The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Communication using brain–computer interfaces (BCIs) can be non-intuitive, often requiring the performance of a conversation-irrelevant task such as hand motor imagery. In this paper, the reliability of electroencephalography (EEG) signals in discriminating between different covert speech tasks is investigated. Twelve participants, across two sessions each, were asked to perform multiple iterations...
The expanding use of deep learning algorithms causes the demands for accelerating neural network (NN) signal processing. For the NN processing, in-memory computation is desired, in which expensive data transfer can be eliminated. In reflection of recently proposed binary neural networks (BNNs), which can reduce the computation resource and area requirements, we designed an in-memory BNN signal processor...
This paper develops new designs for recommender systems inspired by recent advances in graph signal processing. Recommender systems aim to predict unknown ratings by exploiting the information revealed in a subset of user-item observed ratings. Leveraging the notions of graph frequency and graph filters, we demonstrate that a common collaborative filtering method — fc-nearest neighbors — can be modeled...
In this study, classification of 11 different Power Quality (PQ) disturbances with Artificial Neural Networks (ANN) has been done by using the attributes obtained with S-Transform. It was aimed to achieve accurate and high classification performance in noisy environment by using the least number of attributes representing PQ disturbances. The most suitable ones from the attributes were selected by...
Over the last few years, neural networks with values in multidimensional domains have been intensely studied. This paper introduces octonion-valued bidirectional associative memories, for which the states, outputs, weights and thresholds are all octonions. The octonion algebra represents a non-associative normed division algebra which generalizes the complex and quaternion algebras and doesn't fall...
In this study, the features of the seminiferous tubule sections were extracted and the presence of the cells and cell stain types detected with the help of the feed forward artificial neural network. By looking at the section view with a small window, 78 features were extracted from the pixels seen by the window and used as an input to the artificial neural network. Artificial neural network outputs...
Concrete can be molded to any shape and size, and once hardened it can withstand tremendous amount of compressive loads. This ability of concrete makes it the most widely used material in construction and thus, a need for identification and prediction of its compressive strength. Nondestructive tests have been solely preferred for this purpose and a drop-impact test machine prototype named; Material...
This work present new parameters based on biometrie handwritten information for the writer identification. The feature extraction is developed by new algorithms based on image processing techniques. The handwritten parameters will be classified by artificial neural network and fusion strategy in order to increase the accuracy. After experiments, and using a dataset composed by 100 writers, this proposal...
One of the most challenging problems in the field of digital image processing is image denoising. When processing medical images, it is of particular relevance to improve the perceived quality of images, while preserving the diagnostically relevant information. This paper investigates the capacity of a neural network framework for medical image denoising. Specifically, the performance of the proposed...
Efficient compression of medical images is needed to decrease the storage space and enable efficient image transfer over network for access of electronic patient records. Since the medical images contain diagnostically relevant information, it is necessary for the process of image compression to preserve high levels of image fidelity, especially when the images are compressed at low bit rates. This...
For cardiologists, the detection of cardiac abnormalities is a very delicate and crucial task for the treatment of a patient's condition. This task that requires electronic systems of medical assistance that is more precise, faster and reliable to help cardiologists to analyze and make the right decisions. These medical assistance systems tend to model the human expertise and perception using signal...
An Artificial neural network (ANN) is parallel Information processing structure consists of processing units. The processing unit decides while the network is efficient or not. So need to design an efficient processing unit and it also provide better performance. The processing unit consists of MAC unit (Multiplication and Accumulation) and Activation unit. In an existing system, the processing MAC...
Correlation size together with Lyapunov exponents estimated from both electroencephalography (EEG) and electromyography (EMG) signals, are the crucial variables in the classification of mental tasks using an artificial neural network (ANN) classifier for patients suffering from neurological disorders/diseases. The above parameters vary according to the status of the patient, for example: depending...
Sparse representations have been found to provide high classification accuracy in many fields. Their drawback is the high computational load. In this work, we propose a novel cascaded classifier structure to speed up the decision process while utilizing sparse signal representation. In particular, we apply the cascaded decision process for noise robust automatic speech recognition task. The cascaded...
Objects that have been buried underground cannot be recognized due to the opaqueness of the soil. To recognize objects that have been buried, ground penetrating radar (GPR) by the assistance of computer-aided system was used. This paper proposes the latter, which is called the Recognition System of Underground Object Shape using GPR datagram. The hyperbola from cylinder and cube metal object that...
An Artificial Neural Network (ANN) based approach is carried out for power system unsymmetrical fault classification and localization using Continuous Wavelet Transform (CWT) in Hybrid Distributed Generation (HDG) System. In this study, CWT is used as a signal processing tool to extract features of HDG System current signals captured from distribution substation. The extracted features are applied...
Given the importance of an accurate wind speed forecasting for efficient utilization of wind farms, and the volatile nature of wind speed data including its non-linear and uncertain nature, the wind speed forecasting has remained an active field of research. In this study, the non-linearity of wind speed is tackled using artificial neural network and its uncertainty by wavelet transform. To avoid...
In this thesis, a single, separate section, for example, is used to forecast the traffic flow in a long time. The advantage of artificial neural network is its ability of learning or training in other words. By learning, the network can give appropriate output when accepting input. Thus, artificial neural network is a good model for predicting transportation flow. This paper proposes the Bayesian...
These last years, artificial neural networks (ANN) have known a renewed interest since efficient training procedures have emerged to learn the so called deep neural networks (DNN), i.e. ANN with at least two hidden layers. In the same time, the computational auditory scene recognition (CASR) problem which consists in estimating the environment around a device from the received audio signal has been...
Hybrid signal processing technique is discussed in this paper to sense the fault in a 11 kV, 30 km distribution line with R-L load placed at the receiving end. The proposed method uses 1 cycle post fault voltage and current signal wave form sending end of the system under study. Further preprocessing of the collected signal is done by wavelet packet transform and discrete wavelet transform which includes...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.