287 resultados para Video genre classification


Relevância:

20.00% 20.00%

Publicador:

Resumo:

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Sur le plan économique, le système de genre est une pierre angulaire du discours publicitaire. Il intervient dans la segmentation des marchés, dans la sélection des médias et des supports, dans l'apparence extérieure des produits, dans le ton des campagnes, dans le choix des arguments de vente et, bien sûr, dans les scripts des annonces qui mettent en scène, en grand nombre, des êtres humains. En contrepartie, sur le plan symbolique, le discours publicitaire est un dépositaire privilégié des imaginaires de genre qui circulent dans son contexte de production et de diffusion. En cette qualité, confronté aux lois d'un marché toujours plus concurrentiel, à une segmentation plus fine des cibles, à la multiplication des supports, à l'instabilité croissante des consommateurs ainsi qu'à une critique médiatique, académique et publique toujours prompte à relever sa tendance au stéréotypage, le discours publicitaire est amené à proposer des représentations des hommes et des femmes de plus en plus variées et complexes. La présente étude, qui relève de l'analyse linguistique des discours, a pour objectif d'entrer dans la complexité de ces variations publicitaires contemporaines sur le féminin et le masculin et de déchiffrer les imaginaires de genre qu'elles contribuent à construire. Après un état des lieux des travaux consacrés à la représentation publicitaire des sexes ainsi qu'une présentation détaillée des jalons théoriques et méthodologiques de l'approche adoptée, une analyse de contenu, réalisée sur la base d'un corpus de plus 1200 annonces, met en évidence les configurations récurrentes du masculin et du féminin dans la production publicitaire contemporaine de presse magazine. Une analyse textuelle et critique interroge ensuite le rôle de la langue dans le processus de schématisation des imaginaires publicitaires de genre. Dans un premier temps, grâce à une prise en compte des déterminations prédiscursive et discursive des représentations publicitaires du féminin et du masculin, cette analyse montre comment le discours publicitaire, en plus de différencier radicalement le féminin et le masculin, tend à essentialiser cette différenciation au travers de deux procédés discursifs d'idéologisation, la catégorisation et la généralisation. Dans un second temps, un inventaire thématique des variations publicitaires contemporaines sur le genre permet d'évaluer la perméabilité du discours publicitaire à la reconfiguration du système de genre qui est en marche dans notre société depuis la seconde moitié du vingtième siècle. La présente recherche, qui entend globalement déconstruire ce qui prend trop souvent l'apparence d'évidences et soumettre à débat des interprétations, thématise par ailleurs la question de la dimension politique des recherches académiques.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

In this paper, mixed spectral-structural kernel machines are proposed for the classification of very-high resolution images. The simultaneous use of multispectral and structural features (computed using morphological filters) allows a significant increase in classification accuracy of remote sensing images. Subsequently, weighted summation kernel support vector machines are proposed and applied in order to take into account the multiscale nature of the scene considered. Such classifiers use the Mercer property of kernel matrices to compute a new kernel matrix accounting simultaneously for two scale parameters. Tests on a Zurich QuickBird image show the relevance of the proposed method : using the mixed spectral-structural features, the classification accuracy increases of about 5%, achieving a Kappa index of 0.97. The multikernel approach proposed provide an overall accuracy of 98.90% with related Kappa index of 0.985.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

In this paper, we consider active sampling to label pixels grouped with hierarchical clustering. The objective of the method is to match the data relationships discovered by the clustering algorithm with the user's desired class semantics. The first is represented as a complete tree to be pruned and the second is iteratively provided by the user. The active learning algorithm proposed searches the pruning of the tree that best matches the labels of the sampled points. By choosing the part of the tree to sample from according to current pruning's uncertainty, sampling is focused on most uncertain clusters. This way, large clusters for which the class membership is already fixed are no longer queried and sampling is focused on division of clusters showing mixed labels. The model is tested on a VHR image in a multiclass classification setting. The method clearly outperforms random sampling in a transductive setting, but cannot generalize to unseen data, since it aims at optimizing the classification of a given cluster structure.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Mature T-cell and T/NK-cell neoplasms are both uncommon and heterogeneous, among the broad category of non-Hodgkin's lymphomas. Due to the lack of specific genetic alterations in the vast majority of cases, most currently defined entities show overlapping morphologic and immunophenotypic features and therefore pose a challenge to the diagnostic pathologist. The goal of the symposium is to address current criteria for the recognition of specific subtypes of T-cell lymphoma, and to highlight new data regarding emerging immunophenotypic or molecular markers. This activity has been designed to meet the needs of practicing pathologists, and residents and fellows enrolled in training programs in anatomic and clinical pathology. It should be a particular benefit to those with an interest in hematopathology. Upon completion of this activity, participants should be better able to: -To be able to state the basis for the classification of mature T-cell malignancies involving nodal and extranodal sites. -To recognize and accurately diagnose the various subtypes of nodal and extranodal peripheral T-cell lymphomas. -To utilize immunohistochemical and molecular tests to characterize atypical T-cell proliferations. -To recognize and accurately diagnose T-cell lymphoproliferative lesions involving the skin and gastrointestinal tract, and be able to provide guidance regarding their clinical aggressiveness and management -To be able to utilize flow cytometric data to identify diverse functional T-cell subsets.