Wyniki wyszukiwania - BazTech

1

Method of vibration signal processing and load-type identification of a mill based on ACMD-SVD

Tang Wencong, Zhang Fangwei, Luo Xiaoyan, Wan Junliang, Deng Tao

Gospodarka Surowcami Mineralnymi

|

2023

|

T. 39, z. 1

217--233

EN

Green mine construction is the main melody of mining development and problems such as safe production, energy saving and consumption reduction need to be solved urgently. The working conditions of the mill are complex in the process of grinding. Aiming at the problems existing in the feature extraction and load prediction of the mill, a signal-processing method based on adaptive chirp mode decomposition (ACMD) and a standardized variable distance classifier (SVD) is proposed. Firstly, the recursive framework of the ACMD method is used to obtain the initial frequency of mill vibration signals. Secondly, the initial frequency is used to reconstruct the high-resolution component of the mill vibration signal through the iterative frame in the ACMD method. The frequency corresponding to the frequency domain peak of the reconstructed signal is then selected as the mill load feature vector. Finally, with consideration to the influence of standard deviation and standardized variable factors on the feature vectors, a standardized variable distance classifier is proposed. The feature vectors of the mill load are input into the SVD model for training, and the state types of the mill load are obtained. The method is applied to the grinding experiment and the results show that the frequency-domain features obtained by the mill vibration signal-processing method based on ACMD-SVD are obvious, which has high accuracy in the identification of mill load types, and provides a new idea for the extraction of mill load features and prediction of the mill load.

PL

Budowa zielonej kopalni jest główną melodią rozwoju górnictwa, a problemy takie jak: bezpieczna produkcja, oszczędność energii i redukcja zużycia wymagają pilnego rozwiązania. Warunki pracy młyna w procesie mielenia są złożone. Mając na celu rozwiązanie problemów występujących w ekstrakcji cech i przewidywaniu obciążenia młyna, zaproponowano metodę przetwarzania sygnału opartą na dekompozycji w trybie adaptacyjnym ACMD (Adaptive Chirp Made Decomposition) i znormalizowanym klasyfikatorze zmiennej odległości SVD (Variable Distance Classifier). Po pierwsze, rekurencyjna struktura metody ACMD jest wykorzystywana do uzyskania początkowej częstotliwości sygnałów drgań młyna. Po drugie, częstotliwość początkowa jest wykorzystywana do rekonstrukcji wysokorozdzielczej składowej sygnału drgań młyna poprzez ramkę iteracyjną w metodzie ACMD. Częstotliwość odpowiadająca pikowi w dziedzinie częstotliwości rekonstruowanego sygnału jest następnie wybierana jako wektor cech obciążenia młyna. Na koniec, biorąc pod uwagę wpływ odchylenia standardowego i standaryzowanych czynników zmiennych na wektory cech, zaproponowano standaryzowany klasyfikator odległości o zmiennej długości. Wektory cech obciążenia młyna są wprowadzane do modelu SVD w celu uczenia i uzyskiwane są typy stanu obciążenia młyna. Metodę zastosowano w eksperymencie mielenia, a wyniki pokazują, że cechy w dziedzinie częstotliwości uzyskane za pomocą metody przetwarzania sygnału drgań młyna opartej na ACMD-SVD są oczywiste, co ma wysoką dokładność w identyfikacji typów obciążeń młyna i zapewnia nowy pomysł na ekstrakcję cech obciążenia młyna i predykcję obciążenia młyna.

2

Image encryption algorithm based on rear-mounted phase mask and random decomposition

Yadav Shivani, Singh Hukum

Optica Applicata

|

2022

|

Vol. 52, nr 2

195--212

EN

To escalate the image encryption a new method has been devised which includes double random phase encoding (DRPE) using rear phase masking and random decomposition (RD) technique stranded on fractional Fourier transform. Here, asymmetric cryptographic system is developed in fractional Fourier transform (FrFT) mode using two random phase masks (RPM) and a rear mounted phase mask. In the projected scheme a colored image is decomposed into R, G and B channels. The amplitude of each channel is normalized, phase encoded and modulated using RPM. The modulated R, G and B channels of the colored image are individually transformed using FrFT to produce corresponding encrypted image. The proposed scheme is authorized on grayscale image also. The norm behind the development of the suggested scheme has been elaborated by carrying out cryptanalysis on system based on the RD. The method helps in escalations of the protection of double random phase encoding by cumulating the key length and the parameter amount, so that it vigorously can be used against various attacks. The forte of the suggested cryptographic system was verified using simulations with MATLAB 7.9.0 (R2008a). The efficiency of the suggested scheme includes the analysis using singular value decomposition (SVD), histogram and correlation coefficient.

3

Design of Hybrid Precoder for mm-Wave MIMO System Based on Generalized Triangular Decomposition Method

Thurpati Sammaiah, Muthuchidambaranathan P.

International Journal of Electronics and Telecommunications

|

2022

|

Vol. 68. No. 3

519--525

EN

Hybrid precoding techniques are lately involved a lot of interest for millimeter-wave (mmWave) massive MIMO systems is due to the cost and power consumption advantages they provide. However, existing hybrid precoding based on the singular value decomposition (SVD) necessitates a difficult bit allocation to fit the varying signal-to-noise ratios (SNRs) of altered sub-channels. In this paper, we propose a generalized triangular decomposition (GTD)-based hybrid precoding to avoid the complicated bit allocation. The development of analog and digital precoders is the reason for the high level of design complexity in analog precoder architecture, which is based on the OMP algorithm, is very non-convex, and so has a high level of complexity. As a result, we suggest using the GTD method to construct hybrid precoding for mmWave mMIMO systems. Simulated studies as various system configurations are used to examine the proposed design. In addition, the archived findings are compared to a hybrid precoding approach in the classic OMP algorithm. The proposed Matrix Decomposition’s simulation results of signal-to-noise ratio vs spectral efficiencies.

4

Analiza porównawcza metod znakowania wodnego obrazów medycznych

Duda Sylwia, Fijałek Dominik, Kozieł Grzegorz

Journal of Computer Sciences Institute

|

2020

|

Vol. 17

379--383

EN

The article is devoted to the analysis of watermarking algorithms in terms of their use in marking medical images. The algorithms based on the Integer Wavelet Transform (IWT), Discrete Cosine Transform (DCT), and Singular Value Decomposition (SVD) were compared. The algorithms were implemented using the combinations: IWT, IWT-DCT, and IWT-SVD. As part of the research, the level of disturbances caused by embedding the watermark was checked using subjective and objective methods. The attack resistance of the watermarked images was tested and the steganographic capacity was measured. All algorithms are based on IWT, however, each has different advantages. The algorithm based on the IWT showed the highest capacity. The most resistant to attacks is IWT-SVD, and the lowest level of interference was obtained for the IWT-DCT algorithm.

PL

Artykuł poświęcono analizie algorytmów znakowania wodnego pod kątem wykorzystania w znakowaniu obrazów medycznych. Porównano algorytmy oparte o całkowitą transformatę falkową (IWT), dyskretną transformatę kosinusową (DCT) i rozkład według wartości osobliwych (SVD). Zaimplementowano algorytmy stosując kombinacje: IWT, IWT-DCT i IWT-SVD. W ramach badań sprawdzono poziom zakłóceń spowodowanych osadzaniem znaku wodnego przy pomocy metod subiektywnych i obiektywnych. Przeprowadzono badania odporności oznakowanych obrazów na ataki i zmierzono pojemność steganograficzną. Wszystkie algorytmy bazują na IWT, jednakże każdy z nich ma inne zalety. Największą pojemność wykazał algorytm oparty o IWT. Najodporniejszy na ataki jest IWT-SVD, a najmniejszy poziom zakłóceń uzyskano dla algorytmu IWT-DCT.

5

Predicting pairwise relations with neural similarity encoders

Horn F., Müller K. R.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2018

|

Vol. 66, nr 6

821--830

EN

Matrix factorization is at the heart of many machine learning algorithms, for example, dimensionality reduction (e.g. kernel PCA) or recommender systems relying on collaborative filtering. Understanding a singular value decomposition (SVD) of a matrix as a neural network optimization problem enables us to decompose large matrices efficiently while dealing naturally with missing values in the given matrix. But most importantly, it allows us to learn the connection between data points’ feature vectors and the matrix containing information about their pairwise relations. In this paper we introduce a novel neural network architecture termed similarity encoder (SimEc), which is designed to simultaneously factorize a given target matrix while also learning the mapping to project the data points’ feature vectors into a similarity preserving embedding space. This makes it possible to, for example, easily compute out-of-sample solutions for new data points. Additionally, we demonstrate that SimEc can preserve non-metric similarities and even predict multiple pairwise relations between data points at once.

6

Automated quantification of ultrasonic fatty liver texture based on curvelet transform and SVD

Bharath R., Mishra P. K., Rajalakshmi P.

Biocybernetics and Biomedical Engineering

|

2018

|

Vol. 38, no. 1

145--157

EN

Fatty liver is a prevalent disease and is the major cause for the dysfunction of the liver. If fatty liver is untreated, it may progress into chronic diseases like cirrhosis, hepatocellular carcinoma, liver cancer, etc. Early and accurate detection of fatty liver is crucial to prevent the fatty liver progressing into chronic diseases. Based on the severity of fat, the liver is categorized into four classes, namely Normal, Grade I, Grade II and Grade III respectively. Ultrasound scanning is the widely used imaging modality for diagnosing the fatty liver. The ultrasonic texture of liver parenchyma is specific to the severity of fat present in the liver and hence we formulated the quantification of fatty liver as a texture discrimination problem. In this paper, we propose a novel algorithm to discriminate the texture of fatty liver based on curvelet transform and SVD. Initially, the texture image is decomposed into sub-band images with curvelet transform enhancing gradients and curves in the texture, then an absolute mean of the singular values are extracted from each curvelet decomposed image, and used it as a feature representation for the texture. Finally, a cubic SVM classifier is used to classify the texture based on the extracted features. Tested on a database of 1000 image textures with 250 image textures belonging to each class, the proposed algorithm gave an accuracy of 96.9% in classifying the four grades of fat in the liver.

7

Nanosatellite Attitude Estimation from Vector Measurements Using SVD-Aided UKF Algorithm

Cilden D., Soken H. E., Hajiev C.

Metrology and Measurement Systems

|

2017

|

Vol. 24, nr 1

113--125

EN

The integrated Singular Value Decomposition (SVD) and Unscented Kalman Filter (UKF) method can recursively estimate the attitude and attitude rates of a nanosatellite. At first, Wahba’s loss function is minimized using the SVD and the optimal attitude angles are determined on the basis of the magnetometer and Sun sensor measurements. Then, the UKF makes use of the SVD’s attitude estimates as measurement results and provides more accurate attitude information as well as the attitude rate estimates. The elements of “Rotation angle error covariance matrix” calculated for the SVD estimations are used in the UKF as the measurement noise covariance values. The algorithm is compared with the SVD and UKF only methods for estimating the attitude from vector measurements. Possible algorithm switching ideas are discussed especially for the eclipse period, when the Sun sensor measurements are not available.

8

Calculation Models for Acoustic Analysis of St. Elizabeth of Hungary Church in Jaworzno Szczakowa

Kosała K.

Archives of Acoustics

|

2016

|

Vol. 41, No. 3

485--498

EN

In parallel with research conducted using conventional methods, a uniform index method for assessing the acoustic quality of Roman Catholic churches has been developed. The latest version of the index method has been created using the index observation matrix of 12 churches which have been rated by means of the single number global index. Assessments of the acoustic quality of any Roman Catholic church, using two calculation models: the Global Acoustic Properties Index (GAP) and the Global Index (GI), are shown in the article. The verification was performed on the example of one church, showing the way of calculating global indices to assess the acoustic quality of a new facility. The next stages in the development of the index method for assessing the acoustic quality of churches were taking into account the audience, using simulation tests and determining the spatial distribution of the single number GAP index in an examined church. An attempt to use the GAP and GI calculation models to assess the acoustic properties of some churches is also shown in the article.

9

A practical application of kernel-based fuzzy discriminant analysis

Gao J. Q., Fan L. Y., Li L., Xu L. Z.

International Journal of Applied Mathematics and Computer Science

|

2013

|

Vol. 23, no. 4

887--903

EN

A novel method for feature extraction and recognition called Kernel Fuzzy Discriminant Analysis (KFDA) is proposed in this paper to deal with recognition problems, e.g., for images. The KFDA method is obtained by combining the advantages of fuzzy methods and a kernel trick. Based on the orthogonal-triangular decomposition of a matrix and Singular Value Decomposition (SVD), two different variants, KFDA/QR and KFDA/SVD, of KFDA are obtained. In the proposed method, the membership degree is incorporated into the definition of between-class and within-class scatter matrices to get fuzzy between-class and within-class scatter matrices. The membership degree is obtained by combining the measures of features of samples data. In addition, the effects of employing different measures is investigated from a pure mathematical point of view, and the t-test statistical method is used for comparing the robustness of the learning algorithm. Experimental results on ORL and FERET face databases show that KFDA/QR and KFDA/SVD are more effective and feasible than Fuzzy Discriminant Analysis (FDA) and Kernel Discriminant Analysis (KDA) in terms of the mean correct recognition rate.

10

Multidimensional analysis and assessment of combustion engine technical state basis on SVD method with modern engineers software application

Kałaczyński T., Łukasiewicz M.

Journal of Polish CIMAC

|

2012

|

Vol. 7, no 2

65-78

EN

The changes of vibration estimators as a result of engine maladjustment, waste, damage or failure are the main idea of vibrodiagnostic investigation. Diagnostic investigations that use of vibration to determine the technical state of combustion engines are very difficult. Only a few proposed methods could have a wider technical use in diagnostics. This paper showes the validation of research results of the use of the Singular Value Decomposition (SVD) method. The research object is combustion engine No. 138C.2.048 with 1.4l. swept capacity, power 55 kW / 75 KM, generally applied to Fiat. It is possible to introduce generated vibration signals as well as the investigation of its adjustment influence on the combustion engine vibration signals change. Thanks to SVD methods, it is possible to decide which symptom given in the observation matrix is the best to recognize the technical state of combustion engines. The results that are introduced in this paper are only the part of realized investigative project and they do not describe wholes of the investigative question, only chosen aspects.

11

Singular vectors in the synthesis of indices for acoustic assessment of objects

Kosała K.

Diagnostyka

|

2012

|

nr 2(62)

31-36

EN

The SVD (Singular Value Decomposition) technique makes it possible to transform a set of correlated data into uncorrelated ones without the loss of any information. The new system of mutually uncorrelated variables is comparable to the initial one. Apart from decorrelation of data, SVD makes full variable correlation possible, which ensures no copying of certain information during the addition of such new variables. This article presents the application of the SVD technique for the construction of singlenumber indices for assessment of correlated object acoustic parameters using the methods of decorrelation and full correlation. Verification of the proposed single-number assessment indices was carried out using the example of sacral buildings. For one Roman Catholic church with flawed acoustics, the application of a single-number index for assessment of the variant of proposed acoustic adaptation of its interior was shown. The synthetic index applied for sacral objects, being an approximated general measure of assessment, can be a helpful tool for designers and during assessment of church interiors in terms of the acoustic functioning of the object.

PL

Technika rozkładu względem wartości szczególnych (SVD) umożliwia przekształcenie zbioru danych skorelowanych w dane nieskorelowane, bez utraty jakiejkolwiek informacji. Nowy układ zmiennych wzajemnie nieskorelowanych jest porównywalny z układem wyjściowym. Oprócz dekorelacji, SVD umożliwia korelację zupełną zmiennych, zapewniającą brak powielenia pewnych informacji przy sumowaniu doskonale skorelowanych nowych zmiennych. W artykule pokazano wykorzystanie techniki SVD do konstrukcji jednoliczbowych wskaźników oceny skorelowanych parametrów akustycznych obiektów metodami dekorelacji i korelacji zupełnej. Weryfikację zaproponowanych jednoliczbowych ocen wskaźnikowych przeprowadzono na przykładzie kilku obiektów sakralnych. Dla jednego kościoła-rzymsko katolickiego o wadliwej akustyce pokazano zastosowanie jednoliczbowego wskaźnika do oceny wariantu zaproponowanej adaptacji akustycznej tego wnętrza. Syntetyczny wskaźnik zastosowany dla obiektów sakralnych, będący przybliżoną miarą ogólną oceny, może być pomocnym narzędziem dla projektantów oraz przy ocenie wnętrz kościelnych, związanej z prawidłowym pod względem akustycznym funkcjonowaniem obiektu.

12

Zastosowanie ewolucji wartości szczególnych w wielosymptomowej diagnostyce maszyn

Tabaszewski M., Cempel C.

Diagnostyka

|

2011

|

nr 3(59)

65-73

PL

W pracy przedstawiono możliwość obserwacji wartości szczególnych ewoluujących w czasie jako wielkości niosących informację diagnostyczną. Aby wykazać przydatność wartości szczególnych w tym zakresie dokonano wielu symulacji numerycznych. Wykazano w nich, że obserwując zmiany wartości szczególnych uzyskanych z rozkładu SVD w czasie życia maszyny, należy zwrócić uwagę na występowanie punktów zwrotnych. Występowanie tych punktów może świadczyć o skokowej zmianie wartości symptomów (np. pęknięcia struktury). Fakt wystąpienia takiego skoku może zostać łatwo przeoczony ze względu na zmienne parametry robocze maszyny, które z kolei wpływają na wartości mierzonych symptomów a także symptomów uogólnionych po rozkładzie SVD. Wartości szczególne są prawie nie wrażliwe na zmiany parametrów roboczych, tak więc łatwiej wychwycić tego typu skoki w ich ewolucji niż bezpośrednio w symptomie. W pracy przedstawiono także przykład zastosowania proponowanej metody dla rzeczywistych danych diagnostycznych pochodzących z łożysk tocznych.

EN

The paper presents a possibility of observation of singular values evolving in time as quantities carrying diagnostic information. To prove the usefulness of singular values for this purpose many numerical simulations have been conducted. It has been proved that when observing changes of singular values obtained from SVD during the lifetime of a machine, the appearance of reversal points must be taken into account. The appearance of such points may prove that the symptom values change abruptly (eg. structure cracking). The appearance of such an abrupt change can easily be overlooked because of variable working parameters of the machine, which influence the values of measured symptoms and generalized symptoms after SVD. Singular values are almost insensitive to changes of working parameters, so it is easier to pick out such changes in their evolution than directly in symptoms. The paper also presents an example of application of the proposed method for real diagnostic data obtained from ball bearings.

13

Decorrelation of indices for a single number assessment of the acoustic quality of sacral objects

Kosała K.

Diagnostyka

|

2011

|

nr 2(58)

49-52

EN

A new approach for solving the problem of a single number index formula by creating the index from mutually-correlated indices by means of decorrelation of the Index Observation Matrix (IOM) was shown. The orthogonal singular vectors obtained from SVD were used in order to build a single number index. Application of the proposed formula for a local single number index of selected acoustic parameters of the interiors of six Roman Catholic churches was shown. Decorrelation of the indices for a single number assessment of the acoustic quality of sacral objects will be applied for complex global acoustic assessment of such interiors' index method, which is currently being improved.

PL

Pokazane w artykule nowe podejście do rozwiązania problemu opracowania wskaźnika jednoliczbowego, polega na utworzeniu tego wskaźnika ze wskaźników skorelowanych ze sobą, na drodze dekorelacji wskaźnikowej macierzy obserwacji. Do konstrukcji wskaźnika jednoliczbowego wykorzystano, uzyskane z rozkładu SVD, ortogonalne wektory szczególne. Weryfikację zaproponowanego wzoru na jednoliczbowy lokalny wskaźnik wybranych parametrów akustycznych obiektów sakralnych pokazano na sześciu rzeczywistych kościołach rzymskokatolickich. Dekorelacja wskaźników w jednoliczbowej ocenie jakości akustycznej obiektów sakralnych będzie wykorzystana do kompleksowej globalnej oceny jakości akustycznej obiektów sakralnych - metody wskaźnikowej, która jest w dalszym ciągu udoskonalana.

14

The multidimensional analysis of the combustion engine investigations results with SVD method

Łukasiewicz M.

Journal of KONES

|

2010

|

Vol. 17, No. 2

285-292

EN

New approach to investigation of combustion engine technical state is vibroacoustics as a diagnostic tool. The main idea of vibroacoustics investigation is following the changes of vibration estimators as a result of engine maladjustment, waste, damages or its failure. Combustion engines technical state diagnostic investigations with use of vibration are very difficult and only few proposed methods could have wider technical use in diagnostics. The combustion engine No. 138C. 2.048 with 1.41. swept capacity, power 55 kW / 75 KM, generally applied to Fiat and Lancia is the investigation objęci. The engine is situated in the investigative laboratory of combustion engines in UTP Bydgoszcz, li makes possible to introduce generated vibration signals as well as the investigation of his adjustment influence on the combustion engine vibration signals change. As a validation of investigation results in this paper is shown presentation of Singular Value Decomposition (SVD) method. The SVD method is the appropriate tool for analysing a mapping from one vector space into another vector space, possibly with a different dimension. Thanks to SVD methods we could decide which symptom given in observation matrix is the best to recognize a set of combustion engine technical state. Relationships cause - consecutive expressing quantitative relation between studied variable symptoms results were qualified using the function of the multiple regression. Introduced in this paper results of investigations are only the part of realized investigative project and they do not describe wholes of the investigative question, only chosen aspects.

15

An improved self-embedding algorithm: robust protection against lossy compression attacks in digital image watermarking

Yogarajah P., Condell J., Curran K., Cheddad K., Mckevitt P.

Image Processing & Communications

|

2010

|

Vol. 15, no 1

47-57

EN

Watermarking is the process of embedding watermarks into an image such that the embedded watermark can be extracted later. Lossy compression attacks in digital water-marking are one of the major issues in digital watermarking. Cheddad et al. proposed a robust secured self-embedding method which is resistant to a certain amount of JPEG compression. Our experimental results show that the self-embedding method is resistant to JPEG compression attacks and not resistant to other lossy compression attacks such as Block Truncation Coding (ETC) and Singular Value Decomposition (SVD). Therefore we improved Cheddad et al's. method to give better protection against ETC and SVD compression attacks.

16

CORDIC and SVD implementation in digital hardware

Szecówka P. M., Malinowski P.

Elektronika : konstrukcje, technologie, zastosowania

|

2010

|

Vol. 51, nr 11

26-29

EN

Singular Value Decomposition (SVD) is classified among the most effective numeric methods of matrices inversion. The paper presents a study of hardware implementation of SVD and CORDIC algorithms. Various digital architectures were proposed and compared, including low-cost sequential and high-performance pipelined solutions. Fixed point and floating point arithmetic was considered. The concepts were implemented in VHDL, verified and synthesized with Xilinx tools. Selected approach was physically implemented and tested.

PL

Algorytm SVD (Singular Value Decomposition) jest zaliczany do najbardziej efektywnych metod pozwalających odwracać macierze. Artykuł opisuje próbę sprzętowej realizacji algorytmów CORDIC i SVD. Rozważono szereg architektur - warianty bardzo oszczędne sekwencyjne, a także rozwiązania pozwalające uzyskać wysoką wydajność obliczeniową, z przetwarzaniem potokowym. Porównano także rezultaty uzyskane przy zastosowaniu arytmetyki stało- i zmiennoprzecinkowej. Koncepcje zostały zaimplementowane w języku opisu sprzętu (VHDL) poddane weryfikacji i syntezie za pomocą narzędzi Xilinx. Niektóre warianty zostały przetestowane fizycznie.

17

Możliwości zastosowania rozkładu względem wartości szczególnych w analizie wibroakustycznej obiektów budowlanych

Engel Z., Engel J., Kosała K.

Zeszyty Naukowe Politechniki Rzeszowskiej. Mechanika

|

2008

|

z. 74 [258]

55-64

EN

The possibilities of application the mathematical method - Singular Value Decomposition in wibroacoustical analysis of building objects on sacral objects example were shown in the paper. Singular Value Decomposition (SVD) is a technique used in the reduction of matrix sizes and analysis of independencies of variables. Application of this tool in acoustic problems of sacral interiors gave the possibility of analysis of proposed index method of acoustic assessment of sacral objects. The dependencies between partial indices were obtained from SVD as well as the formulae which can approximately assess global acoustic quality of sacral interior. The verification of index method with SVD was performed for six real roman-catholic churches. The approximate global index and partial indices can be used for acoustic assessment of real interior where acoustic adaptation is needed as well as for designed sacral rooms. The proposed indices are calculated from simulation research on created geometrical model. The inverse problem was formulated in the paper, where at the assumed global index, partial indices determining individual properties - are looked for.

18

Using singular value decomposition in textile production quality control

Bednarczuk P.

Theoretical and Applied Informatics

|

2008

|

Vol. 20, No. 3

167-177

PL

Artykuł przedstawia możliwość wykorzystania dekompozycji według wartości osobliwych SVD do kontroli jakości tkanin. Podejście to polega na wykorzystaniu do analizy tylko pierwszych wartości osobliwych i badaniu ich odchylenia standardowego od wartości przyjętych za odniesienie. Badany obiekt to tkanina, przedstawiony w postaci obrazu, zapisanego jako macierz danych. Artykuł zawiera porównanie skuteczności i szybkości różnych algorytmów i metod analizy danych. Metoda SVD pod względem detekcji defektów zapowiada dużą skuteczność. Także szybkość zaproponowanego rozwiązania jest porównywalna z najszybszymi algorytmami i jest najlepsza wśród metod o tej samej skuteczności. Wyniki eksperymentu badania pięć różnych defektów tkanin potwierdzają możliwości badanej metody.

EN

The article presents a possibility of using singular value decomposition in textile quality control. This approach consists in using only first singular values in the analysis and examining their standard deviation from the referential values. The examined object is textile, represented in the form of an image and saved as a data matrix. The article includes comparison of the efficiency and speed of different data analysis algorithms and methods. SVD method, in regard to defect detection, shows a high efficiency. Furthermore, the speed of the proposed solution is comparable with the fastest algorithms and is the best from among the methods with the same efficiency. The results of the conducted and described experiment consisting in examining five different textile defects confirm the potential of the method chosen.

19

The application of the matrix decomposition into singular values method for detecting changes of the system dynamic characteristics

Kromulski J., Kaźmierczak H., Pawłowski T.

Vibrations in Physical Systems

|

2006

|

Vol. 22

191--196

20

Testing dimension reduction methods for image retrieval

Moravec P., Snášel V.

Image Processing & Communications

|

2006

|

Vol. 11, no 2

51-61

EN

In this paper, we compare performance of several dimension reduction techniques, namely LSI, NMF, SDD and FastMap. The qualitative comparison is based on rank lists and evaluated on a collection of faces from the Olivetti Research Lab. We compare the quality of these methods from several standpoints: the visual impact, quality of generated "eigenfaces", size of reduced matrices and retrieval performance.