Wyniki wyszukiwania - Biblioteka Nauki

1

Isolated word descriptors as control parameters of the computer applications

100%

Porwik P.

|

tom Vol. 10

35--46

EN

This paper is an extended version of the MIT'06 conference contribution. During the conference, many inquiries about the used techniques were performed. Hence, in the paper some parts of investigations were explained and discussed, with greater accuracy. It is shown that the computer applications can be controlled by a human voice. The computer controlling processes are available by means of utterance of isolated words, where application events with the aid of user's voice can be serviced. The voice usage can be convenient for blind or partially sighted users or for persons with limb paresis. The Microsoft application events, by means of the practicable Microsoft Windows firmware MSAA® technology can be analysed. Such technology, together with isolated word descriptors, as voice recognition system, has been presented.

2

Efficient Calculation of the Reed-Muller Form by Means of the Walsh Transform

100%

Porwik P.

|

2002

|

tom Vol. 12, no 4

571-579

EN

The paper describes a spectral method for combinational logic synthesis using the Walsh transform and the Reed-Muller form. A new algorithm is presented that allows us to obtain the mixed polarity Reed-Muller expansion of Boolean functions. The most popular minimisation (sub-minimisation) criterion of the Reed-Muller form is obtained by the exhaustive search of all the polarity vectors. This paper presents a non-exhaustive method for Reed-Muller expansions. The new method allows us to build the Reed-Muller form based on the analysis of Walsh-Hadamard coefficients. The presented method has much less complexity than the procedures which have been applied until now. Both the transforms and the presented Walsh-Hadamard spectral characterization of the Reed-Muller expansion are compared. An analysis of the properties of the spectra obtained from these transforms is made.

3

The dedicated decision support system in recognition of some uncertain disease entities

100%

Porwik P.

|

2009

|

tom Vol. 13

97--100

EN

This work presents the principles of image recognition, where quality-based methods are applied. The neural networks and additional software have been proposed. This goal was achieved by using non-parametric recognition algorithms. In this paper the two-state hybrid classification method has been proposed, where artificial intelligence algorithm is included. In recognition process, the learning method, selection and optimization of diagnostic parameters have been introduced. The integrated part of the classifier structure is voting mechanism, which indicates incorrect states of the system – for example the unrecognized images. Effectiveness of the system has been shown by means of examples, where ambiguous data have been incorporated – it is very often a practice of medical diagnostics.

4

Efficient spectral method of identification of linear Boolean function

100%

Porwik P.

|

tom Vol. 33, no 4

663-678

EN

This paper discusses a problem of recognition of the Boolean function's linearity. The article describes the spectral method of analysis of incompletely specified Boolean functions using the Walsh Transform. The linearity and nonlinearity play an important role in design of digital circuits. The analysis of the spectral coefficients' distribution allows to determine the various combinatorial properties of the Boolean functions: redundancy, monotonicity, self-duality, correcting capability, etc. which seems to be more difficult to obtain by means of other methods. In particular, the distribution of spectral coefficients allows us to determine whether Boolean function is linear. The method described in the paper can be easily used in investigations of large Boolean functions (of many variables), what seems to be very attractive for modern digital technologies. Experimental results demonstrate the efficiency of the approach.

5

The Three-Yalued Sign Walsh Coefficients Calculation for an Affine Boolean Function Recognition

100%

Porwik P.

|

tom T. 17, z. 3

199-211

EN

The paper discusses a problem of recognition of the Boolean function linearity. The linearity of function is investigated by means of the threevalued sign Walsh coefficients. The linearity and nonlinearity play an important role in many domains especially in designing of digital circuits. The knowledge about spectral coefflcients distribution allows to determine the various combinatorial properties of the Boolean functions: redundancy, monotonicity, self-duality, correcting capability, etc., which seems to be more difficult in realization by means of others methods. Experimental results demonstrate the efficiency of the approach.

PL

W pracy przedstawiono metodę rozpoznawania afmicznej funkcji boolowskiej na podstawie analizy jej trójwartościowego, znakowego widma Walsha. Boolowskie funkcje liniowe (afmiczne) są ważną klasą funkcji wykorzystywaną w kryptografii, projektowaniu układów arytmetycznych i kodów kontrolnych. Ze względu na ważność zagadnienia, metody znajdowania lub rozpoznawania takich funkcji są ciągle przedmiotem badań i udoskonaleń. Wykazano, że dla celów rozpoznawania w pełni określonych funkcji afinicznych klasyczną transformację Walsha-Hadamarda można uprościć i zrealizować w wersji znakowej, co ogranicza zakres dopuszczalnych wartości widma. W pracy wykazano, że widmo wyznaczone dla funkcji zapisanej w postaci ON-sześcianów (ang. cubes) może zostać również uproszczone. Zaproponowano trójwartościową reprezentację widma, gdzie używane są tylko trzy znaki (symbole) {+1, -1,0},co zmniejsza liczbę bitów niezbędnych do przechowywania wartości widma. Wyznaczono złożoność obliczeniową (O) oraz pamięciową (M) w bajtach obu przedstawionych w pracy metod. Wartości te wynoszą odpowiednio: dla znakowej transformacji Walsha-Hadamarda: O(2n) oraz M(2n+2), natomiast dla znakowej transformacji wykorzystującej zapis w postaci sześcianów O(n2 2n) oraz M(2n-2).

6

Combined fingerprints recognition method basing on image filtering, minutiae analysis and reference point location

100%

Porwik P.

|

tom Vol. 48, nr 4

31-39

EN

The method of fingerprint recognition is proposed. If such method the reference point is determined, appropriate bank of filters and minutiae-based representation has been used. That capturing fingerprints approach gives good results of fingerprint recognition. Proposed analysis of finger imprint allows to compute a unique fingerprint code. The proposed method by means of the NIST database was tested where different quality fingerprints are stored. Some elements of the method with other approaches was also compared.

PL

Przedstawiono metodę rozpoznawania linii papilarnych na podstawie wyznaczonego punktu referencyjnego, filtrowania obrazów oraz minucji. Otrzymane wyniki wykazują dobrą jakość rozpoznawania linii papilarnych. W wyniku odpowiedniej analizy można także otrzymać unikatowy kod linii papilarnych. Metoda była testowana za pomocą obrazów linii papilarnych zgromadzonych w bazie NIST. Niektóre jej elementy zostały porównane z innymi metodami.

7

An attempt to improve Eigenface algorithm efficiency for colour images

63%

Orczyk T. , Porwik P.

|

2010

|

tom Vol. 16

201--207

EN

This article presents an attempt to improve Eigenface algorithm efficiency by using image pre–filtering in order to eliminate background areas of the picture and illumination influence. The background is treated as noise, so when noise is present then efficiency of the algorithm decreases. In order to eliminating this inconvenience, analysed image is pre–filtered by means of the colour classifier. The classifier eliminates pixels which have different colour than an average human skin colour on a digital photo. This causes that the Eigenface algorithm is less sensitive to background noise. The illumination influence was minimized by using hue information instead of traditionally used luminance. The main advantage of the proposed approach is possibility of using in environments where diverse image background texture and scene illumination appears. The eigenfaces technique can be applied in handwriting analysis, voice recognition, hand gestures interpretation and medical imaging.

8

Własności informatyczne transformaty Walsha

63%

Porwik P. , Słota D.

|

1999

|

tom Vol. 45, nr 3-4

287-293

PL

W artykule omówiono nowe możliwości interpretacji współczynników widmowych funkcji boolowskich rozłożonych w dyskretny szereg Walsha. Stosując analizę współczynników można przeprowadzić pewna ogólną, dotychczas nigdzie nie opisaną klasyfikację funkcji boolowskich i szacować pewne ich właściwości. W pracy przedstawiono także w nowym świetle własności samej transformacji opartej na funkcjach Walsha. Na podstawie własności transformaty Walsha można uprościć interpretację współczynników lub na ich podstawie określić właściwości kombinacyjnych funkcji boolowskich na przykład redundancję lub liniowość co wydaje się być trudniejsze do wykonania za pomocą innych, klasycznych metod.

EN

This paper presents the short background about Boolean functions and their relation to Walsh - Hadamard transform. This transform reduces the costs of correlation computations. In this paper has been considered a new method of interpretation of Boolean function spectrum. The analysis methods described in this paper may be generalised to many - valued logical functions. By means of spectrum analysis we obtain information about properties of Boolean functions. For some classes of Boolean functions we can investigate the redundancy by spectrum coefficients analysis only. We have presented discussion of some theorems, which has played important role in analysis of Boolean functions. The presented result can be used in testing and very often plays important role in building of logical functions.

9

User verification based on the analysis of keystrokes while using various software

63%

Wesołowski T. E. , Porwik P.

|

tom Vol. 24

13--22

EN

The article presents the new approach to a computer users verification. The research concerns an analysis of user’s continuous activity related to a keyboard used while working with various software. This type of analysis constitutes a type of free-text analysis. The presented method is based on the analysis of users activity while working with particular computer software (e.g. text editors, utilities). A method of computer user proﬁling is proposed and an attempt to intrusion detection based on k-NN classifier is performed. The obtained results show that the introduced method can be used in the intrusion detection and monitoring systems. Such systems are especially needed in medical facilities where sensitive data are processed.

10

Dedicated spectral method of Boolean function decomposition

63%

Porwik P. , Stanković R. S.

|

2006

|

tom Vol. 16, no 2

271-278

EN

Spectral methods constitute a useful tool in the analysis and synthesis of Boolean functions, especially in cases when other methods reduce to brute-force search procedures. There is renewed interest in the application of spectral methods in this area, which extends also to the closely connected concept of the autocorrelation function, for which spectral methods provide fast calculation algorithms. This paper discusses the problem of spectral decomposition of Boolean functions using the Walsh transform and autocorrelation characteristics.

11

The pixel alignment based algorithm for continuous orientation field estimation

63%

Więcław Ł. , Porwik P.

|

2010

|

tom Vol. 15

47--52

EN

This paper presents a new estimation method of fingerprint orientation field. An accurate estimation of fingerprint orientation fields is an essential step in automatic fingerprint recognition systems (AFIS). Most popular, gradient-based method is very sensitive to noise (image quality). Proposed algorithm is a modification of, more resistant to noise, mask-based method, which provides orientation limited to discrete values. This modification is based on aggregation of pixel values differentiation and was used to more precise estimation of magnitude of orientation vectors. This approach allows to obtain a continuous values of orientation field still maintaining robust to noise.

12

Comparative investigations of similarity measure exploited in medical images preselection

63%

Wróbel K. , Porwik P.

|

2004

|

tom Vol. 8

II25--31

EN

Nowadays, the most significant impact of digital image processing in the area of applications are real–world problems. Many new technological trends in medicine and digital processing have been implemented. Several factors indicate such development. A major one is the perpetually declining cost of the computer equipment required. Both processing unit and capacity of storage devices continue to become less expensive year by year. Another factor is the increasing availability of equipment for digitising and displaying images. In modern image processing, images have to be compared each other because such approach allows us to automate of retrieval process. Computer image retrieving is today especially important in medical diagnostics [1,7] or in preliminary images selection [8,9]. Today, in the digital image processing are used techniques and methods which have well known mathematical backgrounds. It can be observed, that in the area of digital signal processing, the Hough and well known the Fourier transform are exploited very often. These transforms are frequently use in image retrieving and can be implemented as computer applications. In many cases the mentioned methods give promising results in images classification or preselection [1,2,4,5,11]. Special properties of such transforms can be used in statistical or comparative goals, especially when searched information has graphic form. Taking into account the mentioned applications, transforms as methods of preliminary medical images selection have been investigated. From this reason pictures, analysing in the paper, to medical images have been limited.

13

The Haar-wavelet transform in digital image processing: its status and achievements

63%

Porwik P. , Lisowska A.

Machine Graphics and Vision

|

2004

|

tom Vol. 13, No. 1/2

79-98

EN

Image processing and analysis based on continuous or discrete image transforms are classic techniques. The image transforms are widely used in image filtering, data description, etc. Nowadays, wavelet theorems make up very popular methods of image processing, denoising and compression. Considering that Haar functions are the simplest wavelets, these forms are used in many methods of discrete image transforms and processing. The image transform theory is a well known area characterized by a precise mathematical background, but in many cases some transforms have particular properties which have not been investigated yet. This paper presents graphic dependences between parts of Haar and wavelets spectra for the first time. It also presents a method of image analysis by means of the wavelet-Haar spectrum. Some properties of the Haar and wavelet spectrum are investigated. Extraction of image features directly from spectral coefficients distribution is presented. The paper shows that two-dimensional products of both Haar and wavelet functions can be treated as exstractors of particular image features. Furthermore, it is also shown that some coefficients from both the spectra are proportional, which simplifies computations and analyses to some degree.

14

Signature pre - procesing based on walsh coefficients

63%

Porwik P. , Wróbel K.

|

tom Vol. 12

57--61

EN

Recognition and verification of persons are difficult and important tasks today. In many fields of human activities (driver's licenses, passports, electronic cards, etc.), signature recognition of person is needed. Hence, it inspires the development of a wide range of automatic identification systems. Signatures have been used for many centuries as a method of people's identification. Signatures recognition was performed manually by experts in the past. Nowadays, these procedures are very often automatically applied. In this paper the system that automatically authenticates documents based on the owner's handwritten signature is presented.

15

Word extraction method in human speech processing

63%

Porwik P. , Proksa R.

|

tom Vol. 12

209--216

EN

A major problem in isolated-word speech recognition systems is detection of the beginning and ending boundaries of the word. It is an essential of speech recognition algorithms, where signal speech segments should be reliably separated. During speech recognition background noise is also recorded, hence the word isolation is difficult. The parametric representation of the speech must provide enough information to characterize the words and to differentiate between acoustically similar words. In this paper the method of words extraction from human speech will be considered.

16

The voice synthetiser of polish text for blind persons

63%

Porwik P. , Szczepankiewicz M.

|

tom Vol. 4

MT101--109

EN

In this paper we present new method of computer text analyser and computer Polish speech (words) generator. In the described computer program the grammatical characteristics of Polish speech and accents in some words have been taken into consideration. All users' actions are commented by artificial, computer voice. The group of blind students of University of Silesia have examined and tested the presented final program for over one year. Described software tool has in a lot of cases better parameters than others, commercial products.

17

Signature recognition method by means of the windows technique

51%

Porwik P. , Wróbel K. , Dorosz R.

|

tom Vol. 14, no 2-3

43-50

EN

This study presents a new method of determining the similarity of signatures. The method described in this paper allows comparing signatures with the use of any similarity coefficient, without the necessity of using additional algorithms equalizing the lengths of the sequences being compared. The proposed method consists in dividing the signatures into windows with a given size and calculating the value of similarity between individual windows. The performed studies took into account selected similarity coefficients, known from the literature, and different parameters of the method. In addition, there was determined the usability of individual similarity coefficients, taking into account different features of signatures.

18

Instance based kNN modification for classification of medical data

51%

Orczyk T. , Porwik P. , Lewandowski M. , Cholewa M.

|

tom Vol. 24

99--106

EN

Paper describes a novel modification to a well known kNN algorithm, which enables using it for medical data, which often is a class-imbalanced data with randomly missing values. Paper presents the modified algorithm details, experiment setup, results obtained on a cross validated classification of a benchmark database with randomly removed values (missing data) and records (class imbalance), and their comparison with results of the state of the art classification algorithms.

19

Combined off-line type signature recognition method

51%

Porwik P. , Para T. , Smacki Ł. , Żuławiński S.

|

tom Vol. 10

81--90

EN

In this paper the off-line type signature analysis have been presented. The signature recognition is composed of some features. Different influences of such features were tested and stated. Proposed approach gives good signature recognition level, hence described method can be used in many areas, for example in biometric authentication, as biometric computer protection or as method of the analysis of person's behaviour changes.

20

On-line signature recognition method based on linear regression

51%

Para T. , Porwik P. , Wróbel K.

|

2007

|

tom Vol. 11

97--104

EN

Nowadays, automatic signature verification is an active area of researches in numerous applications such as bank check verification, access restriction or special areas such as police investigations. In our researches signature was captured by Topaz SigLite T-LBK750-HSB device, where some dynamic features of signature can be also registered. In many transactions, the electronic verification of a person's identity is beneficial, hence it inspires the development of a wide range of automatic identification systems. In this paper the system that automatically authenticates documents based on the owner's handwritten signature is presented.