Remote sensing and GIS Research Topics?
The objective of this remote sensing study was to compare accuracies of pixel and object-based classification RGB and NIR images of the Kraansvlak coastal dune system in the Netherlands. Therefore, an evaluation of the influence of color balancing was necessary. With Definiens Developer rule sets for different shrub vegetations were tested for the entire Kraansvlak area. Field samples of the various classes were collected in the Kraansvlak. With these training sites a classification of a part of the Kraansvlak was done with ERDAS Imagine 9.1(pixel-based) and Definiens Developer 7.0 (object-based) and the accuracy results were compared. With Definiens Developer color balanced and original images were compared and the entire Kraansvlak was classified with the color images. For the RGB image an overall accuracy of 59.52% is obtained with ERDAS Imagine 9.1 and an overall accuracy of 67.6% is obtained with Definiens Developer. For the NIR image an overall accuracy of 66.67% is obtained with ERDAS Imagine 9.1 and an overall accuracy of 76.5% is achieved with Definiens Developer. For the color balanced images an overall accuracy of 64.8% is acquired. For the entire Kraansvlak an overall accuracy of 73.6% was obtained and rule sets are found for the shrubs which resulted in an overall accuracy of 71.6%. It can be concluded that the highest accuracies are obtained with NIR images in both approaches. However, a higher accuracy was obtained with Definiens Developer, the object based approach. In general, color balancing probably has a negative influence on the classification. The figure shows 4 examples of classifications using RGB and NIR images and corresponding overall accuracy of the Kraansvlak study area.
Remote sensing and GIS Research Topics
White Paper Team Meeting of the was held on January 18-19 at the Institute of Remote Sensing and Digital Earth (RADI) of the Chinese Academy of Sciences in Beijing, organized by the Committee on Data for Science and Technology (CODATA) of the International Council for Science (ICSU). We are currently working on the First White Paper on Gap Analysis on Open Data Interconnectivity for Global Disaster Risk Research, which will be published later this year.
In the Academy, he is working on “A system approach model for Citrus crop growth prediction using ground and remote sensing based information and data assimilation techniques”.
Remote Sensing and GIS MSC (1 year) - Courses Homepage
Chen, J. M., C. H. Menges, and S. G. Leblanc, 2005. Global derivation of the vegetation clumping index from multi-angular satellite data. Remote Sensing of Environment, 97: 447-457
CV-RESUME Bio MSc thesis Remote Sensing Past Reports ..
Earth Surface Processes, River Science, Remote Sensing and GIS, Water Quality, Contaminant Transport in Freshwater System, Micro & Emerging pollutants, Hydrogeochemistry, Urban pollution and sustainability, Metal speciation and Remediation techniques