Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/202

Browse

Now showing 1 - 20 of 350

Planar MEMS Variable Optical Attenuators (VOAs) with Linear Attenuation-Voltage Characteristics
(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 01.01.2019) Hah, Dooyoung; 0000-0002-1290-0597; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
Variable optical attenuators (VOAs) are essential components in wavelength division multiplexing (WDM) networks, light waveform generators, and optical fiber test equipment. Among various types of planar MEMS VOAs, a shutter type and a reflective type have been most frequently studied so far. In a shutter type, a knife-edge-like beam blocker is inserted in between the butt-coupled input and output fibers, partially obstructing the coupling between the fibers. In a reflective type, a mirror that is placed in the optical path controls the alignment between the fibers to result in attenuation. The movement of the shutter or the mirror is controlled by MEMS actuators, such as comb-drive actuators. In most of the planar MEMS VOAs reported, the relationship between the attenuation and the control voltage has been highly nonlinear. This nonlinearity results in uneven resolution throughout the attenuation range. Although this nonlinearity can be addressed by employing a control system, a structure-based solution is preferred, which can curtail the requirement of power consumption, and prevent control instability issues. In this study, shaped-finger comb-drive actuators are used to obtain a linear relationship between the control voltage and the attenuation in planar MEMS VOAs. Two types, i.e. shutter-type and reflective-type, of VOAs are examined. First, the objective differential equation is established based on attenuation-displacement relationships, electrostatic/mechanical force balance equation, and the design objective (linearity) equation. Then, the differential equation (in terms of 2-D comb capacitance) is solved by using the Euler's method, and the finger gaps are calculated by using a conformal mapping method. When a single comb-drive actuator is used, an excluded zone needs to be introduced around the region of small displacement. Effects of the width of the excluded zone to the device characteristics are studied. The issue of zone exclusion can be addressed by adopting dual (control and bias) combs. The effects of design parameters to the VOA performances are studied. It is shown that the planar MEMS VOAs with linear attenuation-voltage relationships can be designed successfully by using the proposed method.
Performance Evaluations of Active Subnetwork Search Methods in Protein-Protein Interaction Networks
(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 01.01.2019) Gunter, Pinar; Bakir-Gungor, Burcu; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
Protein-protein interaction networks are mathematical representations of the physical contacts between proteins in the cell. A group of interconnected proteins in a protein-protein interaction network that contains most of the disease associated proteins and some interacting other proteins is called an active subnetwork. Active subnetwork search is important to understand mechanisms underlying diseases. Active subnetworks are used to discover disease related regulatory pathways, functional modules and to classify diseases. In the literature there arc many methods to search for active subnetworks. The purpose of this study is to compare the performance of different subnetwork identification methods. By using the Rheumatoid Arthritis dataset, the performances of greedy approach, genetic algorithm, simulated annealing algorithm, prize collecting steiner forest and game theory based subnetwork search methods are compared.
94.8 Km-Range Direct Detection Fiber Optic Distributed Acoustic Sensor
(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 01.01.2019) Uyar, Faruk; Onat, Talha; Unal, Canberk; Unal, Canberk; Ozdur, Ibrahim; Ozbay, Ekmel; 0000-0003-3552-1650; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
This work demonstrates an ultra-long range direct detection fiber optic distributed acoustic sensor which can detect vibrations at a distance of 94.8 km with 10 m resolution along the sensing fiber. (C) 2019 The Author(s)
Use of laser-induced bubbles in intraocular pressure measurement: a preliminary study
(IOP PUBLISHING LTD, TEMPLE CIRCUS, TEMPLE WAY, BRISTOL BS1 6BE, ENGLAND, 01.01.2019) Altindis, Fatih; Ozdur, Ibrahim T.; Mutlu, Sait N.; Yilmaz, Bulent; 0000-0002-3891-935X; 0000-0001-6452-0804; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
This work investigates the feasibility of a novel approach for measuring intraocular pressure (IOP) by analyzing micron-level laser-induced bubble characteristics in the intraocular fluid. We believe that this concept may be used as a non-invasive alternative for measuring a patient's IOP by analyzing the laser-induced bubble volume in the intraocular fluid in the anterior chamber of the eye. The behavior of laser-induced bubbles was examined under differing fluid pressure levels and at differing laser pulse energy levels. An intraocular medium-like environment was imitated and an imaging system was designed in order to capture laser-induced bubbles with their movements. The video recordings of the bubbles were processed using custom software, and the volume of the bubbles was estimated using three different approaches. The bubble volumes were estimated more accurately by using the rising velocity of the bubble rather than its direct radii appearances on the images. An inversely proportional relationship was observed between the laser-induced bubble volume and the fluid pressure. IOP can be measured with a non-invasive technique using laser-induced bubble volume. Deeper and detailed studies, including clinical studies, may lead to the use of lasers for measuring IOP.
Parameter investigation of topological data analysis for EEG signals
(ELSEVIER SCI LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, OXON, ENGLAND, 01.01.2021) Altindis, Fatih; Yilmaz, Bulent; Borisenok, Sergey; Icoz, Kutay; 0000-0002-0947-6166; 0000-0002-3891-935X; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
Topological data analysis (TDA) methods have become appealing in EEG signal processing, because they may help the scientists explore new features of complex and large amount of data by simplifying the process from a geometrical perspective. Time delay embedding is a common approach to embed EEG signals into the state space. Parameters of this embedding method are variable and the structure of the state space can be entirely different depending on their selection. Additionally, extracted persistent homologies of the state spaces depend on filtration level and the number of points used. In this study, we showed how to adapt false nearest neighbor (FNN) test to find out the suitable/optimal time embedding parameters (i.e., time delay and embedding dimension) for EEG signals, and compared their effects on different types of artefacts and motor intention waves that are commonly used in brain-computer interfaces. We extracted and compared persistent homologies of state spaces that were reconstructed with four different sets of parameters. Later, the effect of filtration level on extracted persistent homologies was compared, and statistical significance levels were computed between leftand right-hand movement imaginations. Finally, computational cost of the discussed methods was found, and the adaptability of this method to a real-time application was evaluated. We demonstrated that the discussed parameters of the TDA approach were highly crucial to extract true topological features of the EEG signals, and the adapted testing approaches depicted the applicability of this approach on real-time analysis of EEG signals.
Magnetic micro/nanoparticle flocculation-based signal amplification for biosensing
(DOVE MEDICAL PRESS LTD, PO BOX 300-008, ALBANY, AUCKLAND 0752, NEW ZEALAND, 07.07.2016) Mzava, Omary; Tas, Zehra; Icoz, Kutay; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
We report a time and cost efficient signal amplification method for biosensors employing magnetic particles. In this method, magnetic particles in an applied external magnetic field form magnetic dipoles, interact with each other, and accumulate along the magnetic field lines. This magnetic interaction does not need any biomolecular coating for binding and can be controlled with the strength of the applied magnetic field. The accumulation can be used to amplify the corresponding pixel area that is obtained from an image of a single magnetic particle. An application of the method to the Escherichia coli 0157: H7 bacteria samples is demonstrated in order to show the potential of the approach. A minimum of threefold to a maximum of 60-fold amplification is reached from a single bacteria cell under a magnetic field of 20 mT.
An FDTD-based computer simulation platform for shock wave propagation in electrohydraulic lithotripsy
(ELSEVIER, 2013) Yilmaz, Bulent; Ciftci, Emre; 0000-0003-2954-1217; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Yilmaz, Bulent
Extracorporeal Shock Wave Lithotripsy (ESWL) is based on disintegration of the kidney stone by delivering high-energy shock waves that are created outside the body and transmitted through the skin and body tissues. Nowadays high-energy shock waves are also used in orthopedic operations and investigated to be used in the treatment of myocardial infarction and cancer. Because of these new application areas novel lithotriptor designs are needed for different kinds of treatment strategies. In this study our aim was to develop a versatile computer simulation environment which would give the device designers working on various medical applications that use shock wave principle a substantial amount of flexibility while testing the effects of new parameters such as reflector size, material properties of the medium, water temperature, and different clinical scenarios. For this purpose, we created a finite-difference time-domain (FDTD)-based computational model in which most of the physical system parameters were defined as an input and/or as a variable in the simulations. We constructed a realistic computational model of a commercial electrohydraulic lithotriptor and optimized our simulation program using the results that were obtained by the manufacturer in an experimental setup. We, then, compared the simulation results with the results from an experimental setup in which oxygen level in water was varied. Finally, we studied the effects of changing the input parameters like ellipsoid size and material,temperature change in the wave propagation media, and shock wave source point misalignment. The simulation results were consistent with the experimental results and expected effects of variation in physical parameters of the system. The results of this study encourage further investigation and provide adequate evidence that the numerical modeling of a shock wave therapy system is feasible and can provide a practical means to test novel ideas in new device design procedures.
MODELING AND SIMULATION OF ADVANCED NUCLEAR REACTORS
(IEEE, 2013) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Advanced nuclear power plants are currently being proposed with a number of various designs. However, there is a lack of modeling and control strategies to deal with load following operations. This research investigates a possible modeling approach for advanced nuclear power plants in order to provide an assessment way to the concept designs. The modeling approach covers linear and nonlinear reactor modeling and linear modeling of heat exchanger-turbinegenerator systems. Numerical results are presented on an example molten-salt type nuclear power plant system to demonstrate the validity and effectiveness of the modeling approach.
A modeling and control approach to advanced nuclear power plants with gas turbines
(PERGAMON-ELSEVIER, 2013) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Advanced nuclear power plants are currently being proposed with a number of various designs. However, there is a lack of modeling and control strategies to deal with load following operations. This research investigates a possible modeling approach and load following control strategy for gas turbine nuclear power plants in order to provide an assessment way to the concept designs. A load frequency control strategy and average temperature control mechanism are studied to get load following nuclear power plants. The suitability of the control strategies and concept designs are assessed through linear stability analysis methods. Numerical results are presented on an advanced molten salt reactor concept as an example nuclear power plant system to demonstrate the validity and effectiveness of the proposed mod- eling and load following control strategies.
A new tool for QT interval analysis during sleep in healthy and obstructive sleep apnea subjects: a study on women
(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEY, ATATURK BULVARI NO 221, KAVAKLIDERE, ANKARA, 00000, TURKEY, 2013) Kaya, Kemal Alican; Yilmaz, Bulent; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
By monitoring the Q wave/T wave (QT) interval computed from electrocardiography (ECG) signals during sleep, it is possible to create a link between the ventricular repolarization and sleep stages. In this study, we aimed to find a robust and simple approach to automatically determine the fiducials on each 30-s sleep epoch, such as the Q, R, and T-end points, on long sleep ECG recordings in order to statistically analyze the effect of obstructive sleep apnea (OSA) and sleep stages on QT intervals. This is a retrospective study in which the ECG data extracted from the polysomnography recordings of 7 healthy women and 5 women with OSA, acquired in a sleep laboratory, were used. Experts annotated the sleep stage and OSA presence information for each 30-s epoch. Later, we visually selected epochs with clean signals from a total of 8324 epochs. On the selected epochs, we determined R peaks on each heartbeat, and by aligning each ECG portion corresponding to a heartbeat using those R points, we computed an average ECG signal for each epoch. On the average ECG signals, we developed a novel approach to find the Q and T-end points. With the help of Bazzet's formula, we computed the corrected QT interval (QTc) values for each epoch using the QT and the median RR interval. Finally, we analyzed the QTc values for the different sleep stages and healthy or OSA groups. We employed statistical approaches such as the Mann-Whitney U test, Freidman's test, and the Wilcoxon signed-rank test. As a result of this study, we found that OSA has a prolongation effect on the total duration of the ventricular depolarization and repolarization. We also observed that the QTc values computed in each sleep stage were significantly different between the healthy and OSA groups. Additionally, we discovered that within the healthy group, the QTc values were distinctive in the different sleep stages.
Robust reactivity, neutron source, and precursor estimators for nuclear reactors
(ELSEVIER SCIENCE SAPO BOX 564, 1001 LAUSANNE, SWITZERLAND, 2013) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Reactivity, precursor concentration, and external neutron source strength determine control, operation and performance of nuclear reactors. These main reactor quantities are not directly measurable and must be calculated or estimated using reactor kinetics. This study presents efficient and robust nonlinear estimation algorithms for predicting these fundamental reactor quantities. The effectiveness of the proposed estimators is assessed through chirp and step test signals in the presence of parameter uncertainties and measurement noise. (C) 2013 Elsevier B.V. All rights reserved.
Steam Generator Level Control with an Observer-Based Algebraic Approach
(IEEE, 2013) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Level control systems are used to maintain the steam generator water level within the prescribed narrow limits and to provide constant supply of steam during power demand variations. The current level control systems are often found to be unsatisfactory during startup and low power operations. A robust nonlinear estimator-based optimal algebraic control is developed for level control systems to solve the water level tracking problem during power demand variations. It is shown that the proposed control provides an optimal and robust water level tracking with a single automatic controller over the complete range of power operation in the presence of plant uncertainties and noisy measurements.
An Observer-Based Fault Diagnosis in Battery Systems of Hybrid Vehicles
(IEEE, 2013) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Hybrid electric vehicles (HEVs) currently use Nickel-Metal Hydride (Ni-MH) batteries which have advantages of design flexibility, superior power, environmental acceptability and recyclability, long life, wide-range operating temperature and low cost. No matter how good a battery is, a failure can always occur in a battery leading to serious inconvenience, performance deterioration and costly replacement. Thus, it is desirable to be able to detect the underlying degradation and to predict level of unsatisfactory performance. By using current, voltage and temperature measurements of Ni-MH batteries, they can be modeled so that the internal dynamics of the batteries can be estimated and state of health of the batteries can be predicted for secure and long-life operations. An observer-based fault diagnosis approach is designed to analyze the state of health of the Ni-MH battery system of HEVs in this study. Real-world input data is used to assess the efficiency of the approach in the existence of uncertainties. The possible sensor faults and unexpected parameter deviations are diagnosed efficiently with statistical evaluation of the generated residuals.
Multiple slow waves and relevant transverse transmission and confinement in chirped photonic crystals
(OPTICAL SOC AMER, 2010 MASSACHUSETTS AVE NW, WASHINGTON, DC 20036 USA, 2014) Serebryannikov, Andriy E.; Cakmak, A. Ozgur; Colak, Evrim; Caglayan, Humeyra; Kurt, Hamza; Ozbay, Ekmel; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
The dispersion properties of rod-type chirped photonic crystals (PhCs) and non-channeled transmission in the direction of the variation of structural parameters from one cell of such a PhC to another are studied. Two types of configurations that enable multiple slow waves but differ in the utilized chirping scheme are compared. It is demonstrated that the multiple, nearly flat bands with a group index of refraction exceeding 180 can be obtained. For these bands, transmission is characterized by multiple narrow peaks of perfect transmission, strong field enhancement inside the slab, and large values of the Q-factor. Among the bands, there are some that show negative phase velocity. Symmetry with respect to the slab mid-plane must be kept in order to obtain constructive interferences that are necessary for reflection-free transmission. It is shown that 15 and more slow wave bands can be obtained in one configuration. The corresponding transmission peaks are well separated from each other, being the only significant feature of the transmission spectrum, while the Q-factor can exceed 10(5). The observed features are preserved in a wide range of the incidence angle variation. They can be used for tuning the locations and spectral widths of the transmission peaks. Some comparisons with the chirped multilayer structures have been carried out. (C) 2014 Optical Society of America
Online Condition Monitoring of Battery Systems With a Nonlinear Estimator
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2014) Ablay, Gunyaz; 0000-0003-2862-6761; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
The performance of batteries as uninterruptable power sources in any industry cannot be taken for granted. The failures in battery systems of safety-related electric systems can lead to performance deterioration, costly replacement, and, more importantly, serious hazards. The possible failures in battery systems are currently determined through periodic maintenance activities. However, it is desirable to be able to detect the underlying degradation and to predict the level of unsatisfactory performance by an online real-time monitoring system to prevent unexpected failures through early fault diagnosis. Such an online fault diagnosis method can also contribute to better maintenance and optimal battery replacement programs. A robust nonlinear estimator-based online condition monitoring method is proposed to determine the state of health of the battery systems online in industry. Real-world experimental data of a modern battery system are used to assess the efficiency of the proposed approach in the existence of parameter uncertainties.
Validation of Higher-Order Approximations and Boundary Conditions for Lossy Conducting Bodies
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA, 2014) Sukharevsky, Ilya O.; Altintas, Ayhan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
The problem of high-frequency diffraction by a smooth lossy body with high conductivity is considered. In addition to the geometrical optics approximation, additional asymptotic terms are derived to take into account the curvature of the boundary and material properties. Since these higher-order terms are derived by taking into account exact boundary conditions, it is easy to learn about the limitations of impedance conditions and to determine more accurate approximate conditions. The obtained higher-order boundary conditions and their limitations are numerically validated by solving Muller's second-kind integral equations.
Wireless Measurement of Elastic and Plastic Deformation by a Metamaterial-Based Sensor
(MDPI, ST ALBAN-ANLAGE 66, CH-4052 BASEL, SWITZERLAND, 2014) Ozbey, Burak; Demir, Hilmi Volkan; Kurc, Ozgur; Erturk, Vakur B.; Altintas, Ayhan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
We report remote strain and displacement measurement during elastic and plastic deformation using a metamaterial-based wireless and passive sensor. The sensor is made of a comb-like nested split ring resonator (NSRR) probe operating in the near-field of an antenna, which functions as both the transmitter and the receiver. The NSRR probe is fixed on a standard steel reinforcing bar (rebar), and its frequency response is monitored telemetrically by a network analyzer connected to the antenna across the whole stress-strain curve. This wireless measurement includes both the elastic and plastic region deformation together for the first time, where wired technologies, like strain gauges, typically fail to capture. The experiments are further repeated in the presence of a concrete block between the antenna and the probe, and it is shown that the sensing system is capable of functioning through the concrete. The comparison of the wireless sensor measurement with those undertaken using strain gauges and extensometers reveals that the sensor is able to measure both the average strain and the relative displacement on the rebar as a result of the applied force in a considerably accurate way. The performance of the sensor is tested for different types of misalignments that can possibly occur due to the acting force. These results indicate that the metamaterial-based sensor holds great promise for its accurate, robust and wireless measurement of the elastic and plastic deformation of a rebar, providing beneficial information for remote structural health monitoring and post-earthquake damage assessment.
Colloidal Quantum Dot Light-Emitting Diodes Employing Phosphorescent Small Organic Molecules as Efficient Exciton Harvesters
(AMER CHEMICAL SOC, 1155 16TH ST, NW, WASHINGTON, DC 20036 USA, 2014) Mutlugun, Evren; Guzelturk, Burak; Abiyasa, Agus Putu; Gao, Yuan; Sun, Xiao Wei; Demir, Hilmi Volkan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
Nonradiative energy transfer (NRET) is an alternative excitation mechanism in colloidal quantum dot (QD) based electroluminescent devices (QLEDs). Here, we develop hybrid highly spectrally pure QLEDs that facilitate energy transfer pumping via NRET from a phosphorescent small organic molecule-codoped charge transport layer to the adjacent QDs. A partially codoped exciton funnelling electron transport layer is proposed and optimized for enhanced QLED performance while exhibiting very high color purity of 99%. These energy transfer pumped hybrid QLEDs demonstrate a 6-fold enhancement factor in the external quantum efficiency over the conventional QLED structure, in which energy transfer pumping is intrinsically weak.
A joint production and transportation planning problem with heterogeneous vehicles
(PALGRAVE MACMILLAN LTD, BRUNEL RD BLDG, HOUNDMILLS, BASINGSTOKE RG21 6XS, HANTS, ENGLAND, 2014) Toptal, Aysegul; Koc, Utku; Sabuncuoglu, Ihsan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
We consider a manufacturer's planning problem to schedule order production and transportation to respective destinations. The manufacturer in this setting can use two vehicle types for outbound shipments. The first type is available in unlimited numbers. The availability of the second type, which is less expensive, changes over time. Motivated by some industry practices, we present formulations for three different solution approaches: the myopic solution, the hierarchical solution and the coordinated solution. These approaches vary in how the underlying production and transportation subproblems are solved, that is, sequentially versus jointly or heuristically versus optimally. We provide intractability proofs or polynomial-time exact solution procedures for the sub-problems and their special cases. We also compare the three solution approaches over a numerical study to quantify the savings from integration and explicit consideration of transportation availabilities. Our analytical and numerical results set a foundation and a need for a heuristic to solve the integrated problem. We thus propose a tabu search heuristic, which quickly generates near-optimal solutions.
Liver fibrosis staging using CT image texture analysis and soft computing
(ELSEVIER, 2014) Kayaalti, Omer; Aksebzeci, Bekir Hakan; Karahan, Ibrahim Okkes; Deniz, Kemal; Ozturk, Mehmet; Yilmaz, Bulent; Kara, Sadik; Asyali, Musa Hakan; 0000-0001-7476-8141; 0000-0003-2954-1217; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Yilmaz, Bulent; Aksebzeci, Bekir Hakan
Liver biopsy is considered to be the gold standard for analyzing chronic hepatitis and fibrosis; however, it is an invasive and expensive approach, which is also difficult to standardize. Medical imaging techniques such as ultrasonography, computed tomography (CT), and magnetic resonance imaging are non-invasive and helpful methods to interpret liver texture, and may be good alternatives to needle biopsy. Recently, instead of visual inspection of these images, computer-aided image analysis based approaches have become more popular. In this study, a non-invasive, low-cost and relatively accurate method was developed to determine liver fibrosis stage by analyzing some texture features of liver CT images. In this approach, some suitable regions of interests were selected on CT images and a comprehensive set of texture features were obtained from these regions using different methods, such as Gray Level Co-occurrence matrix (GLCM), Laws’ method, Discrete Wavelet Transform (DWT), and Gabor filters. Afterwards, sequential floating forward selection and exhaustive search methods were used in various combinations for the selection of most discriminating features. Finally, those selected texture features were classified using two methods, namely, Support Vector Machines (SVM) and k-nearest neighbors (k-NN). The mean classification accuracy in pairwise group comparisons was approximately 95% for both classification methods using only 5 features. Also, performance of our approach in classifying liver fibrosis stage of subjects in the test set into 7 possible stages was investigated. In this case, both SVM and k-NN methods have returned relatively low classification accuracies. Our pairwise group classification results showed that DWT, Gabor, GLCM, and Laws’ texture features were more successful than the others; as such features extracted from these methods were used in the feature fusion process. Fusing features from these better performing families further improved the classification performance. The results show that our approach can be used as a decision support system in especially pairwise fibrosis stage comparisons.

Browse

Browsing Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu by Issue Date