General Adaptive Neighborhood Image Restoration, Enhancement and Segmentation

This paper aims to outline the General Adaptive Neighborhood Image Processing (GANIP) approach [1–3], which has been recently introduced. An intensity image is represented with a set of local neighborhoods defined for each point of the image to be studied. These so-called General Adaptive Neighborhoods (GANs) are simultaneously adaptive with the spatial structures, the analyzing scales and the physical settings of the image to be addressed and/or the human visual system. After a brief theoretical introductory survey, the GANIP approach will be successfully applied on real application examples in image restoration, enhancement and segmentation. 1 The General Adaptive Neighborhood (GAN) Paradigm This paper deals with 2D intensity images, that is to say image mappings defined on a spatial support D in the Euclidean space R and valued into a gray tone range, which is a real numbers interval. The General Adaptive Neighborhood paradigm has been introduced in order to propose an original image representation for adaptive processing and analysis. The central idea is the notion of adaptivity which is simultaneously associated to the analyzing scales, the spatial structures and the intensity values of the image to be addressed. 1.1 Adaptivity with Analyzing Scales A multiscale image representation such as wavelet decomposition [4] or isotropic scale-space [5], generally takes into account analyzing scales which are global and a priori defined, that is to say extrinsic scales. This kind of multiscale analysis presents a main drawback since a priori knowledge, relating to the features of the studied image, is consequently required. On the contrary, an intrinsic multiscale representation such as anisotropic scale-space [6], takes advantage of scales which are self-determined by the local image structures. Such a decomposition does not need any a priori information. ha l-0 01 28 11 1, v er si on 1 30 J an 2 00 7 Author manuscript, published in "Image Analysis and Recognition, France (2006)"