Remote Sensing for Agriculture, Ecosystems, and Hydrology VII, 2005
The success of any decision support system for managing wildfires lies on its ability to simulate... more The success of any decision support system for managing wildfires lies on its ability to simulate fire evolution. Therefore, accurate information on the natural fuel material in any area of interest is necessary. The present study aims to provide methodological tools to explore in depth the potential of new Earth Observation data for horizontal mapping of vegetated areas. Two approaches
This study investigates the potential of three advanced pixel window classification methods for h... more This study investigates the potential of three advanced pixel window classification methods for habitat mapping, namely Kernel based spatial Re-Classification (KRC), Radial Basis Function (RBF) neural networks (NN) and Support Vector Machines (SVM). KRC classifier takes into account the spatial arrangement and frequency of spectral classes present within a predefined square kernel. On the other hand, RBF-NN and SVM classifiers
Remote Sensing for Agriculture, Ecosystems, and Hydrology V, 2004
Kernel-based reclassification algorithm derives information on specific thematic classes on the b... more Kernel-based reclassification algorithm derives information on specific thematic classes on the basis of the frequency and spatial arrangement of land cover classes within a square kernel. This algorithm has been originally developed and validated for the urban environment. The present work investigates the potential of projecting this technique to the classification of very high spatial resolution satellite imagery of natural
International Journal of Applied Earth Observation and Geoinformation, 2011
Information on burnt area is of critical importance in many applications as for example in assess... more Information on burnt area is of critical importance in many applications as for example in assessing the disturbance of natural ecosystems due to a fire or in proving important information to policy makers on the land cover changes for establishing restoration policies of fire-affected regions. Such information is commonly obtained through remote sensing image thematic classification and a wide range of classifiers have been suggested for this purpose. The objective of the present study has been to investigate the use of Support Vector Machines (SVMs) classifier combined with multispectral Landsat TM image for obtaining burnt area mapping. As a case study a typical Mediterranean landscape in Greece was used, in which occurred one of the most devastating fires during the summer of 2007. Accuracy assessment was based on the classification overall statistical accuracy results and also on comparisons of the derived burnt area estimates versus validated estimates from the Risk-EOS Burnt Scar Mapping service. Results from the implementation of the SVM using diverse kernel functions showed an average overall classification accuracy of 95.87% and a mean kappa coefficient of 0.948, with the burnt area class always clearly separable from all the other classes used in the classification scheme. Total burnt area estimate computed from the SVM was also in close agreement with that from Risk-EOS (mean difference of less than 1%). Analysis also indicated that, at least for the studied here fire, the inclusion of the two middle infrared spectral bands TM5 and TM7 of TM sensor as well as the selection of the kernel function in SVM implementation have a negligible effect in both the overall classification performance and in the delineation of total burnt area. Overall, results exemplified the appropriateness of the spatial and spectral resolution of the Landsat TM imagery combined with the SVM in obtaining rapid and cost-effective post-fire analysis. This is of considerable scientific and practical value, given the present open access to the archived and new observations from this satellite radiometer globally.
The success of any decision support system for managing wildfires lies on its ability to simulate... more The success of any decision support system for managing wildfires lies on its ability to simulate fire evolution. Therefore, accurate information on the natural fuel material in any area of interest is necessary. The present study aims to provide methodological tools to explore in depth the potential of new Earth Observation data for horizontal mapping of vegetated areas. Two approaches are mainly described. The first one deals with the classification of ASTER visible, near- and short-wave infrared images in a detailed nomenclature including both different species and tree densities. This is important for wildfire studies since the same vegetation classes may represent completely different risk ignition levels depending on their morphological characteristics (i.e., trees height and density). The improvement of class separability using hyperspectral images acquired by Hyperion is also investigated. The second approach refers to a pattern recognition software tool for single tree counting using a very high spatial resolution image acquired by IKONOS-2 satellite. According to this approach, the regions dense in plants are identified by applying a suitable thresholding on the image. The resulted regions are further processed in order to estimate the number and location of single trees based on a pre-specified crown size per stratified zone. The outcome of the latter approach may provide direct evidence of tree density relating to ground biomass. Finally, the two approaches are used in a complementary manner to explore the possibilities offered by new sensor technology to override past limitations in species and fuel classification due to inadequate spectral/spatial resolution. The pilot application area is Mount. Pendeli and the east side of Mount. Parnitha, in the prefecture of Attiki, Greece.
ISPRS Journal of Photogrammetry and Remote Sensing, 2006
This study investigates the potential of three advanced pixel window classification methods for h... more This study investigates the potential of three advanced pixel window classification methods for habitat mapping, namely Kernel based spatial Re-Classification (KRC), Radial Basis Function (RBF) neural networks (NN) and Support Vector Machines (SVM). KRC classifier takes into account the spatial arrangement and frequency of spectral classes present within a predefined square kernel. On the other hand, RBF-NN and SVM classifiers
ABSTRACT Recent studies have indicated that for the first time since 1950, intense geophysical ac... more ABSTRACT Recent studies have indicated that for the first time since 1950, intense geophysical activity is occurring at the Santorini volcano. Here, we present and discuss the surface deformation associated with this activity, spanning from January 2011 to February 2012. Analysis of satellite interferometry data was performed using two well-established techniques, namely, Persistent Scatterer Interferometry (PSI) and Small Baseline Subset (SBAS), producing dense line-of-sight (LOS) ground deformation maps. The displacement field was compared with GPS observations from 10 continuous sites installed on Santorini. Results show a clear and large inflation signal, up to 150 mm/yr in the LOS direction, with a radial pattern outward from the center of the caldera. We model the deformation inferred from GPS and InSAR using a Mogi source located north of the Nea Kameni island, at a depth between 3.3 km and 6.3 km and with a volume change rate in the range of 12 million m3 to 24 million m3 per year. The latest InSAR and GPS data suggest that the intense geophysical activity has started to diminish since the end of February 2012.
Kernel-based reclassification algorithm derives information on specific thematic classes on the b... more Kernel-based reclassification algorithm derives information on specific thematic classes on the basis of the frequency and spatial arrangement of land cover classes within a square kernel. This algorithm has been originally developed and validated for the urban environment. The present work investigates the potential of projecting this technique to the classification of very high spatial resolution satellite imagery of natural
Fire monitoring and management in Europe, and in the wider Mediterranean region in particular, is... more Fire monitoring and management in Europe, and in the wider Mediterranean region in particular, is of paramount importance. Almost every summer massive forest wildfires break out in several areas across the Mediterranean, leaving behind severe destruction in forested and agricultural land, infrastructure and private property, and losses of human lives.
Remote Sensing for Agriculture, Ecosystems, and Hydrology VII, 2005
The success of any decision support system for managing wildfires lies on its ability to simulate... more The success of any decision support system for managing wildfires lies on its ability to simulate fire evolution. Therefore, accurate information on the natural fuel material in any area of interest is necessary. The present study aims to provide methodological tools to explore in depth the potential of new Earth Observation data for horizontal mapping of vegetated areas. Two approaches
This study investigates the potential of three advanced pixel window classification methods for h... more This study investigates the potential of three advanced pixel window classification methods for habitat mapping, namely Kernel based spatial Re-Classification (KRC), Radial Basis Function (RBF) neural networks (NN) and Support Vector Machines (SVM). KRC classifier takes into account the spatial arrangement and frequency of spectral classes present within a predefined square kernel. On the other hand, RBF-NN and SVM classifiers
Remote Sensing for Agriculture, Ecosystems, and Hydrology V, 2004
Kernel-based reclassification algorithm derives information on specific thematic classes on the b... more Kernel-based reclassification algorithm derives information on specific thematic classes on the basis of the frequency and spatial arrangement of land cover classes within a square kernel. This algorithm has been originally developed and validated for the urban environment. The present work investigates the potential of projecting this technique to the classification of very high spatial resolution satellite imagery of natural
International Journal of Applied Earth Observation and Geoinformation, 2011
Information on burnt area is of critical importance in many applications as for example in assess... more Information on burnt area is of critical importance in many applications as for example in assessing the disturbance of natural ecosystems due to a fire or in proving important information to policy makers on the land cover changes for establishing restoration policies of fire-affected regions. Such information is commonly obtained through remote sensing image thematic classification and a wide range of classifiers have been suggested for this purpose. The objective of the present study has been to investigate the use of Support Vector Machines (SVMs) classifier combined with multispectral Landsat TM image for obtaining burnt area mapping. As a case study a typical Mediterranean landscape in Greece was used, in which occurred one of the most devastating fires during the summer of 2007. Accuracy assessment was based on the classification overall statistical accuracy results and also on comparisons of the derived burnt area estimates versus validated estimates from the Risk-EOS Burnt Scar Mapping service. Results from the implementation of the SVM using diverse kernel functions showed an average overall classification accuracy of 95.87% and a mean kappa coefficient of 0.948, with the burnt area class always clearly separable from all the other classes used in the classification scheme. Total burnt area estimate computed from the SVM was also in close agreement with that from Risk-EOS (mean difference of less than 1%). Analysis also indicated that, at least for the studied here fire, the inclusion of the two middle infrared spectral bands TM5 and TM7 of TM sensor as well as the selection of the kernel function in SVM implementation have a negligible effect in both the overall classification performance and in the delineation of total burnt area. Overall, results exemplified the appropriateness of the spatial and spectral resolution of the Landsat TM imagery combined with the SVM in obtaining rapid and cost-effective post-fire analysis. This is of considerable scientific and practical value, given the present open access to the archived and new observations from this satellite radiometer globally.
The success of any decision support system for managing wildfires lies on its ability to simulate... more The success of any decision support system for managing wildfires lies on its ability to simulate fire evolution. Therefore, accurate information on the natural fuel material in any area of interest is necessary. The present study aims to provide methodological tools to explore in depth the potential of new Earth Observation data for horizontal mapping of vegetated areas. Two approaches are mainly described. The first one deals with the classification of ASTER visible, near- and short-wave infrared images in a detailed nomenclature including both different species and tree densities. This is important for wildfire studies since the same vegetation classes may represent completely different risk ignition levels depending on their morphological characteristics (i.e., trees height and density). The improvement of class separability using hyperspectral images acquired by Hyperion is also investigated. The second approach refers to a pattern recognition software tool for single tree counting using a very high spatial resolution image acquired by IKONOS-2 satellite. According to this approach, the regions dense in plants are identified by applying a suitable thresholding on the image. The resulted regions are further processed in order to estimate the number and location of single trees based on a pre-specified crown size per stratified zone. The outcome of the latter approach may provide direct evidence of tree density relating to ground biomass. Finally, the two approaches are used in a complementary manner to explore the possibilities offered by new sensor technology to override past limitations in species and fuel classification due to inadequate spectral/spatial resolution. The pilot application area is Mount. Pendeli and the east side of Mount. Parnitha, in the prefecture of Attiki, Greece.
ISPRS Journal of Photogrammetry and Remote Sensing, 2006
This study investigates the potential of three advanced pixel window classification methods for h... more This study investigates the potential of three advanced pixel window classification methods for habitat mapping, namely Kernel based spatial Re-Classification (KRC), Radial Basis Function (RBF) neural networks (NN) and Support Vector Machines (SVM). KRC classifier takes into account the spatial arrangement and frequency of spectral classes present within a predefined square kernel. On the other hand, RBF-NN and SVM classifiers
ABSTRACT Recent studies have indicated that for the first time since 1950, intense geophysical ac... more ABSTRACT Recent studies have indicated that for the first time since 1950, intense geophysical activity is occurring at the Santorini volcano. Here, we present and discuss the surface deformation associated with this activity, spanning from January 2011 to February 2012. Analysis of satellite interferometry data was performed using two well-established techniques, namely, Persistent Scatterer Interferometry (PSI) and Small Baseline Subset (SBAS), producing dense line-of-sight (LOS) ground deformation maps. The displacement field was compared with GPS observations from 10 continuous sites installed on Santorini. Results show a clear and large inflation signal, up to 150 mm/yr in the LOS direction, with a radial pattern outward from the center of the caldera. We model the deformation inferred from GPS and InSAR using a Mogi source located north of the Nea Kameni island, at a depth between 3.3 km and 6.3 km and with a volume change rate in the range of 12 million m3 to 24 million m3 per year. The latest InSAR and GPS data suggest that the intense geophysical activity has started to diminish since the end of February 2012.
Kernel-based reclassification algorithm derives information on specific thematic classes on the b... more Kernel-based reclassification algorithm derives information on specific thematic classes on the basis of the frequency and spatial arrangement of land cover classes within a square kernel. This algorithm has been originally developed and validated for the urban environment. The present work investigates the potential of projecting this technique to the classification of very high spatial resolution satellite imagery of natural
Fire monitoring and management in Europe, and in the wider Mediterranean region in particular, is... more Fire monitoring and management in Europe, and in the wider Mediterranean region in particular, is of paramount importance. Almost every summer massive forest wildfires break out in several areas across the Mediterranean, leaving behind severe destruction in forested and agricultural land, infrastructure and private property, and losses of human lives.
Uploads
Papers by Charalambos Kontoes