Learn More
In this paper, we present a novel multiscale texture model and a related algorithm for the unsupervised segmentation of color images. Elementary textures are characterized by their spatial interactions with neighboring regions along selected directions. Such interactions are modeled, in turn, by means of a set of Markov chains, one for each direction, whose(More)
A new region-based algorithm is proposed for the compression of multispectral images. The image is segmented in homogeneous regions, each of which is subject to spectral KLT, spatial shape-adaptive DWT, and SPIHT encoding. We propose to use a dedicated KLT for each region or for each class rather than a single global KLT. Experiments show that the(More)
The analysis and extraction of textural information from image data is a relevant topic in the image analysis and processing domains, mainly due to the large number of application areas that it concerns. The Hierarchical Multiple Markov Chain (H-MMC) family of models has been recently introduced [2] to provide a simple and effective tool to represent the(More)
Tree-structured Markov random fields have been recently proposed in order to model complex images and to allow for their fast and accurate segmentation. By modeling the image as a tree of regions and subre-gions, the original K-ary segmentation problem can be recast as a sequence of reduced-dimensionality steps, thus reducing computational complexity and(More)
1. EXTENDED ABSTRACT In the remote-sensing field, significant advances in sensor design have been occurring for the last years, thanks to a systematic and fast evolution of the related technologies. As a consequence, a larger and larger amount of different typologies of remotely sensed data is currently available. In this context, an effective synergistic(More)