Change detection of isolated housing using a new hybrid approach based on object classification with optical and TerraSAR-X data

Abstract

Optical and microwave high spatial resolution images are now available for a wide range of applications. In this work, they have been applied for the semi-automatic change detection of isolated housing in agricultural areas. This article presents a new hybrid methodology based on segmentation of high-resolution images and image differencing. This new approach mixes the main techniques used in change detection methods and it also adds a final segmentation process in order to classify the change detection product. First, isolated building classification is carried out using only optical data. Then, synthetic aperture radar (SAR) information is added to the classification process, obtaining excellent results with lower complexity cost. Since the first classification step is improved, the total change detection scheme is also enhanced when the radar data are used for classification. Finally, a comparison between the different methods is presented and some conclusions are extracted from the study.

Acknowledgements

The authors thank the ‘Instituto Cartográfico de Valencia’ for the provision of aerial images. They also want to thank the German Aerospace Centre (DLR), which provided the TerraSAR-X data in the frame of a scientific use agreement.