GEOBIA Labs & People
Kasper has extensive experience in the use of object-based image analysis of high spatial resolution image data sets and in particular the use of the eCognition Developer, Architect and Server software. Kasper has used object-based image analysis for high spatial resolution satellite image data, airborne optical, thermal and LiDAR data as well as several UAV derived data sets. He has applied these techniques in a variety of environments, including riparian zones, agriculture (sugarcane, oil palms, banana, avocado, mango, and macadamia plantations), mine sites, rangelands, mangroves, and urban areas for feature extraction. Application areas in object-based image analysis include but are not limited to change detection, comparison of object- and pixel-based approaches, tree crown delineation, gully erosion mapping, vegetation structural parameter mapping, and land-cover/land-use mapping and monitoring.
Stuart has used object-based image analysis for segmentation and classification of high spatial resolution images of coral reefs and for field survey data to image objects. He is supervising a number of projects using object-based image analysis and the eCognition software to map riparian, mangrove, wetland and seagrass/coral environments from high and moderate spatial resolution images. Stuart has a long standing (>20 years) background in multi-scale image analysis and hierarchy theory and is very keen to use both of these in object-based image analysis.
Since 2008 Chris used object based image analysis using eCognition as a tool to derive biophysical properties describing the habitat of coral reef, seagrass and mangrove environments in space and time. This research included a variety of multi and hyperspectral high and moderate spatial resolution imagery integrated with field data of the various environments. Habitat maps have since been created for coral reefs and coastal zones throughout the Asia Pacific region and at various extent, covering one or several image scenes. His most recent projects focus on time series analysis of seagrass and coral reef environments using semi-automated OBIA approaches, and on developing mapping approaches to map the geomorphic zonation and benthic composition of the Great Barrier Reef.
Peter works with a cross disciplinary team at the Joint Remote Sensing Research Centre (JRSRP) to develop rigorous data management systems and validated mathematical models that measure and map both major and more subtle changes in Australia's vegetation using large earth observation data sets and vegetation survey field data systems.
Jasmine has experience working with the open source package – RSGISLib for object based image analysis, mapping wooded vegetation extent in SEQ. Jasmine has previously used eCognition to classify landuse change in central Vietnam, and classify mangrove community types in northern Australia.
Nick Goodwin and Lisa Collett
Development and maintenance of operational Landsat burnt area (fire scar) mapping program and associated R&D to improve the timeliness and accuracy of burnt area maps. Burnt area mapping, and associated time series based cloud and shadow detection uses a combination of time series change detection, simple image segmentation, and a classification tree to produce maps of burnt vegetation for every image in a time sequence of Landsat TM, ETM+ and OLI surface reflectance images.
The primary area of Professor Lucas’ expertise is in quantifying and understanding the response of terrestrial ecosystems and environments to change, including that associated with climatic variation, through integration of remote sensing data from various sources.