Workshop IMTA 2009 Abstracts


Full Papers
Paper Nr: 3
Title:

Disparity Measure Construction for Comparison of 3D Objects’ Surfaces

Authors:

Natalya Dyshkant

Abstract: In this paper a problem of 3D objects’ surfaces comparison is considered. Each spatial object is given as a set of schlicht surfaces that are described by point clouds. This article discusses a proposed disparity measure to compare such objects and an algorithm to compute it. A method for comparison of mesh functions defined on different point sets is proposed. The theoretical base of the proposed approach is the piecewise-linear approximation of surfaces using Delaunay triangulations for initial point clouds. The presented approach uses Delaunay triangulations of each point clouds, general Delaunay triangulation for both clouds, function interpolation on basis of localization of triangulations in each other and function comparison on single cells of general triangulation. Localization is implemented on basis of minimum spanning trees. As the application of the proposed methodology a problem of 3D face models comparison is considered. It was experimentally verified that the proposed method is numerically efficient.

Paper Nr: 7
Title:

Basic Definitions and Operations for Gestalt Algebra

Authors:

Eckart Michaelsen and Jochen Meidow

Abstract: The gestalt algebra is a mathematical construction designed to capture the perceptual structure of complex patterns. Such patterns occur e.g. in aerial images of urban terrain. The principles of gestalt construction – namely proximity, good continuation, similarity and symmetry – are used in a recursive way to describe images or scene data using terms following an algebraic defini-tion. Such description can be used for recognition, matching or data mining.

Paper Nr: 9
Title:

The Elaboration and Clinical Testing of a New Technique of Image Quality Improvement in Ultrasound Medical Diagnostics

Authors:

N. S. Kulberg, T. V. Yakovleva, Yu. R. Kamalov, V. A. Sandrikov, L. V. Osipov and P. A. Belov

Abstract: The subject of the present research is solving the problem of the noisy and the informative texture elements separation with taking into account the specific traits of the ultrasound visualization. A noise suppression procedure is realized on the basis of the elaborated mathematical model. The elaborated technique has been tested in a clinic. The testing has confirmed its efficiency. The work has been implemented under the support of the Russian Foundation of Basic Research (RFBR), project № 08-01-12011-ofi.

Paper Nr: 10
Title:

Shape and Semantics for 3D Anatomical Structure Retrieval

Authors:

Davide Moroni, Mario Salvetti and Ovidio SSalvetti

Abstract: In this paper, we propose a framework for the description of anatomical structures based on topological and geometrical features and on semantic annotation. The main goal is to enhance the description of an anatomical structure –understood as a 3D model– with other non-geometrical pieces of information relevant to the particular problem-context. Hybrid methods for similarity searches are then introduced and shown to be able to support effective case-based reasoning procedures. The approach is illustrated with examples from several medical application fields in order to discuss its potential impact.

Paper Nr: 11
Title:

Ontology-based Framework to Image Mining

Authors:

Sara Colantonio, I. Gurevich, Gabriele Pieri, Ovidio Salvetti and Yulia Trusova

Abstract: A novel knowledge-based approach for supporting image processing and analysis is presented as well as its use within a framework for image mining. Modern approaches to knowledge representation, ontologies and reasoning, have been combined with techniques for image processing, analysis and understanding within a semantic framework able to support the extraction of novel knowledge for image collections.

Paper Nr: 12
Title:

Image Representation: From Raw Data to Models

Authors:

I. Gurevich and Vera Yashina

Abstract: A space of image representations used in Descriptive Theory for Image Analysis is considered. The main types of image representations are introduced in accordance with image and its reducing to a recognizable form.

Paper Nr: 13
Title:

Prolegomena toward Algebraic Image Analysis

Authors:

I. Gurevich and Vera Yashina

Abstract: The paper is an extended abstract of analytical tutorial devoted to algebraization of image analysis

Paper Nr: 13
Title:

Prolegomena toward Algebraic Image Analysis

Authors:

I. Gurevich and Vera Yashina

Abstract: The paper is an extended abstract of analytical tutorial devoted to algebraization of image analysis

Paper Nr: 14
Title:

Skeleton-Based Shape Segmentation

Authors:

Liudmila Domakhina

Abstract: Shape decomposition into meaningful parts (segmentation) problem is considered in this paper. The shape is given either as a raster object on a homogeneous background or as a polygonal figure. A new shape decomposition approach is called skeleton-based segmentation. The approach is based on continuous skeletons that provides an opportunity to construct visually proper segmentations reflecting the shape structure. The proposed skeleton-based segmentation method stands out against known methods because it is suitable to work correctly with pairs of shapes. For a pair of shapes it is proposed to construct isomorphic skeleton-based segmentations which can be used on shape comparison and morphing applications.

Paper Nr: 16
Title:

Data Improvement using Form Factors Reconstructed in Latent Dimensions

Authors:

Alexander Vinogradov and Yury Laptin

Abstract: A new approach to the problem of enhanced description of cluster boundaries in the sample is developed. New density estimates calculated on the base of nonlinear reconstruction of form factors in latent dimensions are used for resolving critical loci in empirical distributions.

Paper Nr: 19
Title:

Method for Image Transform Selection in Cytological Image Analysis

Authors:

I. Koryabkina and I. Gurevich

Abstract: The paper considers diagnostic analysis of blood system tumours using special methods. Initial data are images of specimens from patients with three diagnoses, including two types of aggressive lymphoid tumours, and an innocent tumour. Analysing feature set, it was found that significant features vary for different diagnoses. Thus the task requires special methods for image analysis and recognition, i.e. methods that allow selecting image transformation depending on informational image nature. The paper shows that applying special methods, the recognition rate can be increased appreciably.

Paper Nr: 21
Title:

A Method for Automatic Extraction of Dopaminergic Neuron Terminals on Striatum Frontal Section Images

Authors:

I. Gurevich, I. Koryabkina, E. Kozina, A. Myagkov, H. Niemann, M. Ugrumov and V. Yashina

Abstract: This work is devoted to the description of an experimental data acquisition automated method which is required to fill the model of Parkinson’s disease preclinical stage. Digital images of the immunostained brain sections of experimental animals are used as a data source. Proposed method: 1) is based on following mathematical morphology operations: opening, grayscale reconstruction, closing, bot-hat transformation, morphological gradient, watershed transformation; 2) enables: to smooth heterogeneous complex background; to select small objects on images depended on given sizes and gray values; to eliminate out-of-focus objects; to separate close objects; to calculate features of selected objects; 3) is intended for automatic extraction of dopaminergic neurons terminals on striatum frontal section images. Experimental investigations confirmed possibility and suitability of section images automated processing and analysis by means of the method. The results of the method use are segmented object contours binary image and object feature list.

Paper Nr: 22
Title:

A Contribution to an Image Mining Oriented Geoprocessing

Authors:

Renato Guadagnin, Levy Santana, Edilson Ferneda and Hércules Prado

Abstract: Since its origin Geoprocessing Information Systems (GIS) are supposed to deal with structured information concerned some geographical localization. So one uses three-dimensional image representation systems in a huge database, where it is possible to insert many data about some interest domain, say, agriculture, economics, industry, demographics and so on. This article presents a new approach, which allows an integration of Geoprocessing and Image Mining not only in typical geographical subjects but also in other domains such as healthcare.

Paper Nr: 25
Title:

Adaptive Committees of Feature-specific Classifiers for Image Classification

Authors:

Tiziano Fagni, Fabrizio Falchi and Fabrizio Sebastiani

Abstract: We present a system for image classification based on an adaptive committee of five classifiers, each specialized on classifying images based on a single MPEG-7 feature. We test four different ways to set up such a committee, and obtain important accuracy improvements with respect to a baseline in which a single classifier, working an all five features at the same time, is employed.