This paper describes an image fusion approach based on CNNs and DWT. According to the suggested method, First Each inputted image is decomposed into approximation coefficients and detail coefficients using DWT. The second step is to maximize the weights using CNN with detailed coefficients. Third, using maximum weight and max pooling, the combined detail images are produced. Fourth, an average pooling of the approximate coefficients is used to determine the final approximation coefficients. Lastly, Inverse DWT is then used to combine the detail and final approximation images to produce the final fused image. Experiments are carried out on four different fusion datasets. Different Quality checking metrics are used to analyze the data, and the results are then contrasted with more recent and usual fusion techniques. The result substantiates that the suggested technique performs better than the existing fusion methods. It is also appropriate for real-time applications due to the proposed method's reasonable computational time and simple yet efficient implementation.
PL
Artykuł dotyczy wielosensorowych konwolucyjnych sieci neuronowych (MS CNN) do fuzji obrazów w oparciu o konwolucyjne sieci neuronowe (CNN) i dyskretną transformację falkową (DWT). Zgodnie z sugerowaną metodą, najpierw każdy wprowadzony obraz jest rozkładany na współczynniki aproksymacji i współczynniki szczegółowości przy użyciu DWT. Drugim krokiem jest maksymalizacja wag za pomocą CNN ze szczegółowymi współczynnikami. W trzecim etapie, przy użyciu maksymalnej wagi i maksymalnego łączenia, tworzone są połączone szczegółowe obrazy. W czwartym etapie stosuje się średnią sumę przybliżonych współczynników w celu określenia ostatecznych współczynników przybliżenia. Na koniec stosuje się odwrotną DWT do łączenia obrazów szczegółowych i końcowych przybliżeń w celu uzyskania ostatecznego połączonego obrazu. Eksperymenty przeprowadzane są na czterech różnych zbiorach danych. Do analizy danych wykorzystuje się różne wskaźniki kontroli jakości, a następnie wyniki porównuje się z nowszymi i typowymi technikami łączenia. Wynik potwierdza, że sugerowana technika działa lepiej niż istniejące metody aglutynacji. Nadaje się również do zastosowań w czasie rzeczywistym ze względu na rozsądny czas obliczeń proponowanej metody oraz prostą, ale efektywną implementację.
This article addresses the problem of fault detection in robot manipulator systems. In the production field, online detection and prevention of unexpected robot stops avoids disruption to the entire manufacturing line. A number of researchers have proposed fault diagnosis architectures for electrical systems such as induction motor, DC motor, etc..., utilising the technique of discrete wavelet transform. The results obtained from the use of this technique in the field of diagnosis are very encouraging. Inspired by previous work, The objective of this paper is to present a methodology that enables accurate fault detection in the actuator of a two-degree of freedom robot arm to avoid system performance degradation. A partial reduction in joint torque constitutes the actuator fault, resulting in a deviation from the desired end-effector motion. The actuator fault detection is carried out by analysing the torques signals using the wavelet transform. The stored energy at each level of the transform contains information which can be used as a fault indicator. A Matlab/Simulink simulation of the manipulator robot demonstrates the effectiveness of the proposed technique.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The diagnosis of urinary tract infections and kidney diseases using urine microscopy images has gained significant attention of medical community in recent years. These images are usually created by physicians’ own rule of thumb manually. However, this manual urine sediment analysis is usually labor-intensive and time-consuming. In addition, even when physicians carefully examine an image, an erroneous cell recognition may occur due to some optical illusions. In order to achieve cell recognition in low-resolution urine microscopy images with a higher level of accuracy, a new super resolution Faster Region-based Convolutional Neural Network (Faster R-CNN) method is proposed. It aims to increase resolution in low-resolution urine microscopy images using self-similarity based single image super resolution which was used during the pre-processing. Denoising based Wiener filter and Discrete Wavelet Transform (DWT) are used to de-noise high resolution images, respectively, to increase the level of accuracy for image recognition. Finally, for the feature extraction and classification stages, AlexNet, VGFG16 and VGG19 based Faster R-CNN models are used for the recognition and detection of multi-class cells. The model yielded accuracy rates are 98.6%, 96.4% and 96.2% respectively.
Parkinson's disease is a recognizable clinical syndrome with a variety of causes and clinical presentations; it represents a rapidly growing neurodegenerative disorder. Since about 90 percent of Parkinson's disease sufferers have some form of early speech impairment, recent studies on tele diagnosis of Parkinson's disease have focused on the recognition of voice impairments from vowel phonations or the subjects' discourse. This paper presents a new approach for Parkinson's disease detection from speech sounds that are based on CNN and LSTM and uses two categories of characteristics. These are Mel Frequency Cepstral Coefficients (MFCC) and Gammatone Cepstral Coefficients (GTCC) obtained from noise-removed speech signals with comparative EMD-DWT and DWT-EMD analysis. The proposed model is divided into three stages. In the first step, noise is removed from the signals using the EMD-DWT and DWT-EMD methods. In the second step, the GTCC and MFCC are extracted from the enhanced audio signals. The classification process is carried out in the third step by feeding these features into the LSTM and CNN models, which are designed to define sequential information from the extracted features. The experiments are performed using PC-GITA and Sakar datasets and 10-fold cross validation method, the highest classification accuracy for the Sakar dataset reached 100% for both EMD-DWT-GTCC-CNN and DWT-EMD-GTCC-CNN, and for the PC-GITA dataset, the accuracy is reached 100% for EMD-DWT-GTCC-CNN and 96.55% for DWT-EMD-GTCC-CNN. The results of this study indicate that the characteristics of GTCC are more appropriate and accurate for the assessment of PD than MFCC.
The aim of this publication was to propose a method to determine changes in fatigue in selected muscle groups of the lower extremity during dynamic and cyclical motion performed on a rowing ergometer. The study aimed to use the discrete wavelet transform (DWT) to analyze electromyographic signals (EMG) recorded during diagnostic assessment of muscle and peripheral nerve electrical activity (electroneurography) using an electromyography device (EMG). Methods: The analysis involved implementing calculations such as mean frequency (MNF) and median frequency (MDF) using the reconstructed EMG signal through DWT. The study examined the efficacy of DWT analysis in assessing muscle fatigue after physical exertion. Results: The study obtained a negative regression coefficient for DWT analysis in all muscles except for the right gastrocnemius (GAS). The results suggest that DWT analysis can be an effective tool for evaluating muscle fatigue after physical exertion. Conclusions: The use of DWT in the analysis of EMG signals during rowing ergometer exercises has shown promising results in assessing muscle fatigue. However, additional investigations are necessary to confirm and expand these findings. This publication addresses the literature gap on the determination of muscle fatigue considering motion analysis on a rowing ergometer using the discrete wavelet transform. Previous studies have extensively compared and analyzed methods such as the Fourier transform (FFT), short-time Fourier transform (STFT), and wavelet transform (WT) for muscle fatigue analysis. However, no previous work has specifically examined the assessment of muscle fatigue by incorporating DWT analysis with motion analysis on a rowing ergometer.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The compression of image using analyzing techniques give us q high quality in the reconstructed image however in the case of transmission produce a sensitive (to the channel noise) image .In this paper we are going to use combination between error detection , source and channel coding with unequal distribution in the code rate our approach shows a high efficiency and optimization in the use of the code rate using Whale Algorithm (WA) (minimization in the redundant bits) compared to other approaches. The results of the work carried out in this article are mainly focused on the medical images compression by the (DWT+SPIHT) method, which, in fact, allow a significant reduction for data. We are also interested in the transmission of these images on an channel in a way that can provide a high bit rate with good transmission quality, by exploiting the channel coding technique, which is effective in combating the noise introduced during the transmission of these images.
PL
Kompresja obrazu przy użyciu technik analitycznych daje nam q wysoką jakość rekonstruowanego obrazu, jednak w przypadku transmisji wytwarzamy obraz wrażliwy (na szum kanału). przy nierównym rozkładzie współczynnika kodowania nasze podejście wykazuje wysoką wydajność i optymalizację w wykorzystaniu współczynnika kodowania przy użyciu algorytmu wieloryba (WA) (minimalizacja w nadmiarowych bitach) w porównaniu z innymi podejściami. Wyniki prac przeprowadzonych w niniejszym artykule koncentrują się głównie na kompresji obrazów medycznych metodą (DWT+SPIHT), która w rzeczywistości pozwala na znaczną redukcję danych. Interesuje nas również transmisja tych obrazów na kanale w sposób, który może zapewnić wysoką przepływność przy dobrej jakości transmisji, wykorzystując technikę kodowania kanałów, która skutecznie zwalcza szumy wprowadzane podczas transmisji tych obrazów.
Parallel realizations of discrete transforms (DTs) computation algorithms (DTCAs) performed on graphics processing units (GPUs) play a significant role in many modern data processing methods utilized in numerous areas of human activity. In this paper the authors propose a novel execution time prediction model, which allows for accurate and rapid estimation of execution times of various kinds of structurally different DTCAs performed on GPUs of distinct architectures, without the necessity of conducting the actual experiments on physical hardware. The model can serve as a guide for the system analyst in making the optimal choice of the GPU hardware solution for a given computational task involving particular DT calculation, or can help in choosing the best appropriate parallel implementation of the selected DT, given the limitations imposed by available hardware. Restricting the model to exhaustively adhere only to the key common features of DTCAs enables the authors to significantly simplify its structure, leading consequently to its design as a hybrid, analytically–simulational method, exploiting jointly the main advantages of both of the mentioned techniques, namely: time-effectiveness and high prediction accuracy, while, at the same time, causing mutual elimination of the major weaknesses of both of the specified approaches within the proposed solution. The model is validated experimentally on two structurally different parallel methods of discrete wavelet transform (DWT) computation, i.e. the direct convolutionbased and lattice structure-based schemes, by comparing its prediction results with the actual measurements taken for 6 different graphics cards, representing a fairly broad spectrum of GPUs compute architectures. Experimental results reveal the overall average execution time and prediction accuracy of the model to be at a level of 97.2%, with global maximum prediction error of 14.5%, recorded throughout all the conducted experiments, maintaining at the same time high average evaluation speed of 3.5 ms for single simulation duration. The results facilitate inferring the model generality and possibility of extrapolation to other DTCAs and different GPU architectures, which along with the proposed model straightforwardness, time-effectiveness and ease of practical application, makes it, in the authors’ opinion, a very interesting alternative to the related existing solutions.
Detecting high impedance faults (HIFs) is one of the challenging issues for electrical engineers. This type of fault occurs often when one of the overhead conductors is downed and makes contact with the ground, causing a high-voltage conductor to be within the reach of personnel. As the wavelet transform (WT) technique is a powerful tool for transient analysis of fault signals and gives information both on the time domain and frequency domain, this technique has been considered for an unconventional fault like high impedance fault. This paper presents a new technique that utilizes the features of energy contents in detail coefficients (D4 and D5) from the extracted current signal using a discrete wavelet transform in the multiresolution analysis (MRA). The adaptive neurofuzzy inference system (ANFIS) is utilized as a machine learning technique to discriminate HIF from other transient phenomena such as capacitor or load switching, the new protection designed scheme is fully analyzed using MATLAB feeding practical fault data. Simulation studies reveal that the proposed protection is able to detect HIFs in a distribution network with high reliability and can successfully differentiate high impedance faults from other transients.
This paper presents mechanical fault detection in squirrel cage induction motors (SCIMs) by means of two recent techniques. More precisely, we have analyzed the rolling element bearing (REB) faults in SCIM. Rolling element bearing faults constitute a major problem among different faults which cause catastrophic damage to rotating machinery. Thus early detection of REB faults in SCIMs is of crucial importance. Vibration analysis is among the key concepts for mechanical vibrations of rotating electrical machines. Today, there is massive competition between researchers in the diagnosis field. They all have as their aim to replace the vibration analysis technique. Among them, stator current analysis has become one of the most important subjects in the fault detection field. Motor current signature analysis (MCSA) has become popular for detection and localization of numerous faults. It is generally based on fast Fourier transform (FFT) of the stator current signal. We have detailed the analysis by means of MCSA-FFT, which is based on the stator current spectrum. Another goal in this work is the use of the discrete wavelet transform (DWT) technique in order to detect REB faults. In addition, a new indicator based on the MCSA-DWT technique has been developed in this study. This new indicator has the advantage of expressing itself in the quantity and quality form. The acquisition data are presented and a comparative study is carried out between these recent techniques in order to ensure a final decision. The proposed subject is examined experimentally using a 3 kW squirrel cage induction motor test bed.
10
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule opisano algorytm opracowany do wyznaczania zmienności rytmu serca (tzw. sygnału HRV) na podstawie wartości chwilowych okresu sygnału PPG, który reprezentuje falę tętna obwodowego. Sygnał PPG został zarejestrowany podczas oddziaływania muzyki. Do wydzielenia składowych sygnału HRV (tj. fluktuacji i nieliniowego trendu) zastosowano dyskretną transformatę falkową. Do oceny wpływu muzyki na częstość pracy serca przyjęto parametry opisujące zarówno zmienność fluktuacji rytmu serca, jak i wolnozmiennego trendu.
EN
In this article, the algorithm developed for determination of HRV based on the PPG signal representing the peripheral pulse wave was described. The PPG signal was recorded under the influence of music. The components of HRV signal (i.e. a nonlinear trend and fluctuations) were extracted by using the DWT. The parameters representing variability of the HRV fluctuations as well as trend were applied to assessment of HRV.
11
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The applications of different dimension of multimedia have been grown rapidly on daily basis. Digital media brings about the changes in the conveniences to the people, The cons of this technology is security threat if that security issue exist there is no meaning of conveniences We have segmented the proposed work in such a way unlike conventional approach the module of work includes sub-plotting the image in three directional coordinates plot(x), plot (g), plot(y). The security of information may have the distinct dimension in growing effective techniques to discourage the unauthenticated technique of duplication of virtual signals. Digital watermarking is the mathematical technique of embedding information right into a virtual signal in a way that it is difficult to eliminate. in order to overcome this, robust dwt watermarking approach is proposed. We've contemplated a way a good way to use t the temporary statistics to apply the inversion of dwt in row way and decompose the picture in the same length of width and height. Robustness can be defined as a watermark to stay unaffected even if digital content is passed thru diverse approaches and attacks. we've got proposed invisible sturdy watermarking for you to proved to be is the most accurate method. in conjunction with conventional technique embed a watermark containing key facts consisting of authentication codes.We have considered different dynamic conditions where the copyright and basic constituent of distributed images/photos is violated. Handiest one picture in active mode and all different is inactive mode. Since the images/pics posted on social networks or any networks are normally changed and/or compressed to the original template supplied through the carrier carriers, we suggest a digital photo watermarking based on dwt coefficients modification for use for such snapshots/images. we've got carry out the watermarking with proposed approach for the bilateral method which means for encrypting the digital information and for retrieval of the original information from the encoded dataset have alerted the convention concept of Dwt by adding an Adaptive filter into it for extracting the data.The contemplated approach has reduced the noise and unnecessary constituent to provide better efficiency and retrieve the accurate original image without distorting the pixel vectors.
Cavitation is a common cause of failure in centrifugal pumps. Because of interaction of several mechanical parts and fluid, the vibration signal of a centrifugal pump is complicated. In this paper, the vibrations of a transparent-casing centrifugal pump are studied. Three states are studied experimentally: no cavitation, limited cavitation and developed cavitation. Each case was also confirmed by visually inspecting the cavitation bubbles. The vibrations of the pump was acquired by using an accelerometer that was attached to the casing. Discrete wavelet transform (DWT) analysis and empirical mode decomposition (EMD) are used to extract classification features from the acquired signals. Using these features, an artificial neural network (ANN) successfully diagnosed the cavitation condition of the pump. Finally, EEMD is also implemented. The results showed the success of EMD and DWT in cavitation diagnosis. The output of EEMD does not show significant change comparing to EMD.
Gender recognition, across different races and regardless of age, is becoming an increasingly important technology in the domains of marketing, human-computer interaction and security. Most state-of-the-art systems consider either highly constrained conditions or relatively large databases. In either case, often not enough attention is paid to cross-racial age-invariant applications. This paper proposes a method of hybrid classification, which performs well even with a small training set. The design of the classifier enables the construction of reliable decision boundaries insensitive to an aging model as well as to race variation. For a training set consisting of one hundred images, the proposed method reached an accuracy level of 90%, whereas the best method known from the literature, tested under the restrictions imposed on the database, achieved only 78% accuracy.
Celem artykułu jest prześledzenia historii budowy suszarek węgla brunatnego począwszy od pierwszej połowy XX wieku aż do chwili obecnej. Pozwala to na zrozumienie obecnego stanu rynku komercyjnego tego typu suszarek. Szczegółowo pokazano suszarkę fluidalną na parę przegrzaną prof. Owena Pottera z Monash University w Australii oraz suszarki typu DWT lub DDWT wytwarzane na terenie Niemiec wschodnich w drugiej połowie XX wieku. Pokazano również schematy suszarki MTE z politechniki TU Dortmund. Wymienione typy suszarek wpłynęły na obecny rynek komercyjny. Zarówno suszarka typu WTA firmy RWE (Niemcy) jak i suszarka typu DryFining firmy GRE Energy (USA) korzystają z pewnych idei technicznych zawartych w poprzednich typach suszarek. Pokazano również nietypowe techniczne pomysły różnych badaczy związanych z rynkiem suszarek.
EN
The purpose of the article time to track the history of the construction of lignite dryers from the first half of the 20th century until the present. This allows the reader to understand the current state of the commercial market of lignite dryers. Article presents details showing a fluidized bed dryer constructed through prof. Owen Potter from Monash University in Australia and driers type of DWT or DDWT produced in East Germany in the second half of the 20th century. Article also shows schema of MTE dryer from University of Dortmund. These types of dryers influenced the current commercial market. Both dryer type WTA from company RWE (Germany) and dryer type DryFining from GRE Energy (USA) use certain technical ideas contained in previous types of dryers. Article also shows unusual technical ideas of different researchers related to the dryers constructions, too.
In building speech recognition based applications, robustness to different noisy background condition is an important challenge. In this paper bimodal approach is proposed to improve the robustness of Hindi speech recognition system. Also an importance of different types of visual features is studied for audio visual automatic speech recognition (AVASR) system under diverse noisy audio conditions. Four sets of visual feature based on Two-Dimensional Discrete Cosine Transform feature (2D-DCT), Principal Component Analysis (PCA), Two-Dimensional Discrete Wavelet Transform followed by DCT (2D-DWT-DCT) and Two-Dimensional Discrete Wavelet Transform followed by PCA (2D-DWT-PCA) are reported. The audio features are extracted using Mel Frequency Cepstral coefficients (MFCC) followed by static and dynamic feature. Overall, 48 features, i.e. 39 audio features and 9 visual features are used for measuring the performance of the AVASR system. Also, the performance of the AVASR using noisy speech signal generated by using NOISEX database is evaluated for different Signal to Noise ratio (SNR: 30 dB to -10 dB) using Aligarh Muslim University Audio Visual (AMUAV) Hindi corpus. AMUAV corpus is Hindi continuous speech high quality audio visual databases of Hindi sentences spoken by different subjects.
In this paper authors present a simple method for recognizing blurred regions in the image. Proposed algorithm is based on 81 simple features — moments of histogram of image subbands, that were obtained during image decomposition, and ratio derived from gray level co-occurrence matrix (GLCM) are used. The method is compared with a different method, that is based on approaches found in literature. To increase the efficiency of algorithms, authors combined three solutions (edge-detection, gray level co-occurrence matrix and fast image sharpness). The aim of the research was to verify whether it is possible to use simpler methods of feature extraction to achieve similar, or even better, results.
In this paper some preliminary investigation on combination of watermarking technique with biometric data to increase security of digital images in case of medical images is proposed. Performance of watermarking algorithm, based on discrete wavelet transform (DWT) decomposition, that incorporates biometric watermark is elaborated. The frequency domain were chosen as it is proven, that this domain provides better robustness against attacks and leads to less perceptibility of an embedded watermark. To assure confidentiality of patient data their hand geometry features are embedded instead of patient’s name. Proposed system is evaluated by measuring the similarity between embedded and extracted biometric codes.
18
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
This paper provides a preliminary investigation on digital watermarking as an effective technology to protect property rights and limit distribution of multimedia data. First, crucial properties and design requirements of watermarking schemes are discussed. Then, as watermarking techniques finds many applications in healthcare industry, aspects of medical image watermarking are raised. Nowadays, the transmission of digitized medical information has become very easy due to the generality of Internet. However, the digital form of these images can easily be manipulated and degraded. This causes problems of medical security and copyright protection and poses a great challenge to privacy protection using watermarking techniques.
The detection of vehicles, in video streams from road cameras, is generally performed by analyses of the occupancy of virtual detection fields defined in image frames. This principle of detection is sensitive to ambient light variations, vehicle shadows, and camera movement. The paper presents a method for detection of vehicles that uses transformed image frames. To facilitate detection each frame is converted into a vector of pixel values. Consecutive video vectors are transformed using one-dimensional DWT. Stopped vehicles are represented by stripes, whereas moving ones by checked patches. The width of a stripe indicates vehicle size, while the length shows how long the vehicle waited at the approach to the intersection.
PL
Wykrywanie pojazdów w strumieniu wideo z kamer drogowych oparte jest zwykle na analizie zajętości wirtualnych pól detekcji. Ten sposób wykrywania jest czuły na zmiany oświetlenia, cienie pojazdów i ruchy kamery. Artykuł przedstawia metodę wykrywania, która wykorzystuje transformaty klatek obrazów. W celu umożliwienia sprawnej analizy zawartość klatki zamieniana jest najpierw na wektor wartości pikseli. Kolejne wektory wideo są transformowane z użyciem jednowymiarowego, dyskretnego przekształcenia falkowego. Zatrzymane pojazdy są reprezentowane przez paski, a ruchome przez kratkowane placki. Szerokość paska wskazuje na rozmiar pojazdu, a długość określa, jak długo pojazd oczekiwał na wlocie skrzyżowania.
20
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The paper presents a digital signal processor (DSP) based system for segmentation of speakers of a telephone conversation. The TMS320C6713 DSP by Texas Instruments in real-time watermarks one interlocutor voice and therefore precise segmentation of both conversation sides is made on a PC without any speaker recognition techniques. The authors also solved the problem of data blocks synchronization and beats caused by differences in the digital-to-analog and the analog-to-digital sampling clock frequencies.
PL
Artykuł prezentuje, zrealizowany na procesorze sygnałowym, system do segmentacji mówców rozmowy telefonicznej. Użyto procesora TMS320C6713 firmy Texas Instruments, który podczas rozmowy oznacza znakiem wodnym jednego z rozmówców. Umożliwia to późniejszą separację mówców bez użycia algorytmów ich rozpoznawania. Autorzy dodatkowo rozwiązali problemy związane z synchronizacją bloków danych i dudnieniami wywołanymi różnicą częstotliwości zegarów taktujących przetworniki analogowo-cyfrowe i cyfrowo-analogowe.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.