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.
Feed-forward artificial neural networks (ANNs) have been applied to the diagnosis of mixed-mode electronic circuit. In order to tackle the circuit complexity and to reduce the number of test points hierarchical approach to the diagnosis generation was implemented with two levels of decision: the system level and the circuit level. For every level, using the simulation-before-test (SBT) approach, fault...
This paper presents a method to automatically process the education quality assessment quiz test. The propose technique use Pattern Recognition methodology and the final decision is taken using a MLP neural network. The Xj subject's answers are numerically encoded in a descriptor vector VXj. This vector is fed to the net and it decides the Xj's degree of satisfaction or dissatisfaction over the most...
This paper investigates the effects of using limited precision for efficient implementations of the RBF-M neural network. This architecture employs only simple arithmetic operators and is characterized by simple LMS training in an expanded feature space generated by RBF functions centered around support vectors selected via a simple algorithm. The classification performances of our low complexity,...
In the paper a new approach to a design of fuzzy controllers is proposed, based on an automatic selection of membership functions shapes. The automatization is achieved by optimization using a genetic algorithm relative to a chosen system performance criterion. The software system for fuzzy control of the cart-ball system is also described. The described system is developed in such a way that modules...
The aim of this paper is to show that the data stored in companies data warehouses can be used in order to improve business. By application of data mining method, neural clustering, we investigate age structure of employees and its influence on business companies. This would enable improvement in employment policy for small and medium-sized companies. The criteria while employing new people in retail...
In this paper, the author's previous work is extended and a new neural network is utilized to solve the stability problem of multidimensional systems. In the original authors work the problem is transformed into an optimization problem. Using the DeCarlo-Strintzis Theorem one has to check if |B(Z1,..., 1, Zm)| ≠ 0 for |Z1| = ... = |Zm| = 1 or equivalently if the min |B(Z1, ..., 1, Zm)| is 0 or not,...
Nonnegative tensor factorization (NTF) is a recent multiway (multilinear) extension of negative matrix factorization (NMF), where nonnegativity constraints are mainly imposed on CANDECOMP/PARAFAC model and recently, also, on Tucker model. Nonnegative tensor factorization algorithms have many potential applications, including multiway clustering, multi-sensory or multidimensional data analysis and...
Wireless mesh networks (WMN) represent a type of mobile ad-hoc networks. These networks are very important in providing the Internet access to fixed and mobile terminal equipment. The main problem in WMNs (regarding to mesh routers and mesh clients) is a routing protocol, especially because it has to enable the access to network for both mesh and conventional clients. Access to Internet for wireless...
Influence of the interaction time-delay on the noise induced system size resonance in a system of all-to-all electrically coupled FitzHugh-Nagumo excitable neurons is studied. It is observed that small time-lags decrease and that large time-lags increase the coherence of spiking. Bifurcations of the system's stationary state are used to explain the observed non-monotonic dependence of coherence on...
This work presents the technical aspects for the development of a firmware for a Nd:YAG laser based ophthalmic surgery microscope. The software designed for a Microchip® PIC18F4550 microcontroller interfaces the electronic blocks of the device and allows the calibration of the laser energy attenuator. This paper mainly describes the experimental setup and presents the experimental data for an innovative...
Advancements in digital electronics and signal processing algorithms for various purposes generated the possibility and need for designing artificial neural networks in hardware. The selected platform, FPGA, enables fulfillment of their demands and provides comfortable work and test environment. This paper presents development cycle and specifics for implementation of high speed Hamming artificial...
We investigate performance of neural networks for classification of laser-induced breakdown spectroscopic data of four proteins: Bovine Serum Albumin, Osteopontin, Leptin and Insulin-like Growth Factor II. We utilize principal component analysis algorithm for feature extraction and multilayer perceptrons algorithms with one and two hidden layers. We employ leave-one-out procedure for classifier evaluation...
This paper addresses the problem of vowels recognition in patients after total laryngectomy using combined visual and acoustic features. The linear prediction coefficients were estimated from speech signal using weighted recursive least squares algorithm. Ten cross-sectional areas of vocal tract model were calculated. Face expression parameters related to the spoken vowel were extracted from video...
This paper presents investigation on speech recognition classification performance when using different standard neural networks structures as a classifier. Those cases include usage of a Feed-forward Neural Network (NN) with back propagation algorithm and a Radial Basis Functions (RBF) Neural Network.
In this paper we describe a database, noted as RadEch Database, containing radar echoes from various targets. The data has been collected in controlled test environments at the premises of Military Academy - Republic of Serbia. Our goal is to provide a balanced and comprehensive database to enable reproducible research results in the field of classification of ground moving targets (pattern recognition)...
This paper proposes novel approach in coding single phonemes based on mel-frequency cepstral coefficients (MFCC) in order to simplify the neural network used to recognize those phonemes. The efficiency and effectiveness of proposed algorithm are demonstrated for both male and female speakers.
A fundamental challenge in today's arena of complex systems is the design and development of accurate and robust signal processing methods. These methods should be capable to adapt quickly to unexpected changes in the data and operate under minimal model assumptions. Systems in Nature also do signal processing and often do it optimally. Therefore, it makes much sense to understand what Nature does...
Consistent Boolean generalization of two-valued into a real-valued theory means preservation of all of its algebraic - value indifferent characteristics: Boolean axioms and theorems. Actually two-valued theories in Boolean frame (classical logic, theory of classical sets, theory of classical relations, etc.) are based on the celebrated two-valued realization of Boolean algebra (BA) and their real-valued...
The efficiency of a human face recognition system depends on the capability of face recognition in presence of different changes in the appearance of face. One of the main difficulties regarding the face recognition systems is to recognize face in different views and poses. In this paper we propose a new algorithm which utilizes the combination of texture and depth information to overcome the problem...
This paper discusses the problem of object area detection of video frames. The goal is to design a pixel accurate detector for grass, which could be used for object adaptive video enhancement. A boosting neural network is used for creating such a detector. The resulted detector uses both textural features and color features of the frames.
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.