Search

Data security concerns safeguarding the data from unauthorized users. Numerous research works have been conducted using Reversible Data Hiding (RDH) technique, but not been found to be fooproof. In this paper, a novel reversible data hiding method, based on an Artificial Neural Network, called Histogram Transpose and Naor Pseduorandom Arnold (HT-NPA) data security system for handling health care reports and securely transferring patient information, is proposed to evaluate the embedding capacity, improve the visual quality of the recovered image and ensure security via duplex scramble. This entire process is performed by means of an Artificial Neural Network (i.e. multilayer perceptron). The experimental results reveal that the HT-NPA performs better with 29%, 16%, 19%, and 9% improvement in embedding rate, PSNR, SSIM, entropy, and NPCR, and 26% reduction of bit error rate, respectively for retrieved medical images compared to existing works. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) can track the brain waves which contain the neural activity of the brain. EEG signals help to understand the physiological and functional details and activities of the brain. In the era of Artificial Intelligence (AI), machine learning algorithms were useful in brain disorder detection and classification. Recently, a rapid increase in using Deep Learning (DL) methods in various applications in EEG signals not only helps in the detection of brain disorders but also facilitates the recognition of human emotions and various psycho-neuro disorders. In order to offer a beneficial and broad perspective, a detailed survey on the application of deep learning architecture in EEG signals has been carried out in this paper. Different deep learning methods, using varied architecture in EEG signal analysis, offer an understanding to develop the next level of AI-based systems. This review will provide information about how deep learning methods are used in EEG signals and the challenges and limitations of each method in classification; moreover making it helpful for those who are exploring EEG signals using DL algorithms. [ABSTRACT FROM AUTHOR]

In this paper, the design of a novel efficient QCA-based universal logic gate (ULG) is presented. Compact and efficient implementation layout with only one clock phase delay and also the implementation of the 13 standard functions without extra ULG or gate are the advantages of the proposed ULG. The results show that the proposed ULG is an efficient tool in designing any QCA circuit. Also, an accurate artificial neural network (ANN) model of the proposed ULG is presented. This model can be simply imported into any circuit simulators such as HSPICE, MATLAB Simulink and etc., for modeling and simulation of QCA circuits at the Gate level with high precision in a short time. [ABSTRACT FROM AUTHOR]

This paper emphasizes on the hand movement recognition using artificial neural network (ANN). It shows the methodology that analyses and classifies the electromyography (EMG) signals using neural networks to control multifunction prostheses. The control of these prostheses can be made using myoelectric signals taken from surface electrodes. The linear neural network was employed in the EMG signal classification system as EMG signals contain many artifacts and noise, so linear classification helps to give the accurate classification results which can be further used for myoelectric controls. The experimental results show a promising performance in the classification of hand movements based on EMG pattern. The hand movements used were Flexion and Extension of wrist, Flexion and Extension of elbow, Closing and opening of hand, and the Supination and Pronation of forearm. [ABSTRACT FROM AUTHOR]

Antinuclear antibody (ANA) testing is best performed using the indirect immunofluorescence (IIF) method with human epithelial type-2 (HEp-2) cells as the substrate. IIF is a subjective procedure in which HEp-2 patterns are analyzed manually from the microscope. Therefore, ANA test results greatly rely on the experience and expertise of pathologists. Hence, complete automation of the ANA test is required to avoid incorrect diagnoses. This paper represents an algorithm for the complex HEp-2 cell classification problem. The proposed algorithm used a small hybrid feature set that characterizes the texture and morphology of the HEp-2 cells along with artificial neural network (ANN). The hybrid features were extracted by breaking up the image into eight binary images. The proposed hybrid descriptors were more efficient than the popular co-occurrence matrix descriptor and local binary pattern descriptors for texture analysis. The proposed algorithm was evaluated on the ICPR 2016 IIF HEp-2 cell image dataset. The results indicated that the hybrid descriptor with an ANN approach achieved improved performance, with "96.8%" mean class accuracy. [ABSTRACT FROM AUTHOR]

In this paper, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) are used for the modelling and simulation of quantum-dot cellular automata (QCA) circuits. For this purpose, all QCA basis components and gates are modelled using ANN and ANFIS. The accuracy and performance of the proposed methods are analysed through a few circuits. Finally, we compared the simulation speed of the proposed methods with QCADesigner software. The results show that the proposed ANN and ANFIS models are much faster than QCADesigner. Also, these models are imported into HSPICE software to design and simulate complex QCA logic circuits. [ABSTRACT FROM AUTHOR]

In this paper, two well-known artificial intelligence techniques, artificial neural network (ANN) and upport vector machine (SVM) have been modelled for fault type classification for a transmission line equipped with thyristor controlled series compensator (TCSC) to compare their performance. Due to the non-linear changes introduced by the compensation devices, protection of the series compensated transmission line has been a difficult task. The problem is still complex for advanced series compensated transmission lines. The relaying procedure for a compensated transmission line usually involves fault type identification and fault section identification to conclude for the final decision. The algorithm modelled in this paper has been devised on the analysis of a half cycle post-fault current signals, with the help of discrete wavelet transform as a feature extraction tool. The training and testing of ANN and SVM have been performed with an identical fault data set for comparison. The accuracy of the classifiers had been tested with a large number of fault cases encircling different system conditions and fault conditions. These system conditions are simulated on Power System Computer Aided Design/ElectroMagnetic Transients Program (PSCAD/EMTP) with variation in system parameters such as change in firing angle of TCSC, fault resistance, fault inception angle, loading angle of generator, and source impedance variation for all 10 types of faults. The classifiers are compared in detail with their performance at various firing angles of TCSC and for different fault types. [ABSTRACT FROM PUBLISHER]

Hand gesture recognition system can be a particularly challenging task because of the presence of several spatio-temporal variations like hand trembling, self co-articulation within a gesture, etc. during the segmentation stage. In this paper, we focus our attention on self co-articulation problem which has not been addressed before. With the addition of a new feature, i.e. removing the hand movement during pause state along with the velocity features within the gestures, we were able to perform the gesture spotting more smoothly for the isolated gestures. Moreover, a new set of novel features was added in the feature extraction stage: (1) number of self co-articulated stroke; (2) orientation of the self co-articulated strokes; (3) position of the hand; (4) distance between the start and end points of a gesture; and (5) ratio between the longest and shortest distances from centre. These features were fed to neural network classifier. The results of experimentation were carried out with 40 gestures and it showed an average true positive rate of 91.25%. [ABSTRACT FROM PUBLISHER]

This paper deals with an artificial neural network based control of the voltage and frequency (VF) of autonomous wind power generation using an isolated asynchronous generator (IAG) feeding three-phase four-wire loads. The reference generator currents are estimated using the adaptive linear neuron and its training through least mean square (LMS) algorithm to control the VF of an IAG system. Three-leg voltage source converter (VSC) with an isolated T-connected transformer is used as an integrated VSC. The integrated VSC with a battery energy storage system is used to control the active power of the wind energy conversion system (WECS). Bi-directional power flow capabilities of the proposed VF controller in WECS are demonstrated through simulation results. The proposed controller functions as a VF regulator, a load leveler, a load balancer and a harmonic eliminator in the WECS. The WECS is modeled and simulated in the MATLAB using the Simulink and the sim power system toolboxes. [ABSTRACT FROM AUTHOR]