Vassili Kovalev | National Institute of Informatics (original) (raw)
Uploads
Papers by Vassili Kovalev
The paper presents image description and classification methods which were used by United Institu... more The paper presents image description and classification methods which were used by United Institute of Informatics Problems (UIIP) group for tuberculosis image classification task. A method based on cooccurrence of adjacent supervoxels in 3D computed tomography (CT) images was used for subtask #1 which was dedicated to image-based recognition of multi-drug resistant tuberculosis. For subtask #2 which is dedicated to automated categorization of tuberculosis patients into one of five types of tuberculosis, extended multidimensional multi-sort co-occurrence matrices were used for describing the CT scans. Both two submitted runs were ranked 7th in both subtasks.
With this study, a method for quantitative description of biomedical images based on splitting th... more With this study, a method for quantitative description of biomedical images based on splitting the target image into superpixels followed by categorization using pre-calculated superpixel dictionaries and calculation of co-occurrence matrices is proposed. The method has been tested on the classification of biomedical images of three types: lung CT images, histology images of ovary and thyroid tissues.
Lecture Notes in Computer Science, 2011
Lecture Notes in Computer Science, 1997
ABSTRACT
Lecture Notes in Computer Science, 1997
This paper describes an image registration method for deformable objects like objects in medical ... more This paper describes an image registration method for deformable objects like objects in medical images, in face or fingerprint recognitions tasks, etc. The method uses a global optimisation approach where a cost function is defined, capable of incorporating all available data and knowledge concerning the task. The method is demonstrated within the context of face recognition.
Lecture Notes in Computer Science, 2006
Lecture Notes in Computer Science, 2005
ABSTRACT
Medical Imaging 1994: Image Processing, 1994
Lecture Notes in Computer Science, 1995
Rule-based method is considered for .recognition of arbitrary 64• 64 pixel regions selected in li... more Rule-based method is considered for .recognition of arbitrary 64• 64 pixel regions selected in liver ultrasound images. Recognition rules are based on parameters describing spatial distribution of different gradient levels and anisotropy of liver texture. High recognition accuracy has been obtained in case of the same image acquisition conditions,
2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06), 2000
Feature space reduction was performed using the Principal Component Analysis and Independent Comp... more Feature space reduction was performed using the Principal Component Analysis and Independent Component Analysis methods. Six classification methods were examined including unsupervised clustering algorithms such as K-means, Hard Competitive Learning, and Neural Gas as well as Hierarchical Clustering, Support Vector Machines, and Random Forests classifiers. Detailed results on the cell image classification accuracy and computational efficiency achieved using different feature sets and different classification methods are reported.
Proceedings of the 17th International Conference on Pattern Recognition 2004 Icpr 2004, Aug 23, 2004
About 8% of men (but not women) are suffering from color blindness. The objective of this work wa... more About 8% of men (but not women) are suffering from color blindness. The objective of this work was to investigate the problem of image retrieval based on color cooccurrence features when comparing normal vision and three kinds of color blindness (dichromasia): protanopia, deuteranopia, and tritanopia. Original database comprises 12000 images that were also converted into three dichromatic versions using the Vischeck simulation tool. Results of 48000 queries were used to study influence of color blindness on retrieval results. Principal Component Analysis, Multidimensional Scaling, Hierarchical Clustering, Support Vector Machines, and statistical methods were employed for investigating feature space distortions associated with color blindness.
Proceedings of the 5th International Conference on Computer Analysis of Images and Patterns, Sep 13, 1993
2](https://attachments.academia-assets.com/61249889/thumbnails/1.jpg)
](Russian Journal of Coordination Chemistry, Sep 1, 2009
The complex [ Co (2-Me -Pyz ) 2 ( H 2 O ) 4 ]( NO 3 ) 2 is synthesized and its structure is deter... more The complex [ Co (2-Me -Pyz ) 2 ( H 2 O ) 4 ]( NO 3 ) 2 is synthesized and its structure is determined. The crystals are monoclinic: space group P 2 1 / n , a = 10.685(2) Å, b = 6.837(1), c = 12.515(3) Å , β = 91.84(3)°, V = 913.8(3) Å 3 , ρ calcd = 1.042 g/cm 3 , Z = 2. The Co 2+ ion (in the inversion center) is coordinated at the vertices of the distorted octahedron by two nitrogen atoms of methylpyrazine and four oxygen atoms of the water molecules (Co(1)-N(1) 2.180(3), average Co(1)-O( w ) 2.079(3) Å, angles at the Co atom 87.9(1)-92.1(1)° ). Supramolecular pseudometallocycles are formed in the structure through the O( w )-H ···N (1) hydrogen bonds between the coordinated H 2 O molecules and the terminal nitrogen atoms of the 2-methylpyrazine molecules. Their interaction results in the formation of supramolecular layers joined by the N O 3 groups into a three-dimensional framework.
Tmi, 2001
A method is proposed for three-dimensional (3-D) texture analysis of magnetic resonance imaging b... more A method is proposed for three-dimensional (3-D) texture analysis of magnetic resonance imaging brain datasets. It is based on extended, multisort co-occurrence matrices that employ intensity, gradient and anisotropy image features in a uniform way. Basic properties of matrices as well as their sensitivity and dependence on spatial image scaling are evaluated. The ability of the suggested 3-D texture descriptors is demonstrated on nontrivial classification tasks for pathologic findings in brain datasets.
Eur J Nucl Med Mol Imaging, 2004
A new approach for improving diagnostic accuracy in Alzheimer's disease and frontal lobe dementia... more A new approach for improving diagnostic accuracy in Alzheimer's disease and frontal lobe dementia utilising the intrinsic properties of the SPET dataset Authors:
Graphical Models and Image Processing, 1996
Proceedings of 13th International Conference on Pattern Recognition, 1996
ABSTRACT The objective of this work is to investigate the white blood cell (WBC) image recognitio... more ABSTRACT The objective of this work is to investigate the white blood cell (WBC) image recognition problem at all stages. A robust and effective method for automatic WBC differentiation, based on both statistical pattern recognition and neural net approaches, is presented. We demonstrate well-evaluated results ranging from image scene segmentation techniques to recognition details. Recognition accuracy on the test set of 662 images of five WBC types obtained by different imaging systems from 22 bloodstains is not less than 98%
Cit Journal of Computing and Information Technology, 1998
The paper presents image description and classification methods which were used by United Institu... more The paper presents image description and classification methods which were used by United Institute of Informatics Problems (UIIP) group for tuberculosis image classification task. A method based on cooccurrence of adjacent supervoxels in 3D computed tomography (CT) images was used for subtask #1 which was dedicated to image-based recognition of multi-drug resistant tuberculosis. For subtask #2 which is dedicated to automated categorization of tuberculosis patients into one of five types of tuberculosis, extended multidimensional multi-sort co-occurrence matrices were used for describing the CT scans. Both two submitted runs were ranked 7th in both subtasks.
With this study, a method for quantitative description of biomedical images based on splitting th... more With this study, a method for quantitative description of biomedical images based on splitting the target image into superpixels followed by categorization using pre-calculated superpixel dictionaries and calculation of co-occurrence matrices is proposed. The method has been tested on the classification of biomedical images of three types: lung CT images, histology images of ovary and thyroid tissues.
Lecture Notes in Computer Science, 2011
Lecture Notes in Computer Science, 1997
ABSTRACT
Lecture Notes in Computer Science, 1997
This paper describes an image registration method for deformable objects like objects in medical ... more This paper describes an image registration method for deformable objects like objects in medical images, in face or fingerprint recognitions tasks, etc. The method uses a global optimisation approach where a cost function is defined, capable of incorporating all available data and knowledge concerning the task. The method is demonstrated within the context of face recognition.
Lecture Notes in Computer Science, 2006
Lecture Notes in Computer Science, 2005
ABSTRACT
Medical Imaging 1994: Image Processing, 1994
Lecture Notes in Computer Science, 1995
Rule-based method is considered for .recognition of arbitrary 64• 64 pixel regions selected in li... more Rule-based method is considered for .recognition of arbitrary 64• 64 pixel regions selected in liver ultrasound images. Recognition rules are based on parameters describing spatial distribution of different gradient levels and anisotropy of liver texture. High recognition accuracy has been obtained in case of the same image acquisition conditions,
2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06), 2000
Feature space reduction was performed using the Principal Component Analysis and Independent Comp... more Feature space reduction was performed using the Principal Component Analysis and Independent Component Analysis methods. Six classification methods were examined including unsupervised clustering algorithms such as K-means, Hard Competitive Learning, and Neural Gas as well as Hierarchical Clustering, Support Vector Machines, and Random Forests classifiers. Detailed results on the cell image classification accuracy and computational efficiency achieved using different feature sets and different classification methods are reported.
Proceedings of the 17th International Conference on Pattern Recognition 2004 Icpr 2004, Aug 23, 2004
About 8% of men (but not women) are suffering from color blindness. The objective of this work wa... more About 8% of men (but not women) are suffering from color blindness. The objective of this work was to investigate the problem of image retrieval based on color cooccurrence features when comparing normal vision and three kinds of color blindness (dichromasia): protanopia, deuteranopia, and tritanopia. Original database comprises 12000 images that were also converted into three dichromatic versions using the Vischeck simulation tool. Results of 48000 queries were used to study influence of color blindness on retrieval results. Principal Component Analysis, Multidimensional Scaling, Hierarchical Clustering, Support Vector Machines, and statistical methods were employed for investigating feature space distortions associated with color blindness.
Proceedings of the 5th International Conference on Computer Analysis of Images and Patterns, Sep 13, 1993
Russian Journal of Coordination Chemistry, Sep 1, 2009
The complex [ Co (2-Me -Pyz ) 2 ( H 2 O ) 4 ]( NO 3 ) 2 is synthesized and its structure is deter... more The complex [ Co (2-Me -Pyz ) 2 ( H 2 O ) 4 ]( NO 3 ) 2 is synthesized and its structure is determined. The crystals are monoclinic: space group P 2 1 / n , a = 10.685(2) Å, b = 6.837(1), c = 12.515(3) Å , β = 91.84(3)°, V = 913.8(3) Å 3 , ρ calcd = 1.042 g/cm 3 , Z = 2. The Co 2+ ion (in the inversion center) is coordinated at the vertices of the distorted octahedron by two nitrogen atoms of methylpyrazine and four oxygen atoms of the water molecules (Co(1)-N(1) 2.180(3), average Co(1)-O( w ) 2.079(3) Å, angles at the Co atom 87.9(1)-92.1(1)° ). Supramolecular pseudometallocycles are formed in the structure through the O( w )-H ···N (1) hydrogen bonds between the coordinated H 2 O molecules and the terminal nitrogen atoms of the 2-methylpyrazine molecules. Their interaction results in the formation of supramolecular layers joined by the N O 3 groups into a three-dimensional framework.
Tmi, 2001
A method is proposed for three-dimensional (3-D) texture analysis of magnetic resonance imaging b... more A method is proposed for three-dimensional (3-D) texture analysis of magnetic resonance imaging brain datasets. It is based on extended, multisort co-occurrence matrices that employ intensity, gradient and anisotropy image features in a uniform way. Basic properties of matrices as well as their sensitivity and dependence on spatial image scaling are evaluated. The ability of the suggested 3-D texture descriptors is demonstrated on nontrivial classification tasks for pathologic findings in brain datasets.
Eur J Nucl Med Mol Imaging, 2004
A new approach for improving diagnostic accuracy in Alzheimer's disease and frontal lobe dementia... more A new approach for improving diagnostic accuracy in Alzheimer's disease and frontal lobe dementia utilising the intrinsic properties of the SPET dataset Authors:
Graphical Models and Image Processing, 1996
Proceedings of 13th International Conference on Pattern Recognition, 1996
ABSTRACT The objective of this work is to investigate the white blood cell (WBC) image recognitio... more ABSTRACT The objective of this work is to investigate the white blood cell (WBC) image recognition problem at all stages. A robust and effective method for automatic WBC differentiation, based on both statistical pattern recognition and neural net approaches, is presented. We demonstrate well-evaluated results ranging from image scene segmentation techniques to recognition details. Recognition accuracy on the test set of 662 images of five WBC types obtained by different imaging systems from 22 bloodstains is not less than 98%
Cit Journal of Computing and Information Technology, 1998