Juliusz L. Kulikowski

The Role of Ontological Models in Pattern Recognition

There are considered the role and applications of ontological models in advanced pattern recognition methods. Formal definition of ontological models, a general taxonomy, and specification of some typical ontological models are presented. Examples of a simple, a composite and an extended ontological model are given. The role of ontological models in composite patterns recognition is described and illustrated by examples.

Morphological Spectra as Tools for Texture Analysis

There are presented formal properties of morphological spectra as a novel tool for analysis of textures. It is described a multi-level structure of the system of morphological spectra and the method of calculation of spectral components. Formal properties of morphological spectra: symmetries, ability to describe parallel shifts and rotations of analyzed images are presented. Several comments concerning practical aspects of using morphological spectra to analysis of textures are also given.

Biomedical Structures Representation by Morphological Spectra

It is considered the problem of representation of irregular structures, typical in biomedical images, by morphological spectra. Some basic properties of morphological spectra have been reminded. Then several possible methods of morphological spectra presentation are illustrated by a numerical example. The problem of representation of selected classes of irregular structures is considered and illustrated by the examples of vertically elongated, compact and branching structures representation. It is shown that for each class of irregular structures a hierarchy of spectral components indicating their role in representation of the class can be statistically established.

Pattern recognition based on similarity in linear semi-ordered spaces

In the paper an approach to pattern recognition based on a notion of similarity in linear semi-ordered (Kantorovitsch) space is presented. It is compared with other approaches based on the metric distance and on angular dilation measures in observation spaces. Basic assumptions of the Kantorovitsch space are shortly presented. It is shown that finite reference sets for pattern recognition take on in Kantorovitsch space formal structures presented by connectivity graphs which facilitate finding the reference vectors for pattern recognition.

A Comparative Analysis of SPECT Images of the Left and Right Cerebral Hemispheres in Patients with Diagnosed Epileptic Symptoms

The aim of his work was examination of asymmetries in activity of the left and right cerebral hemispheres as well as localization and contouring of the regions of reduced or increased activity on the basis of single photon emission computer tomography (SPECT) images. Advantage of this technique lies in possibility of brain activity map acquisition at the time of radiotracer injection during seizures though the image registration is done one hour after seizure. The SPECT imaging method makes possible a more accurate spatial localization of seizure source than analysis of EEG signals. Simultaneous EEG signal registration allows to qualify exactly the moment of seizure onset when radiotracer injection could be done to register an unequivocal image. The mean and standard deviation of normalized intensities inside the contoured areas of images were calculated.

Biomedical Image Segmentation Based on Aggregated Morphological Spectra

It is described a method of biomedical images segmentation by discrimination of textures based on their morphological spectra. Basic notions concerning morphological spectra are given. Their properties making possible to characterize basic morphological structures independently on spatial orientation or shifts of the analyzed specimens are described. It is shown that spectral components can be chosen and used in aggregated form so as to make discrimination of textures invariant with respect to scale changing and to basic geometrical image transformations. Analysis of two types of biomedical images: aorta tissue and pancreas tissue, based on comparison of histograms of selected spectral components values illustrate the methods presented in the paper.

Direct Filtering and Enhancement of Biomedical Images Based on Morphological Spectra

In the paper a method of filtering of biomedical images aimed at their enhancement for direct visual examination or for automatic segmentation of regions covered by typical textures is presented. For this purpose morphological spectra (being a modification of the systems of orthogonal 2D Walsh functions) are used. Filtering consists in assigning relative weights coefficients to spectral components representing typical morphological micro-structures. However, direct filtering makes possible elimination of calculation of the components of morphological spectra, because filtered values of image elements are given as linear combinations of the values of the original image in fixed basic windows. The method of calculation of the transformation coefficients in details is described. Application of the method is illustrated by an example of cerebral SPECT image examination.

Structural Image Analysis Based on Ontological Models

A concept of structural image analysis and interpretation based on superrelations and on ontological models is presented. The system of image interpretation should contain the structural analysis (SA), ontological models (OM) and semantic relationships (SR) modules. The role of modules is described. The proposed approach is illustrated by an example of cardiac USG images interpretation.

Morphologic-Statistical Approach to Detection of Lesions in Liver Tissue in Fish

The problem of light microscope images enhancement by filtering for recognition pathologic liver tissues in fish is considered in the paper. The problem follows from the necessity of monitoring the sea water pollutions caused by mercury compounds and their influence on living organisms. It is proposed to use image filtering based on morphological spectra to enhance visibility of liver lesions in the images in order to extract morphologic-statistical parameters useful in automatic tissues classification into normal and pathologic classes. It is shown that selected components of the 4th range morphologic spectra (MS4) are the most suitable to discriminate normal and pathologic liver tissues. The selected spectral components are characterized by their estimated mean values, standard deviations and kurtoses. The so-obtained morphologic-statistical parameters have been used to construct the learning sets for two types of image classifiers: based on the nearest mean and k nearest neighbors rules. It is shown that preliminary image filtering by morphological spectra-based filters improves spatial distribution of the recognized normal and pathologic objects in the parameter space.

Object Recognition Based on Comparative Similarity Assessment

In the paper a concept of object recognition based on their similarity assessment in case of nonhomogenous qualitative and quantitative objects’ features is presented. Moreover, it is assumed that the features’ intensity values are not given directly but by their pairwise comparative assessment. This corresponds to an intuitive, on human experience-based assessment of the objects’ properties. The proposed object recognition method is based on reference sets divided into credibility layers, according to a relative logical model and conceptual classes of similarity. This concept is illustrated by an example of a conceptual class of “irregular” objects, the “irregularity” being intuitively assessed. The method is presented in the form of an algorithm.