Yazar "Ozerdem, Mehmet Sirac" seçeneğine göre listele
Listeleniyor 1 - 20 / 35
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Analysis of Permanent Magnet Synchronous Motor Current in Healthy and Short Circuit Failure Cases With Discrete Wavelet Transform(Ieee, 2019) Lale, Timur; Ozerdem, Mehmet Sirac; Gumus, BilalIn permanent magnet synchronous motors, stator inter-turn short circuit fault caused by insulation in windings causes an overcurrent in phase winding. Purpose of this study, discrete wavelet transform technique is used to diagnose stator inter-turn short circuit fault in permanent magnet synchronous motor. In this study, stator current in healthy and two different fault conditions was evaluated. Variation of the detail components of the stator current obtained by the wavelet analysis according to the stator short circuit fault intensity was investigated. Stator phase currents d5, d6, d7, d8 and d9 detail coefficients in the frequency band by looking at the change in the average peak amplitude stator winding has been found to contain information about the occurrence of a short circuit fault.Öğe Application of ANN in the prediction of the pore concentration of aluminum metal foams manufactured by powder metallurgy methods(Springer London Ltd, 2008) Ozan, Sermin; Taskin, Mustafa; Kolukisa, Sedat; Ozerdem, Mehmet SiracIn this work, the effect of fabrication parameters on the pore concentration of aluminum metal foam, manufactured by the powder metallurgy process, has been studied. The artificial neural network (ANN) technique has been used to predict pore concentration as a function of some key fabrication parameters. Aluminum metal foam specimens were fabricated from a mixture of aluminum powders (mean particle size 60 mu m) and NaCl at 10, 20, 30, 40(wt)% content under a pressure of 200, 250, and 300 MPa. All specimens were then sintered at 630 degrees C for 2.5 h in argon atmosphere. For pore formation (foaming), sintered specimens were immersed into 70 degrees C hot running water. Finally, the pore concentration of specimens was recorded to analyze the effect of fabrication parameters (namely, NaCl ratio, NaCl particle size, and compacting pressure) on the foaming behavior of compacted specimens. It has been recorded that the above-mentioned fabrication parameters are effective on pore concentration profile while pore diameters remain unchanged. In the ANN training module, NaCl content (wt)%, NaCl particle size (mu m), and compacting pressure (MPA) were employed as inputs, while pore concentration % (volume) of compacts related to fabrication parameters was employed as output. The ANN program was successfully used to predict the pore concentration % (volume) of compacts related to fabrication parameters.Öğe Application of Higuchi's Fractal Dimension for the Statistical Analysis of Human EEG Responses to Odors(Ieee, 2018) Seker, Mesut; Ozerdem, Mehmet SiracThe purpose of this study is to evaluate the fractal dimension (FD) analysis of pleasant-unpleasant odor EEGs. Higuchi's fractal dimension (HFD), which calculates FD directly in time domain, was preferred as a nonlinear signal analysis method. EEG recordings of 6 males were collected for dataset. Results showed that HFD values of pleasant and unpleasant odor EEGs were not in normal distribution and variances between independent variables were not homogeneous in all EEG channels. HFD values of pleasant odors were significantly different from unpleasant ones in O2 (right occipital) and T7 (left temporal) channels. There was a strong correlation between frontal lobes (F3-F4) for all class of odors. HFD values of unpleasant odors were higher than pleasant odors. It showed that unpleasant odor EEGs have more complex patterns than pleasant ones. The proposed work was performed with statistical tools.Öğe Artificial Neural Network approach to predict mechanical properties of hot rolled, nonresulfurized, AISI 10xx series carbon steel bars(Elsevier Science Sa, 2008) Ozerdem, Mehmet Sirac; Kolukisa, SedatIn this study, Artificial Neural Network approach to predict mechanical properties of, hot rolled, nonresulfurized, AISI 10xx series carbon steel bars were obtained using a back-propagation neural network that uses gradient descent learning algorithm. In Artificial Neural Network training module, C%, Si%, Mn% contents were employed as input and tensile strength, yield strength, elongation, reduction in area, hardness were used as outputs. ANN system was trained using the prepared training set (also known as learning set). After training process, the test data were used to check system accuracy. As a result the neural network was found successful for the prediction of mechanical properties of, hot rolled, nonresulfurized, AISI 10xx series carbon steels under given conditions. (C) 2007 Elsevier B.V. All rights reserved.Öğe Artificial neural network approach to predict the electrical conductivity and density of Ag-Ni binary alloys(Elsevier Science Sa, 2008) Ozerdem, Mehmet Siracin this study, artificial neural network (ANN) approach was done to predict electrical conductivity and density of silver-nickel binary alloys using aback-propagation neural network that uses gradient descent learning algorithm. in ANN training module, Ag%. and Ni% (weight) contents were employed as input and electrical conductivity, calculated and typical density were used as outputs. ANN system was trained using the prepared training set (also known as learning set). After training process, the test data were used to check system accuracy. As a result the neural network was found successful for the prediction of electrical conductivity and density of silver nickel binary alloys. (C) 2008 Elsevier B.V. All rights reserved.Öğe Artificial neural network approach to predict the mechanical properties of Cu-Sn-Pb-Zn-Ni cast alloys(Elsevier Sci Ltd, 2009) Ozerdem, Mehmet Sirac; Kolukisa, SedatIn this study, an artificial neural network approach is employed to predict the mechanical properties of Cu-Sn-Pb-Zn-Ni cast alloys. In artificial neural network (ANN), multi layer perceptron (MLP) architecture with back-propagation algorithm is utilized. In Artificial Neural Network training module, Cu-Sn-Pb-Zn-Ni (wt%) contents were employed as input while yield strength, tensile strength and elongation were employed as outputs. ANN system was trained using the prepared training set (also known as learning set). After training process, the test data were used to check system accuracy, As a result of the study neural network was found successful for the prediction of yield strength, tensile strength and elongation of Cu-Sn-Pb-Zn-Ni alloys. (C) 2008 Elsevier Ltd. All rights reserved.Öğe Autoencoders Based Deep Learning Approach for Focal-Nonfocal EEG Classification Problem(Ieee, 2019) Seker, Mesut; Ozerdem, Mehmet SiracEEG markers are the records of brain electrical activity which gives meaningful notice about individuals status. Detection of neurological diseases is only possible with effective analysis of EEG records. Epilepsy is a such neurological disease that has been a serious health problem affects life quality of human being. EEG based epilepsy detection in an effective and reliable way is a crucial issue for researchers. Effective feature extraction techniques to diminish input vector is a significant point in applications. In this study, auto-encoder based unsupervised feature extraction method was used and a deep learning approach was investigated to classify focal-non-focal EEG records. Proposed work has superiority in contrast to conventional methods because dataset was classified without using pre-processing and dimensionality-reduction methods. It has been thought that this work proposes an effective approach to diagnose epilepsy by using deep neural networks.Öğe Automatic Detection of Emotional State from EEG Signal by Gamma Coherence Approach(Ieee, 2018) Polat, Hasan; Ozerdem, Mehmet SiracElectroencephalogram coherence analysis is an important measure to help us to assess functional cortical connections and to learn about regional cortical synchronization. In this study, it was aimed to automatically detect emotions related to audio-visual stimuli by electroencephalogram coherence approach. First, the synchronizations of EEG recorded from different regions of the scalp have been analyzed with each other. Coherence analysis was performed for the gamma band of the electroencephalogram signals. Electrode pairs were identified in which the changing emotional state can be observed clearly. The coherence features extracted from the electrode pairs were given to input of the classifier algorithm. The average classification accuracy for the four different participants was obtained as 83.5%.Öğe Automatic Identification of Adenoid Hypertrophy via Ensemble Deep Learning Models Employing X-ray Adenoid Images(Springer, 2025) Orenc, Sedat; Acar, Emrullah; Ozerdem, Mehmet Sirac; Sahin, Sefer; Kaya, AbdullahAdenoid hypertrophy, characterized by the abnormal enlargement of adenoid tissue, is a condition that can cause significant breathing and sleep disturbances, particularly in children. Accurate diagnosis of adenoid hypertrophy is critical, yet traditional methods, such as imaging and manual interpretation, are prone to errors. This study uses an ensemble deep learning-based approach for adenoid classification. It utilizes a unique dataset sourced from Batman Training and Research Hospital. The dataset is composed of masked and non-masked X-ray images. It is used to train and compare the performance of multiple convolutional neural network (CNN) models. By comparing classification accuracy between masked and non-masked datasets, the study reveals the importance of image preprocessing. Six deep learning models-EfficientNet, MobileNet, ResNet50, ResNet152, VGG16, and Xception-are tested, with ResNet50 achieving the highest accuracy (100% on masked images), while Xception performs the worst (65% F1-score). The results indicate that masking significantly enhances the accuracy and reliability of adenoid classification. ResNet50 and EfficientNet show strong generalization capabilities. Conversely, the lower performance of models like Xception highlights the variability in model suitability for this task. This research provides valuable insights into optimizing deep learning models for medical image classification and it advances the field of AI-based adenoid detection.Öğe CLASSIFICATION OF AGRESSIVE ACTION EMG SIGNALS BY AR BASED K-NN METHOD(Ieee, 2014) Acar, Emrullah; Ozerdem, Mehmet SiracThe fields of EMG signal processing technology has been effective in the application of prosthetic control and clinical medicine or sport science. The main purpose of this study is to classify two aggressive action EMG signals which are taken from different people, according to their texture feature vectors. The physical action EMG set is derived from UCI database. The power spectral density (PSD) estimation of EMG signals is calculated by using AR Burg Method. The texture feature vectors of EMG signals are extracted by applying statistical methods to PSD maps of each signal PSD based feature vectors are given to different type of k-NN classifier as inputs and the performance results of each system are compared. Finally, the best average performance is observed as 97.92 % in k=7, 9 and 10 neighbors structure of k-NN classifier.Öğe CLASSIFICATION OF ECoG PATTERNS RELATED TO FINGER MOVEMENTS WITH WAVELET BASED SVM METHODS(Ieee, 2014) Karadag, Kerim; Ozerdem, Mehmet SiracClassification of finger movement related to (electrocorticography) ECoG records is the main purpose of this study. Data set IV presented in BCI Competition IV was used in this study. This data set contains brain signals from three epileptic subjects and the data records consist of both ECoG and electronic glove data. ECoG segments related finger movements were extracted by means of finger movement records generated by electronic glove. Features of segments with different lengths were extracted using wavelets and the channels having high performance were determined. The coefficients were classified with Support Vector Machine (SVM) classifier. The mean performances of three subjects were obtained as follows; classification rate 91.76% for two fingers, classification rate 76.16% for three fingers, classification rate 61.34% for four fingers and classification rate 48.51% for five fingers.Öğe Classification of EEG Records for the Cursor Movement with the Convolutional Neural Network(Ieee, 2018) Turk, Omer; Ozerdem, Mehmet SiracNowadays, very successful results are obtained with deep learning architectures which can be applied to many fields. Because of the high performances it provides in many areas, deep learning has come to a central position in machine learning and pattern recognition. In this study, electroencephalogram (EEG) signals related to up and down cursor movements were represented as image pattern by using obtained approximation coefficients after wavelet transform. The Obtained image patterns were classified by applying Convolutional Neural Network. In this study, EEG records related to cursor movements were classified and classification accuracy was obtained as 88.13%.Öğe Classification of EEG Signals Using Hilbert-Huang Transform-Based Deep Neural Networks(Ieee, 2019) Zan, Hasan; Yildiz, Abdulnasir; Ozerdem, Mehmet SiracEpilepsy is one of the most common neurologic disease. Electroencephalogram (EEG) contains physiologic and pathological information related human nervous system. EEG signals used in this study are obtained from Bonn University, Department of Epileptology EEG database. Original database has five subsets (A, B, C, D, E). Data have been reorganized into three groups which are healthy (AB), interictal (CD) and ictal EEG signals. Furthermore, in order to examine effect of signal length on classification performance, three different lengths are used. Hilbert-Huang transform is applied to the signals and they are represented as image files. Then, generated images are fed into deep neural networks with five different structures for classification. Accuracy is calculated for all cases to asses performance of proposed method. it is clear that successful results could be obtained using Hilbert-Huang transform along with deep learning networks.Öğe Classification of Epilepsy Types from Electroencephalogram Time Series Using Continuous Wavelet Transform Scalogram-Based Convolutional Neural Network(Amer Soc Testing Materials, 2021) Turk, Omer; Akpolat, Veysi; Varol, Sefer; Aluclu, Mehmet Ufuk; Ozerdem, Mehmet SiracDuring the supervisory activities of the brain, the electrical activities of nerve cell clusters produce oscillations. These complex biopotential oscillations are called electroencephalogram (EEG) signals. Certain diseases, such as epilepsy, can be detected by measuring these signals. Epilepsy is a disease that manifests itself as seizures. These seizures manifest themselves in different characteristics. These different characteristics divide epilepsy seizure types into two main groups: generalized and partial epilepsy. This study aimed to classify different types of epilepsy from EEG signals. For this purpose, a scalogram-based, deep learning approach has been developed. The utilized classification process had the following main steps: the scalogram images were obtained by using the continuous wavelet transform (CWT) method. So, a one-dimension EEG time series was converted to a two-dimensional time-frequency data set in order to extract more features. Then, the increased dimension data set (CWT scalogram images) was applied to the convolutional neural network (CNN) as input patterns for classifying the images. The EEG signals were taken from Dicle University, Neurology Clinic of Medical School. This data consisted of four classes: healthy brain waves, generalized preseizure, generalized seizure, and partial epilepsy brain waves. With the proposed method, the average accuracy performance of three of the EEG records' classes (healthy, generalized preseizure, and generalized seizure), and that of all four classes of EEG records were 90.16 % (+/- 0.20) and 84.66 % (+/- 0.48). According to these results, regarding the specific accuracy ratings of the recordings, the healthy EEG records scored 91.29 %, generalized epileptic seizure records were at 96.50 %, partial seizure EEG records scored 89.63 %, and the preseizure EEG records had a 90.44 % rating. The results of the proposed method were compared to the results of both similar studies and conventional methods. As a result, the performance of the proposed method was found to be acceptable.Öğe Classification of imaginary movements in ECoG with a hybrid approach based on multi-dimensional Hilbert-SVM solution(Elsevier, 2009) Demirer, R. Murat; Ozerdem, Mehmet Sirac; Bayrak, CoskunThe study presented in this paper shows that electrocorticographic (ECoG) signals can be classified for making use of a human brain-computer interface (BCI) field. The results show that certain invariant phase transition features can be reliably used to classify two types of imagined movements accurately. Those are the left small-finger and tongue movements. Our approach consists of two main parts: channel selection based on Tsallis entropy in Hilbert domain and the nonlinear classification of motor imagery with support vector machines (SVMs). The new approach, based on Hilbert and statistical/entropy measurements, were combined with SVMs based on admissible kernels for classification purposes. The classification accuracy rates were 95% (264/278) and 73% (73/100) for training and testing sets, respectively. The results support the use of classification methods for ECoG-based BCIs. Published by Elsevier B.V.Öğe Comparison of NDVI and RVI Vegetation Indices Using Satellite Images(Ieee, 2019) Gonenc, Abdurrahman; Ozerdem, Mehmet Sirac; Acar, EmrullahRemote Sensing is the acquisition of information about its physical properties without direct contact with an object. This information is obtained through sensors. These sensors do not come into contact with objects. There are two different systems for remote sensing. These are Active and Passive Sensor Systems. Passive Sensor Systems measure the energy of the rays reflected from the objects by the rays sent by the sun. On the other hand, Active Sensor Systems measure the energy reflected from the objects by transmitting their rays to the object. Passive Sensor Systems can be shown as an example of optical sensor systems. The Landsat-8 satellite works with an optical sensor system. Synthetic Aperture Radar (SAR) systems are examples of active sensor systems. SAR systems have a wide range of usage in all weather conditions and they are a radar system that displays the earth in high resolution. Radarsat-2 satellite has SAR sensor systems. The aim of this study is to compare each of the vegetation indices by using Landsat-8 and Radarsat-2 satellite images with two different types of sensors. In this study, Radar Vegetation Index (RVI) and Normalized Difference Vegetation Index (NDVI) were investigated. For the calculation of the RVI index, the back-scattering coefficient of the four different bands (HH, HV, VH, VV) of the multi-time full-polarimetric Radarsat-2 FQ satellite image dated 8 April 2015 was used. In the calculation of NDVI index, Band 5 (Near Infrared) and Band 4 (Red) of the Landsat-8 satellite image of May 25, 2015 were used. Dicle University agricultural areas were chosen as the study area. 100 different GPS points belonging to this agricultural area were determined and RVI and NDVI values of these points were calculated. A good correlation was observed between RVI and NDVI indices with the aid of statistically approach.Öğe The Comparison of Wavelet and Empirical Mode Decomposition Method in Prediction of Sleep Stages from EEG Signals(Ieee, 2017) Polat, Hasan; Akin, Mehmet; Ozerdem, Mehmet SiracThe aim of this study was to detect sleep stages of human by using EEG signals. In accordance with this purpose, discrete wavelet transforms (DWT) and empirical mode decomposition (EMD) were separately used for feature extraction. Subcomponents of EEG signals obtained by the two methods were assumed as feature vectors. Statistical parameters were used to reduce dimension of feature vectors. The same statistical parameters were used to compare performance of methods related to DWT and EMD. K nearest neighborhood (kNN) algorithm was used in classification final feature vectors that obtained EEG segments related to different sleep stages. The classification accuracies for feature vectors based on DWT and EMD were obtained as 100% and 88.13%, respectively.Öğe Determination Of Changes in Frequencies of EEG Signal in Eyes Open/Closed Duration(Ieee, 2015) Turk, Omer; Ozerdem, Mehmet SiracIn this study, the changes of the power spectral density (PSD) in the EEG data during eyes-closed and eyes-open states were analyzed. In the analysis, the interval of dominant frequencies was roughly determined with different approaches. The EEG signal is separated into sub bands with wavelet transform (WT). The Welch method which is the one of the classical methods was used for PSD prediction and the Burg and Yule-Walker parametric methods were used also for PSD prediction of the EEG signal. It was observed that the alpha rhythm is dominant band in the eyes closed state compared to eyes open state.Öğe Determination of ECoG information flow activity based on Granger causality and Hilbert transformation(Elsevier Ireland Ltd, 2013) Demirer, R. Murat; Ozerdem, Mehmet Sirac; Bayrak, Coskun; Mendi, EnginAnalysis of directional information flow patterns among different regions of the brain is important for investigating the relation between ECoG (electrocorticographic) and mental activity. The objective is to study and evaluate the information flow activity at different frequencies in the primary motor cortex. We employed Granger causality for capturing the future state of the propagation path and direction between recording electrode sites on the cerebral cortex. A grid covered the right motor cortex completely due to its size (approx. 8 cm x 8 cm) but grid area extends to the surrounding cortex areas. During the experiment, a subject was asked to imagine performing two activities: movement of the left small finger and/or movement of the tongue. The time series of the electrical brain activity was recorded during these trials using an 8 x 8 (0.016-300 Hz band with) ECoG platinum electrode grid, which was placed on the contralateral (right) motor cortex. For detection of information flow activity and communication frequencies among the electrodes, we have proposed a method based on following steps: (i) calculation of analytical time series such as amplitude and phase difference acquired from Hilbert transformation, (ii) selection of frequency having highest interdependence for the electrode pairs for the concerned time series over a sliding window in which we assumed time series were stationary, (iii) calculation of Granger causality values for each pair with selected frequency. The information flow (causal influence) activity and communication frequencies between the electrodes in grid were determined and shown successfully. It is supposed that information flow activity and communication frequencies between the electrodes in the grid are approximately the same for the same pattern. The successful employment of Granger causality and Hilbert transformation for the detection of the propagation path and direction of each component of ECoG among different subcortex areas were capable of determining the information flow (causal influence) activity and communication frequencies between the populations of neurons successfully. (C) 2013 Elsevier Ireland Ltd. All rights reserved.Öğe Displacement prediction of precast concrete under vibration using artificial neural networks(Techno-Press, 2020) Aktas, Gultekin; Ozerdem, Mehmet SiracThis paper intends to progress models to accurately estimate the behavior of fresh concrete under vibration using artificial neural networks (ANNs). To this end, behavior of a full scale precast concrete mold was investigated numerically. Experimental study was carried out under vibration with the use of a computer-based data acquisition system. In this study measurements were taken at three points using two vibrators. Transducers were used to measure time-dependent lateral displacements at these points on mold while both mold is empty and full of fresh concrete. Modeling of empty and full mold was made using ANNs. Benefiting ANNs used in this study for modeling fresh concrete, mold design can be performed. For the modeling of ANNs: Experimental data were divided randomly into two parts such as training set and testing set. Training set was used for ANN's learning stage. And the remaining part was used for testing the ANNs. Finally, ANN modeling was compared with measured data. The comparisons show that the experimental data and ANN results are compatible.
