Search

The work presented here tests an automatic procedure able to recognize the presence of built-up areas in the satellite images with the output nominal scale of 1:50,000. The input data is a set of 54 Ikonos and Quick Bird scenes considered as representative of the variety of human settlement patterns in large cities at global level. The methodology for automatic image information extraction is based on calculation of anisotropic rotation-invariant textural grey-level co-occurrence measures, also called PANTEX methodology. The total area analyzed covers 35,000 km2. The data under test shows high variety in latitude, season, sun elevation and sun azimuth at the time of image data collection. The output of the automatic image information retrieval is evaluated by comparison with a collection of reference information visually interpreted from the same satellite data input. Two complementary evaluation strategies are presented here: i) interactive selection of one threshold level in the textural measurement and then unsupervised application of the same threshold level to all the datasets under test, and ii) per-scene optimization of the threshold based on the available reference samples. This work briefly summarizes the nature of the errors and implications for global settlement classification.

Recent improvements in the spatial resolution of commercial satellite imagery make it possible to apply very highresolution (VHR) satellite data for assessing structural damage in the aftermath of humanitarian crises, such as, armed conflicts. Visual interpretation of pre- and post-crisis very high-resolution satellite imagery is the most straightforward method for discriminating structural damage and assessing its extent. However, the feasibility of using visual interpretation alone diminishes in the cases of large and dense urban settlements and spatial resolutions in the range of 2 m to 3 meters and larger. Visual interpretation can be further complicated at spatial resolutions greater than 1 m if accompanied by shadow formation and differences in sensor and solar conditions between the pre- and post-conflict images. In this study, we address these problems through investigating the use of traditional change techniques, namely, image differencing and principle component analysis, with an object-oriented image classification software, e-Cognition. Pre-conflict Ikonos (2 m resolution) images of Jenin in the Palestinian territories and Brest (1 m resolution) in FYROM were classified using the e-Cognition software. Thereafter, the pre-conflict classification was used to guide the classification, using e-Cognition, of the pixel-based change detection analysis. The second part of the study examines the feasibility of using mathematical morphological operators to automatically identify likely structurally damaged zones in dense urban settings. The overall results are promising and show that object-oriented segmentation and classification systems facilitate the interpretation of change detection results derived from very high-resolution (1 m and 2 m) commercial satellite data. The results show that objectoriented classification techniques enhance quantitative analysis of traditional pixel-based change detection applied to very high-resolution satellite data and facilitate the interpretation of changes in urban features. Finally, the results suggest that mathematical morphological methods are a potential new avenue for automatically extracting likely damaged zones from very high-resolution satellite imagery in the aftermath of disasters.