Showing 14 results for Fea
A. Ghaffari, M. R. Homaeinezhad, M. Akraminia,
Volume 6, Issue 1 (3-2010)
Abstract
The aim of this study is to address a new feature extraction method in the area of the heart arrhythmia classification based on a metric with simple mathematical calculation called Curve-Length Method (CLM). In the presented method, curve length of the under study excerpted segment of signal is considered as an informative feature in which the effect of important geometric parameters of the original signal can be found. To show merits of the presented method, first the original electrocardiogram (ECG) in lead I is pre-processed by removing its baseline wander then by scaling it in the [-1,1] interval. In the next step, using a trous method, discrete wavelet scales 23 and 24 and smoothing function scale 22 are extracted. Afterwards, segments including samples of the QRS complex, P and T waves are estimated via an approximation criterion and CLM is implemented to extract corresponding features from aforementioned scales, smoothing function and also from each original segment. The resulted feature vector (including 12 components) is used to tune an Adaptive Network Fuzzy Inference System (ANFIS) classifier. The presented strategy is applied to classify four categories found in the MIT-BIH Arrhythmia Database namely as Atrial Premature Beat (APB), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB) and Premature Ventricular Contraction (PVC) and average values of Se = 99.81%, P+ = 99.80%, Sp = 99.81% and Acc = 99.72% are obtained for sensitivity, positive predictivity, specifity and accuracy respectively showing marginal improvement of the heart arrhythmia classification performance.
M. Aghamohammadi, S. S. Hashemi, M. S. Ghazizadeh,
Volume 7, Issue 1 (3-2011)
Abstract
This paper presents a new approach for estimating and improving voltage stability margin from phase and magnitude profile of bus voltages using sensitivity analysis of Voltage Stability Assessment Neural Network (VSANN). Bus voltage profile contains useful information about system stability margin including the effect of load-generation, line outage and reactive power compensation so, it is adopted as input pattern for VSANN. In fact, VSANN establishes a functionality for VSM with respect to voltage profile. Sensitivity analysis of VSM with respect to voltage profile and reactive power compensation extracted from information stored in the weighting factor of VSANN, is the most dominant feature of the proposed approach. Sensitivity of VSM helps one to select most effective buses for reactive power compensation aimed enhancing VSM. The proposed approach has been applied on IEEE 39-bus test system which demonstrated applicability of the proposed approach.
M. R. Homaeinezhad, E. Tavakkoli, A. Afshar, A. Atyabi, A. Ghaffari,
Volume 7, Issue 2 (6-2011)
Abstract
The paper addresses a new QRS complex geometrical feature extraction technique as well as its application for electrocardiogram (ECG) supervised hybrid (fusion) beat-type classification. To this end, after detection and delineation of the major events of ECG signal via a robust algorithm, each QRS region and also its corresponding discrete wavelet transform (DWT) are supposed as virtual images and each of them is divided into eight polar sectors. Then, the curve length of each excerpted segment is calculated and is used as the element of the feature space. To increase the robustness of the proposed classification algorithm versus noise, artifacts and arrhythmic outliers, a fusion structure consisting of three Multi Layer Perceptron-Back Propagation (MLP-BP) neural networks with different topologies and one Adaptive Network Fuzzy Inference System (ANFIS) were designed and implemented. To show the merit of the new proposed algorithm, it was applied to all MIT-BIH Arrhythmia Database records and the discrimination power of the classifier in isolation of different beat types of each record was assessed and as the result, the average accuracy value Acc=98.27% was obtained. Also, the proposed method was applied to 8 number of arrhythmias (Normal, LBBB, RBBB, PVC, APB, VE, PB, VF) belonging to 19 number of the aforementioned database and the average value of Acc=98.08% was achieved. To evaluate performance quality of the new proposed hybrid learning machine, the obtained results were compared with similar peer-reviewed studies in this area.
D. S. Javan, H. Rajabi Mashhadi,
Volume 7, Issue 4 (12-2011)
Abstract
Deregulation of power system in recent years has changed static security assessment to the major concerns for which fast and accurate evaluation methodology is needed. Contingencies related to voltage violations and power line overloading have been responsible for power system collapse. This paper presents an enhanced radial basis function neural network (RBFNN) approach for on-line ranking of the contingencies expected to cause steady state bus voltage and power flow violations. Hidden layer units (neurons) have been selected with the growing and pruning algorithm which has the superiority of being able to choose optimal unit’s center and width (radius). A feature preference technique-based class separability index and correlation coefficient has been employed to identify the relevant inputs for the neural network. The advantages of this method are simplicity of algorithm and high accuracy in classification. The effectiveness of the proposed approach has been demonstrated on IEEE 14-bus power system.
F. Tootoonchian, K. Abbaszadeh, M. Ardebili,
Volume 8, Issue 3 (9-2012)
Abstract
Resolvers are widely used in electric driven systems especially in high precision servomechanisms. Both encapsulated and pancake resolvers suffer from a major drawback: static eccentricity (SE). This drawback causes a significant increase in resolver output position error (RPE) which could not be corrected electronically. To reduce RPE, this paper proposes a novel structure with axial flux. Proposed topology, design guidelines, optimization procedure and several key features to improve the sensitivity of axial flux resolver (AFR) against SE are studied. Furthermore, to minimize RPE an optimized design is attained. The machines are investigated in detail by using d-q model and 3D time stepping finite-element analysis. The results of theses two methods are compared and both prototype machines (proposed and optimized) are built. In order to evaluate proposed topologies, an experimental test setup is devised. Finally, the experimental results of the prototype machines verified the analysis results.
M. Bashirpour, M. Geravanchizadeh,
Volume 12, Issue 3 (9-2016)
Abstract
Automatic recognition of speech emotional states in noisy conditions has become an important research topic in the emotional speech recognition area, in recent years. This paper considers the recognition of emotional states via speech in real environments. For this task, we employ the power normalized cepstral coefficients (PNCC) in a speech emotion recognition system. We investigate its performance in emotion recognition using clean and noisy speech materials and compare it with the performances of the well-known MFCC, LPCC, RASTA-PLP, and also TEMFCC features. Speech samples are extracted from the Berlin emotional speech database (Emo DB) and Persian emotional speech database (Persian ESD) which are corrupted with 4 different noise types under various SNR levels. The experiments are conducted in clean train/noisy test scenarios to simulate practical conditions with noise sources. Simulation results show that higher recognition rates are achieved for PNCC as compared with the conventional features under noisy conditions.
S. Tannaz, T. Sedghi,
Volume 14, Issue 2 (6-2018)
Abstract
In this article, a fabulous method for database retrieval is proposed. The multi-resolution modified wavelet transform for each of image is computed and the standard deviation and average are utilized as the textural features. Then, the proposed modified bit-based color histogram and edge detectors were utilized to define the high level features. A feedback-based dynamic weighting of shape, color and textural features composition produce a resistant feature vectors for image retrieval and recall. A comprehensive and unified matching scheme based on matrix error rate technique was accomplished for similarity of image and retrieval procedure. The feature vectors size in our algorithm is the least one evaluated to the different techniques. Furthermore, the calculation time of previously published techniques is much more than the presented algorithm which is a benefit in proposed retrieval method. The experimental results illustrates that novel algorithm obtains more precious in retrieval and the efficiency in evaluating with the other techniques and algorithms at Corel color image database.
M. Naderan, E. Namjoo, S. Mohammadi,
Volume 15, Issue 3 (9-2019)
Abstract
Social networks have become the main infrastructure of today’s daily activities of people during the last decade. In these networks, users interact with each other, share their interests on resources and present their opinions about these resources or spread their information. Since each user has a limited knowledge of other users and most of them are anonymous, the trust factor plays an important role on recognizing a suitable product or specific user. The inference mechanism of trust in social media refers to utilizing available information of a specific user who intends to contact an unknown user. This mostly occurs when purchasing a product, deciding to have friendship or other applications which require predicting the reliability of the second party. In this paper, first the raw data of the real world dataset, Epinions, is examined, and the feature vector is calculated for each pair of social network users. Next, fuzzy logic is incorporated to rank the membership of trust to a specific class, according to two-, three- and five-classes classification. Finally, to classify the trust values of users, three machine learning techniques, namely Support Vector Machine (SVM), Decision Tree (DT), and k-Nearest Neighbors (kNN), are used instead of traditional weighted sum methods, to express the trust between any two users in the presence of a special pattern. The results of simulation show that the accuracy of the proposed method reaches to 91%, and unlike other methods, does not decrease by increasing the number of samples.
A. Amiri, S. Mirzakuchaki,
Volume 16, Issue 3 (9-2020)
Abstract
Watermarking has increased dramatically in recent years in the Internet and digital media. Watermarking is one of the powerful tools to protect copyright. Local image features have been widely used in watermarking techniques based on feature points. In various papers, the invariance feature has been used to obtain the robustness against attacks. The purpose of this research was based on local feature-based stability as the second-generation of watermarking due to invariance feature to achieve robustness against attacks. In the proposed algorithm, initially, the points were identified by the proposed function in the extraction and Harris and Surf algorithms. Then, an optimal selection process, formulated in the form of a Knapsack problem. That the Knapsack problem algorithm selects non-overlapping areas as they are more robust to embed watermark bits. The results are compared with each of the mentioned feature extraction algorithms and finally, we use the OPAP algorithm to increase the amount of PSNR. The evaluation of the results is based on most of the StirMark criterion.
A. Fattahi, S. Emadi,
Volume 16, Issue 4 (12-2020)
Abstract
Increased popularity of digital media and image editing software has led to the spread of multimedia content forgery for various purposes. Undoubtedly, law and forensic medicine experts require trustworthy and non-forged images to enforce rights. Copy-move forgery is the most common type of manipulation of digital images. Copy-move forgery is used to hide an area of the image or to repeat a portion in the same image. In this paper, a method is presented for detecting copy-move forgery using the Scale-Invariant Feature Transform (SIFT) algorithm. The spearman relationship and ward clustering algorithm are used to measure the similarity between key-points, also to increase the accuracy of forgery detection. This method is invariant to changes such as rotation, scale change, deformation, and light change; it falls into the category of blind forgery detection methods. The experimental results show that with its high resistance to apparent changes, the proposed method correctly detects 99.56 percent of the forged images in the dataset and reveals the forged areas.
B. Nasersharif, N. Naderi,
Volume 17, Issue 2 (6-2021)
Abstract
Convolutional Neural Networks (CNNs) have been shown their performance in speech recognition systems for extracting features, and also acoustic modeling. In addition, CNNs have been used for robust speech recognition and competitive results have been reported. Convolutive Bottleneck Network (CBN) is a kind of CNNs which has a bottleneck layer among its fully connected layers. The bottleneck features extracted by CBNs contain discriminative and rich context information. In this paper, we discuss these bottleneck features from an information theory viewpoint and use them as robust features for noisy speech recognition. In the proposed method, CBN inputs are the noisy logarithm of Mel filter bank energies (LMFBs) in a number of neighbor frames and its outputs are corresponding phone labels. In such a system, we showed that the mutual information between the bottleneck layer and labels are higher than the mutual information between noisy input features and labels. Thus, the bottleneck features are a denoised compressed form of input features which are more representative than input features for discriminating phone classes. Experimental results on the Aurora2 database show that bottleneck features extracted by CBN outperform some conventional speech features and also robust features extracted by CNN.
A. Jabbari,
Volume 18, Issue 2 (6-2022)
Abstract
Low-speed brushless permanent magnet machines are ideal for use in gearless propulsion systems. It is important to provide a precise analytical model to determine the performance characteristics of these machines. One of the challenges in designing permanent magnet machines is the elimination of the pulsating torque due to the presence of cogging torque and torque ripple components. The use of dummy slots (auxiliary teeth) is one of the most common methods of reducing pulsating torque phenomenon. In this paper, an accurate two-dimensional analytical model for calculating the magnetic vector potential in brushless permanent magnet machines is presented, taking into account the effect of stator slots, stator dummy slots, the magnetic direction of permanent magnets and phase winding style. The proposed analytical method is based on solving Laplace’s and Poisson’s equations using the separation of variables method for given regions in the subdomain approach. In the proposed method, to achieve a simpler analytical model, by changing the variable, the polar coordinate system is converted to a quasi-Cartesian coordinate system. Therefore, in mathematical terms, the hyperbolic functions are used instead of exponential ones. To validate the proposed model accuracy, the performance of a 14 kW low-speed brushless permanent magnet motor is calculated analytically and compared with the results of the numerical method and the experimental tests. Comparison of the performance results of this motor shows the consistency of analytical, numerical, and experimental results.
Ali Jabbari, Ali Badran,
Volume 19, Issue 3 (9-2023)
Abstract
Cost reduction, increased efficiency and reliability, extended service life, reduced noise and vibration, and environmental friendliness are critical for new generation wind turbines and electric vehicles. Segmented Hybrid Permanent Magnet (SHPM) machines, on the other hand, which are primarily segmented PMs combined with different materials, dimensions, and magnetization directions, offer a way to meet these needs. In this study, we present nine topologies of segmented PM-rotor SHPM generators based on the Taguchi experimental design method, while presenting a simple and accurate model based on subdomain method for estimating the magnetic performance characteristics of SHPM machines. An analytical model is provided. Magnetic partial differential equations (MPDEs) are represented in a pseudo-Cartesian coordinate system, and with appropriate boundary conditions (BC) and interface conditions (IC), the general solution and its Fourier coefficients are extracted using a variable separation approach. The performance characteristics of nine of the SHPM machines studied were compared semi-analytically and numerically. Two prototype SHPM machines were manufactured and semi-analytical modeling results were compared with finite element analysis (FEA) methods and experimental testing (load mode) on a generator. The FEA simulation and experimental test results have a maximum error rate of about 3, confirming the high accuracy of the provided semi-analytical model. We compare the induced voltage, torque ripple and magnetic torque among the investigated topologies.
Mohammad Hasheminejad,
Volume 19, Issue 4 (12-2023)
Abstract
The Nonparametric Speech Kernel (NSK), a nonparametric kernel technique, is presented in this study as a novel way to improve Speech Emotion Recognition (SER). The method aims to effectively reduce the size of speech features to improve recognition accuracy. The proposed approach addresses the need for efficient and compact low-dimensional features for speech emotion recognition. Having acknowledged the intrinsic distinctions between speech and picture data, we have refined the Kernel Nonparametric Weighted Feature Extraction (KNWFE) formulation to suggest NSK, which is especially intended for speech emotion identification. The output of NSK can be used as input features for deep learning models such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), or hybrid architectures. In deep learning, NSK can also be used as a kernel function for kernel-based methods such as kernelized support vector machines (SVM) or kernelized neural networks. Our tests demonstrate that NSK outperforms current techniques, outperforming the best-tested approach by 5.02% and 3.05%, respectively, with an average accuracy of 96.568% for the Persian speech emotion dataset and 82.56% for the Berlin speech emotion dataset.
