Segment Based Image Analysis and Image Fusion

Over the last years, image fusion techniques have gained a renewed interest within the remote sensing community. The reason for this is that in most cases the new generation of remote sensors with very high spatial resolution records image datasets in two separate modes: the highest spatial resolution is obtained for panchromatic images whereas multispectral information is associated with lower spatial resolution. The ratios between panchromatic and multispectral imaging modes of one sensor vary between 1:2 and 1:5. For multisensor fusion, ratios can exceed 1:20 (e.g. Ikonos and SPOT merge). An emerging technique for the analysis of high-resolution images, on the other hand, is the use of segmentation algorithms as a pre-processing step. In this paper, we investigate techniques for the combination of image fusion and segment based image analysis. We will present a color preserving iconic fusion with subsequent segmentation and classification, a ‘cookie cutter’ approach for the integration of high resolution RGB and low resolution hyperspectral image data and a GIS based fusion technique for rapid image enhancement.