, optimal segments were used in a knowledge-based classification approach with the thresholds of diagnostic parameters derived from If-means cluster analysis, to detect landslides of five different types, with an overall recognition accuracy of 76.9%. The approach, when tested in a geomorphologically dissimilar area, recognized landslides with an overall accuracy of 77.7%, without modification to the methodology. The multiscale classification-based segment optimization procedure was also able to reduce the Error of Commission significantly in comparison to a single-optimal-scale approach.

Tung Fung - One of the best experts on this subject based on the ideXlab platform.