Adaptive segmentation system

We propose an adaptive image segmentation system composed of three processing modules. The first module enables to determine the global context of the image to process (image mainly composed of uniform regions and textured ones) and to localize textured and uniform areas. The second module regards the local analysis of the image to segment in order to characterize each detected area considering different types of attributes. This more precise analysis of each region allows to make the choice of the segmentation method easier and secondly to adapt the analysis window size of the region to segment. Finally, the third module triggers the segmentation method which is adapted to the local context of the image by using an unsupervised classification method. We show the efficiency of the system through some experimental results.