A new way of image data fusion: the multi-spectral cooperative segmentation

The paper focuses on the problem of the multi-spectral image segmentation. A multi-spectral image is composed by several mono-spectral images (black and white for example), coming from different wavelength bands. Consequently, the complementarity and/or redundancy of data-through the data fusion-make reliable and robust military or industrial systems. This method is a new approach of image data fusion. It does not enter the classification of image data fusion methods, because its level of fusion is between two classical levels. The image data fusion is performed while segmenting together the different images. The image data fusion is treated as an extension of segmentation methods. The classical or mono-spectral image features: edges, regions are being extended to the multi-spectral framework. Their attributes gather the information coming from all spectral images