Search

The monitoring of land cover requires that stable land cover classes be distinguished from changes over time. Within this paper, a postclassification method is presented that provides land cover change information, based on a time series of land cover maps. The method applies a kernel filter to sequential land cover maps. Under some basic assumptions, it shows robustness against classification errors. Despite seasonality, land cover changes often occur at a low temporal frequency (e.g., maximum once every 5–10 years). If land cover maps are available more frequently, some of the information will become redundant (oversampling). The proposed method uses this redundancy for tolerating (nonsystematic) misclassifications. In order to demonstrate the benefits and limitations of the proposed method, analytical expressions have been derived. When compared to a simple postclassification comparison, one of the key strengths of the proposed approach is that it is able to improve both the overall and user's accuracy of change, while also maintaining the same level of producer's accuracy. As a case study, MODerate Resolution Imaging Spectroradiometer remote sensing data from 2006-2010 were classified into forest (F)/nonforest (NF) at pan-European scale. Promising results were obtained for detecting forest loss due to natural disasters. Quality was assessed using burnt area maps in southern Europe and a forest damage report after a windstorm in France. Results indicated a considerable reduction of change detection errors, confirming the theoretical results. [ABSTRACT FROM PUBLISHER]

European Remote Sensing satellite (ERS) ENVISAT coherence is a new repeat-pass interferometric synthetic aperture radar product characterized by a very short repeat-pass interval between acquisitions (28 mm). In this paper, we investigate the properties of two 28-mins coherence images acquired over the regions of Paris, France, and Prague, Czech Republic, as well as a 35-day coherence image from Prague. Bare soils showed higher coherence compared to forests and urban areas. The range of coherence was larger for 28-mm pairs, perpendicular baseline closer to 2 km, and stable weather conditions between acquisitions. Coherence modeling shows that because of the long baseline: 1) significant surface decorrelation occurs when slopes are not accounted for in the common-band filtering and for variation of topography in the resolution cell and 2) volume decorrelation is significant. Temporal decorrelation seems to be relevant for vegetation even at the 28-mm interval time scale. Observations and modeled coherence were found to be in good agreement. Land cover classification using coherence and backscatter has been tested for the Paris scene for mapping the four major classes. The classification accuracy was 86% when an edge-eroded version of the reference land cover map was used. It decreases to 68% when the full land cover map was used as reference because of local mismatches between the coherence image and the land cover map, and the reduced common band in range. The high coherence difference between forests and bare fields suggests the possibility to use the ERS-ENVISAT coherence for forest/nonforest mapping and estimation of biophysical properties of short vegetation. [ABSTRACT FROM AUTHOR]

The article presents information on the 2003 IEEE Geoscience and Remote Sensing Society's (GRS-S) Awards that were presented at the IGARSS'03 banquet on July 24 at Hotel Dieu in Toulouse, France. Charles Luther, the president of the society, presented those awards with the assistance of the Awards Committee Chairman Werner Wiesbeck. The award ceremony was started with the recognition of the IEEE Fellows who have declared the GRS-Society as their home society. The GRS-S Transactions Editor, based on recommendations of associate editors and reviewers, compiled the list of nominees for the Transactions Prize Paper Award.