T-SWEETS: An Alternative to the Stimulus Collaboration from Trust Inference in Social Networks

Evaluating an adaptive windowing scheme in speckle noise MAP filtering

Milhomem Silva, E.Oliveira Rodrigues, D.Gomes de Souza, J.Salgado, A.C.Lemos Meira, S.R.

F. N. S. MedeirosN. D. A. MascarenhasR. C. P. MarquesC. M. Laprano

This paper presents a novel method to infer trustiness in social networks entitled T-SWEETS. It is also presents its application at Konnen, a knowledge management platform on social networks based. The features explored in T-SWEETS come from research done on related works from the literature and from an experiment development with a group of 34 people. T-SWEETS have been main purpose to act as an incentive to collaboration in social networks and, therefore, naturally increase the frequency of knowledge dissemination among the users. Thus, it provides background to others automated systems (e.g. recommender systems) and, therefore, the knowledge produced by these users can be explored more efficiently. SAR images are corrupted by speckle noise, which degrades the quality and interpretation of the images. Speckle removal provides a better interpretability of SAR images if the techniques perform the filtering without loss of spatial resolution and preserve fine details and edges. This work aims to redefine the neighborhood areas around the noisy pixel and in this area the local mean and variance are computed to estimate the MAP filter parameters. The proposed modified MAP algorithm improves the ability to filter the speckle noise without blurring edges and targets by applying the MAP estimator in the current adaptive window that is controlled by a measure of homogeneity in the area around the noisy pixel. The effectiveness in reducing speckle by the modified MAP filter is evaluated in terms of qualitative and quantitative aspects such as line and edge preservation and the improvement of the signal to noise ratio.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: