Vol. 16 (2007):
  Abstracts of Papers

15 (2006)	main	17 (2008)

No. 1/2: Special Issue on Creativity and Visual Design Computing,
No. 3/4: Special Issue on Novel Methods in Computer-Aided Biomedical Image Processing and Interpretation. Part I.

Machine GRAPHICS & VISION, Vol. 16 (2007), No. 1/2:

Grabska E.:
Guest Editorial: Creativity and Visual Design Computing.
MGV vol. 16, no. 1/2, 2007, pp. 3-4.

Borkowski A., Grabska E., Ozimek A., Ozimek P., Papiernik K.:
Supporting conceptual design in architecture by linguistic model.
MGV vol. 16, no. 1/2, 2007, pp. 5-22.

It is shown that linguistic approach can be very useful when developing software supporting conceptual design in architecture. Treating primitives, like rooms, as letters of a certain language and capturing the knowledge about the architecture in a grammar allows us to implement automatic reasoning. It turns out that grammars describing hierarchical graphs are best suited for this purpose. The formalism underlying this approach is presented together with a prototype software that allows the user to convert functional requirements into layouts of the designed building, to visualise the resulting alternative solutions and to select the best one for the detailed design. The usage of this software is demonstrated on two examples.
Key words: hierarchical graph, realisation scheme, conceptual design.

Grabska E., Lachwa A., Slusarczyk G., Grzesiak-Kopec K., Lembas J.:
Hierarchical layout hypergraph operations and diagrammatic reasoning.
MGV vol. 16, no. 1/2, 2007, pp. 23-38.

This paper deals with a new, computer-aided approach to floor-layout design. The approach proposes: firstly, a specific layout language with a syntactic knowledge defined by means of hierarchical hypergraphs; secondly, a correspondence between layout modifications and hypergraph operations. An initial layout drawn by the designer is automatically converted into a hypergraph and each designer's modification to the layout is reflected in the hypergraph structure. Our new approach is illustrated by a step by step example, where two complementary representations of the same floor-layout design are used.
Key words: floor-layout, innovative design, hypergraph, diagrammatic reasoning.

Bielecka M.:
Syntactic segmentation of function graph type of curves.
MGV vol. 16, no. 1/2, 2007, pp. 39-55.

In this paper structural analysis of patterns being a special type of curves is presented. The recognized objects are segmented according to their structural properties into primitive patterns representing generic shapes. Then, an algebraic formula describing the structure of the analyzed curve is created. The description is unique.
Key words: curve description, shape analysis, structural description, curve segmentation.

Cowell J., Hussain F.:
A syntactic recognizer for Arabic characters.
MGV vol. 16, no. 1/2, 2007, pp. 57-83.

Commercial OCR systems for high quality Latin characters are widely available. However, the recognition of Arabic characters, particularly those of poor quality, is still an important research area. This paper gives details of two recognition systems for isolated Arabic characters based on the extraction of critical features. The limitations of this approach and its successful extension to a syntactic recognition system are discussed. The syntactic system uses a new web grammar, in which the critical features are nodes and a set of operators describes the spatial relationships between them. The approach used is insensitive to variations in the size and orientation of the characters and even the relative lengths of strokes. The described system has been implemented and the theoretical behaviour of this approach is supported by the experimental results.
Key words: syntactic pattern recognition, web grammars, OCR, Arabic characters.

Slusarczyk G.:
Computer animation in creative design.
MGV vol. 16, no. 1/2, 2007, pp. 85-104.

The paper proposes a new approach to graphic design with the use of computer animation. The class of designs considered are periodic patterns. First, the characterization of rosettes, borders and periodic planar designs is given. Then, a generative system which produces a great variety of periodic designs is equipped with animation. As in-between animation frames ensure the continuity and fluidity of pattern movement, a dynamic continuous design space is obtained. The changes in pattern symmetry groups which result from the use of animation are described. The role of animation in supporting perception of the emergent shapes in the generated patterns is also mentioned. The paper is illustrated by periodic designs produced by the DARTAN animated generative system.
Key words: graphic design, periodic patterns, generative system, computer animation.

Goinski A., Nikiel S.:
The motion of impostors.
MGV vol. 16, no. 1/2, 2007, pp. 105-116.

We describe a method that brings to life impostor-based environments. In a typical scene, supporting objects are rendered as two-dimensional texture maps always facing the camera. Billboards reduce complexity of objects to high extent. In such a representation, spatial properties of depicted objects are lost. Billboards are usually motionless in order to compress the video memory space. In our technique, we introduce 2.5D morphing with respect to the memory footprint. Minimum two textures are required to animate the billboard. Moreover, the whole process is automated and exploits a programmable GPU. As a result, the main application overhead is reduced. The method is designed for vegetation modeling, but can be easily extended to far- and middle-distance shots of humans.
Key words: 3D graphics and realism, morphing, surface deformation, image-based rendering, impostors, animation, billboards.

Janaszewski M., Kacki E.:
Feature generation from digital images using pseudo-fractal algorithm and its four modifications.
MGV vol. 16, no. 1/2, 2007, pp. 117-138.

The main aim of the paper is to present the authors' original method of feature generation from digital images and to report on a comparison of five various algorithms, which implemented that method. The algorithms are based on an idea by the same authors', which consists in producing a quantitative description of similarity intensity between various parts of an image in various scales. To develop it the algorithms take advantage of fractal coding based on an Iterated Function System. Therefore, the generated features can rightly be called similarity features. In this paper we show that similarity features, when combined with other well known ones, can improve recognition results in some image classification tasks. After presenting how the algorithm works, we compare their properties and report the classification results obtained in two different pattern recognition experiments. Moreover, the paper contains a discussion of the obtained results, and of possible future applications of the similarity features.
Key words: pattern recognition, feature generation, texture analysis, self-affinity.

Wang K., Zheng N., Qi H.:
Least square image matting.
MGV vol. 16, no. 1/2, 2007, pp. 139-152.

This paper addresses the well-known problem of natural image matting. Most of the previous matting algorithms require the user to define the tri-map, which is an inconvenient work and sometimes a burden, especially in a complex situation. This paper uses ceratain user defined foreground and background strokes to estimate the image matte. First we use a Gauss Markov random field to model the matting problem. Then we use the least square optimization approach to solve it. Experimental results show that our approach could properly handle confused boundaries. It also could deal with semi-transparent conditions such as fire etc.
Key words: matting, image processing, least square.

Ali Md. H., Rahman I.S., Islam M., Shahiduzzaman M.:
Mathematical morphology based automated control point detection from human facial image.
MGV vol. 16, no. 1/2, 2007, pp. 153-170.

The ultimate goal of this research is to incorporate facial animation based on image morphing in a very narrow bandwidth video transmission, especially in video conferencing, news telecast etc., where the background as well as the object in the image change little. As a part of the whole work, in this paper, an efficient mathematical morphology-based facial feature control point detection technique is proposed. By facial feature control point we mean facial feature (i.e. eye, lip etc.) surrounding points and other important points in the face which can be utilized to create facial animation based on image metamorphosis. In the experiment, mathematical morphology tools are used both for filtering and pattern matching. At first, intensity-independent, color-based segmentation is used with some morphological processing on the input image to separate skin regions. Then, the parallel eye segments are searched by erosion of the edge-thinned image with eye corner structuring elements. By combining them, the probable eye segment pair is identified. Then, using facial structural knowledge, the lips and other control points are detected. The accuracy of the proposed method is within quite acceptable limits; moreover, the method is capable of working with images of average quality or close to average quality.
Key words: facial feature detection, mathematical morphology.

Bielecki A., Strug B.:
Finding an iterated function systems based representation for complex visual structures using an evolutionary algorithm.
MGV vol. 16, no. 1/2, 2007, pp. 171-189.

This paper presents an approach to the IFS inverse problem based on evolutionary computations. Having a bitmap image, we look for a set of functions that can reproduce a good approximation of a given image. A method using a variable number of mappings is proposed. A number of different crossover operators is described and tested. The possibility of enriching evolutionary algorithms by a specific type mechanism characteristic for replication of influenza viruses is discussed. The genetic material of the influenza type A virus consists of eight separate segments. In some types of tasks, such a structure of a genome can be more adequate than representation that consists of one sequence only. If influenza virus strains infect the same cell, then their RNA segments can mix freely, producing progeny viruses which represents the reasortment mechanism. Furthermore, mistakes leading to new mutations are common. The structure of problems for which such viral reproduction mechanisms can be effective are analyzed. The paper ends with some experimental results showing the images we were able to generate with the proposed method. The preliminary experimental results suggest that the introduction of the reasortment operator results in achieving satisfactory images in a smaller number of generations.
Key words: hierarchical evolutionary algorithms, image generation, iterated function systems inverse problem, viral replication, reasortment.

Rataj A.:
Generalization of raster images containing patterns featuring stochastic repetitiveness.
MGV vol. 16, no. 1/2, 2007, pp. 191-201.

The issue of reconstruction of missing or unreliable parts of an image is one of the basic problems in image processing. For example, there are a number of methods for texture generation on the basis of a small sample. This paper presents a method that `bottlenecks' an image processing feedforward neural network so that only some basic traits of the image are preserved. These basic traits are in turn used to generalize the image, thus filtering out any unusual parts of the image.
The ability of neural networks and several other learning machines to generalize is based on the premise of smoothness of the generalizing function. Thus, in order to detect advanced patterns that exhibit complex traits like repetitiveness, instead of training these machines directly with raw data, transforms of the patterns like the Fast Fourier Transform are sometimes performed. In this paper it is shown, that a simple feedforward neural network, without any pre--processing of the training data, using the described `bottleneck' architecture, can properly predict a stochastically repetitive pattern in a raster image.
Key words: generalization, feedforward neural networks, nonlinear regression.

Machine GRAPHICS & VISION, Vol. 16 (2007), No. 3/4:

Kulikowski J.L.:
Guest Editorial: Novel Methods in Computer-Aided Biomedical Image Processing and Interpretation.
MGV vol. 16, no. 3/4, 2007, pp. 205-206.

Drapikowski P., Czwojdzinski A.:
Geometrical and morphological validation of medical parameters measurements based on 3D surface models .
MGV vol. 16, no. 3/4, 2007, pp. 207-219.

Medical diagnostics necessitates performing quantitative analysis and measurements of 2D or 3D data. The length, angle, region area, 3D surface area, and volume are measured to determine medical parameters. This paper presents an uncertainty estimation for a 3D surface model created from object boundaries using CT, MRI series of images. Next, thirty dry bone pelvises underwent the morphological, classical radiological and CT tests, and were reconstructed in 3D. Then, the obtained results for selected parameters describing the pelvis and the orientation of coxal acetabulum were compared. Using dray human pelvises for validation study is extremely important to convince physicians that measurements results based on virtual models are comparable to the same results obtained in the classical way.
Key words: medical diagnostics, 3D surface models, validation study.

Bialasiewicz J.T., Redmond L.S.:
Medical image compression and analysis using wavelet modulus maxima decomposition.
MGV vol. 16, no. 3/4, 2007, pp. 221-235.

A method of highly effective biomedical image compression that includes the reconstruction process with a good convergence rate is presented in the paper. It represents an image in the form of its wavelet modulus maxima decomposition. The technique allows the compressed image representation to include only those wavelet transform coefficients that correspond to the wavelet transform modulus maxima that are determined for each resolution level.
The proposed approach to analysis of medical images uses the wavelet modulus maxima decomposition to enhance image features that are not visually apparent. The transient behavior of pixel intensities (that corresponds to edges and singular points) is used for image enhancement. The detection of edges is realized by detecting modulus maxima in a two-dimensional dyadic wavelet transform at the proper scale. This approach to image analysis aims at determining structures of the diseased tissue that are represented by the image edges. It is expected that this technique will help with early detection of cancer when routine interpretation of CT scans is inconclusive and biopsy would be required.
Key words: medical image processing, image compression, wavelet modulus maxima, edge detection, reconstruction from edges.

Sarker M.H., Sloane A.:
TGSF / TLoG filter with optical flow technique for large motion detection.
MGV vol. 16, no. 3/4, 2007, pp. 237-249.

In this paper, detection and segmentation of large motion in moving image sequences is presented. For detecting motion, the intensity of each pixel is convolved with the second derivative of the Temporal Gaussian Smoothing Function (TGSF) or the Temporal Laplacian of Gaussian (TLoG) filter. The zero-crossing in a single frame of the resulting function indicates the positions of moving edges. An intensity change over time due to a small illumination effect does not produce a zero crossing. Therefore, such changes are not interpreted as human motion by this method. The optical flow velocity is computed by using the spatial and temporal derivatives of this function, and it is normal to the zero crossing contours. Pixels belonging to the normal velocities are projected back to the original color image sequences to achieve a segmented color image. Experiments show that a moving object is detected correctly, and good segmentation results are achieved.
Key words: motion detection , TLoG filter, zero crossing, motion segmentation.

Koprowski R., Wojaczynska-Stanek K., Wrobel Z.:
Automatic segmentation of characteristic areas of the human head on thermographic images.
MGV vol. 16, no. 3/4, 2007, pp. 251-274.

In the paper a new algorithm for the automatic segmentation of characteristic areas of the human head in a thermovision image is presented. The analysed regions include: forehead, area of sinuses and back of the neck.
The covered set of approaches, which are mainly algorithms employing morphological operations, neural networks and methods based on human head proportions, can provide satisfying completely results for a doctor's practice in a fully automatic way.
Key words: head, eye, nose, segmentation, morphology, neural networks, headache, thermovision.

Przytulska M.:
Analysis of left cardiac ventricle shape variations based on examination of ultrasound images.
MGV vol. 16, no. 3/4, 2007, pp. 275-292.

The changes in the shape of the left cardiac ventricle within full heart evolution cycle in norm and in pathology are described using a kinetic model programmed and verified on real clinical data. In the model, information is represented by the so-called shape coefficients. The time series of the shape coefficients as well as of differential and normalized shape coefficients of the left heart ventricle have been obtained experimentally. They are then analyzed using three different approaches: spectral analysis, direct analysis of shape coefficients and direct visualization of contractions. In each of those cases, the aim of the analysis is detection and recognition of various types of abnormalities, as well as evaluation of the distance between the detected pathology and the norm.
Key words: cardiac imaging, computer-aided image processing, left cardiac ventricle's contractility, cardiac walls akinesis, cardiac walls hypokinesis, cardiac walls diskinesis, spectral methods.

Nedzved A., Zalesky B., Ablameyko S., Drozd V., Fridman M.:
Joint Analysis of histological and ultrasonic images to lean state of thyroid gland.
MGV vol. 16, no. 3/4, 2007, pp. 293-304.

Simultaneous analysis of histological and ultrasonic (US) images of human thyroid glands for thyroid cancer diagnostics is proposed in the paper. It allows to explain the characteristics of US pictures of the thyroid gland via the sizes of its follicles.
To show the dependence of US image features on the state of follicles, statistical analysis of US-texture is performed. In addition, the size of follicles in histological images is calculated by analysis of a distance map for the nuclei of cells. It is shown that echogenicity of the thyroid gland in US images depends essentially on the size of its follicles. The organ regions that contain many follicles of a size smaller than the size of healthy follicles, or contain many destroyed follicles, have low echogenicity. The same effect is observed for regions with oversized follicles. This information can be used to avoid a surgical procedure, including histological analysis.
Key words: ultrasonic and histological images, thyroid gland, simultaneous analysis.

Yapa R.D., Harada K.:
A connected component labelling algorithm for greyscale mammography image processing as a pre-processing tool.
MGV vol. 16, no. 3/4, 2007, pp. 305-327.

A new algorithm for connected component-labelling is presented in this paper. The proposed algorithm requires only one scan through an image for labelling connected components. Once this algorithm encounters a starting pixel of a component, it traces in full all the contour pixels and all internal pixels of that particular component. The algorithm recognizes components of the image one at a time while scanning in the raster order. This property will be useful in areas such as image matching, image registration, content-based information retrieval and image segmentation. It is also capable of extracting the contour pixels of an image and storing them in a clock-wise directional order, which will provide useful information in many applications. The algorithm assigns consecutive label numbers to different components, and therefore requires a minimum number of labels. We have used the algorithm in mammography image processing as a pre-processing tool, and have demonstrated the possibility of using it for breast tissue segmentation and for detecting regions of interest in breast tissue. Another important advantage of the algorithm is that it can be used as a content-based image retrieval tool for retrieving images based on the visual contents of a given image. This would be very useful in retrieving related images from large scale medical databases.
Key words: connected components, blob detection, breast tissue segmentation, mammogram.

Bator M., Nieniewski M.:
Template matching by means of correlation coefficient for detecting cancerous masses in mammograms.
MGV vol. 16, no. 3/4, 2007, pp. 329-345.

The paper presents the authors' experiences with the detection of cancerous masses in mammograms. The described detection method is based on the use of multiscale template matching and multiresolution. As a measure of similarity, the correlation coefficient is adapted. The main conclusion drawn from the conducted experiments is that by sufficiently dense scaling of the templates one can achieve FROC (\emph{Free Response Operating Characteristics}) curves of the same quality as the curves obtained in the literature with considerably more sophisticated methods. The results were calculated for full mammograms of the entire MIAS database, in contrast to the literature, where the results are often given for regions of interest or for selected images. Several options for the templates were investigated, including three variants based on the hemispherical gray level distribution, as well as the optimal choice of the increasing scale of templates covering the whole range of diameters of masses.
Key words: cancerous mass detection, template matching, multiscale, multiresolution, mammography, MIAS.

Przelaskowski A., Podsiadly-Marczykowska T., Wroblewska A., Boninski P., Bargiel P.:
Computer-aided interpretation of medical images: mammography case study.
MGV vol. 16, no. 3/4, 2007, pp. 347-375.

This paper presents the current limitations and challenges of computer-aided interpretation of radiological examinations. The analysis and the proposed improvements in interpretation arose from our experience, knowledge and observations with the collected suggestions and conclusions. The emphasized topics are as follows: computer understanding of human determinants of diagnosis, characteristics and enhancement of observer performance, diagnostic accuracy measures of image examinations, computer-aided diagnosis (CAD) systems, and numerical description of medical image-based content. All of these diagnosis support concepts can be integrated into an intelligent diagnosis interface and enhanced, basing on a formal description of semantic image content, i.e. ontology implied as a reliable, dynamic platform of medical knowledge, useful for diagnosis. CAD for mammography and content--based image indexing supported by the ontology were integrated for the needs of an enhanced diagnostic workstation applied in tele-information medical systems. A design of an effective human-machine interface has arisen as the leading problem of the current challenges.
Key words: radiological interpretation, ontology, diagnostic accuracy, computer-aided diagnosis, content-based indexing.

Nomani P., Rahman W.R., Ali Md.H.:
An efficient facial expression detection system.
MGV vol. 16, no. 3/4, 2007, pp. 377-399.

In this paper we present an effective and robust approach for detecting a Bangladeshi facial expression in a 2D image. The facial expression is one of the most powerful, natural and immediate means for human beings to communicate their emotions and intentions. Most of the existing works on expression analysis have been focused on the facial expressions of European and American people, but these basic expressions vary subtly with races all over the world. We have tried to explore this diversity and to automate the facial expression detection for people of the Indian sub-continent, especially Bangladeshi people. Consequently, we have recommended a modified set of AUs (Action Units) defined in the FACS (the Facial Action Coding System), which is the leading standard for measuring facial expressions in the behavioral sciences. In this work, we propose a method to combine feature detection and extraction and facial expression detection into an integrated system. The main aspect of our system is that it covers all the criteria essential for detecting the facial expression of the people inhabiting sub-continent especially Bangladeshi people. Though the system concentrates on a particular race, it is also successful to a great extent with other races in the world, which proves its flexibility and robustness.
Key words: facial expression detection, Bangladeshi expressions.

Kukharev G., Forczmanski P.:
Facial images dimensionality reduction and recognition by means of 2DKLT .
MGV vol. 16, no. 3/4, 2007, pp. 401-425.

Paper presents an efficient dimensionality reduction method for images (e.g. human faces databases). It does not require any usual pre-processing stage (like down-scaling or filtering). Its main advantage is associated with efficient representation of images leading to accurate recognition. Analysis is performed using two-dimensional Principal Component Analysis and Linear Discriminant Analysis and reduction by means of two-dimensional Karhunen-Loeve Transform. The paper presents mathematical principles together with some results of recognition experiments on popular facial databases. The experiments performed on several facial image databases (BioID, ORL/AT&T, FERET, Face94 and Face95) showed that face recognition using this type of feature space dimensionality reduction is particularly convenient and efficient, giving high recognition performance.
Key words: image recognition, image compression, dimensionality reduction, Linear Discriminant Analysis.

Contest for the best Ph.D. thesis in the domain of image processing in 2006-2007

Reviewers' index

Authors' index

Contents of volume 16, 2007

15 (2006)	main	17 (2008)