DSpace Arşivi :: by Yazar "Akdemir, Bayram" değerine göre listeleniyor

Yazar "Akdemir, Bayram" seçeneğine göre listele

Listeleniyor 1 - 20 / 40

3 fazlı asenkron motorların MC3PHAC mikrodenetleyicisi kullanılarak hız kontrolü
(Selçuk Üniversitesi Fen Bilimleri Enstitüsü, 2003-06-24) Akdemir, Bayram; Ürkmez, Abdullah
Yapılan bu tez çalışmasında, 3 fazlı asenkron motorların hız kontrolü gerçekleştirilmiştir. Motorun hız kontrolünde V/f yöntemi uygulanmış ve motora belirlenen rampa boyunca yumuşak olarak yol verilmiştir. Mikrodenetleyici olarak asenkron motor kontrolüne yönelik olarak tasarlanmış DSP tabanlı bir mikrokontrolör olan MC3PHAC entegresi kullanılmıştır. MC3PHAC entegresinin üretmiş olduğu kontrol sinyalleri sinüs PWM şeklinde olup motor stator akımı, sinüs formundadır. DC giriş beslemesi olarak 3 fazlı yarım dalga doğrultulmuş şehir şebekesi kullanılmıştır. Motor sargıları üçgen bağlantılı olarak gerçekleştirilmiş ve deney sonuçlan bu bağlantı için elde edilmiştir. Sürücü katı 3 kollu 6 adet IGBT ile tasarlanmış ve sinüs örnekleme frekansı olarak 10.582 kHz seçilmiştir. Besleme girişine filtre ilave edilerek ortaya çıkan harmoniklerin şehir şebekesine geçmesi engellenmiştir. Tasarlanan devre, SIMCAD programı ile simüle edilmiş ve elde edilen değerler gerçek değerlerle karşılaştırılmıştır.
Adaptive neural fuzzy inference system (ANFIS) to determine possible relation between bricks porosity and weight losing
(2011) Başar, Mehmet Emin; Akdemir, Bayram
As in the architectural history, brick is still a common building material for building structures. This study offers a relation between the porosity of brick materials and their weight losing in acid treatment in order to evaluate brick features in restoration works sensitively. As for all matters, bricks get older with time by means of some causatives. Acids and acids derivatives may be the most important causatives for this phenomenon. In order to keep on the historical monuments in good conditions through the time, these monuments must be restored according to right architectural rules. With regard to the monuments built of brick material, there is no chance to take a piece from bricks to analyze in laboratory. For this study, the samples are obtained from some historical buildings of Anatolian heritage. These valuable data sets are evaluated by using adaptive neural fuzzy inference system to search a relation between the porosity of bricks and their weight losing in acidic conditions. In order to evaluate the results of tests, mean absolute error and leave one out cross validation methods were used to improve the reliability of the results. Leave one out cross validation result was obtained as 4.06 and mean absolute error 2,583. © 2011 Academic Journals.
Application of Feature Extraction and Classification Methods for Histopathological Image using GLCM, LBP, LBGLCM, GLRLM and SFTA
(Elsevier B.V., 2018) Öztürk, Şaban; Akdemir, Bayram
Classification of histopathologic images and identification of cancerous areas is quite challenging due to image background complexity and resolution. The difference between normal tissue and cancerous tissue is very small in some cases. So, the features of the tissue patches in the image have key importance for automatic classification. Using only one feature or using a few features leads to poor classification results because of the small difference between the textures. In this study, the classification results are compared using different feature extraction algorithms that can extract various features from histopathological image texture. For this study, GLCM, LBP, LBGLCM, GLRLM and SFTA algorithms which are successful feature extraction algorithms have been chosen. The features obtained from these methods are classified with SVM, KNN, LDA and Boosted Tree classifiers. The most successful feature extraction algorithm for histopathological images is determined and the most successful classification algorithm is determined. © 2018 The Authors. Published by Elsevier Ltd.
Artificial Frame Filling Using Adaptive Neural Fuzzy Inference System for Particle Image Velocimetry Dataset
(SPIE-INT SOC OPTICAL ENGINEERING, 2015) Akdemir, Bayram; Dogan, Sercan; Aksoy, Muharrem Hilmi; Canli, Eyup; Ozgoren, Muammer
Liquid behaviors are very important for many areas especially for Mechanical Engineering. Fast camera is a way to observe and search the liquid behaviors. Camera traces the dust or colored markers travelling in the liquid and takes many pictures in a second as possible as. Every image has large data structure due to resolution. For fast liquid velocity, there is not easy to evaluate or make a fluent frame after the taken images. Artificial intelligence has much popularity in science to solve the nonlinear problems. Adaptive neural fuzzy inference system is a common artificial intelligence in literature. Any particle velocity in a liquid has two dimension speed and its derivatives. Adaptive Neural Fuzzy Inference System has been used to create an artificial frame between previous and post frames as offline. Adaptive neural fuzzy inference system uses velocities and vorticities to create a crossing point vector between previous and post points. In this study, Adaptive Neural Fuzzy Inference System has been used to fill virtual frames among the real frames in order to improve image continuity. So this evaluation makes the images much understandable at chaotic or vorticity points. After executed adaptive neural fuzzy inference system, the image dataset increase two times and has a sequence as virtual and real, respectively. The obtained success is evaluated using R-2 testing and mean squared error. R2 testing has a statistical importance about similarity and 0.82, 0.81, 0.85 and 0.8 were obtained for velocities and derivatives, respectively.
Automatic leaf segmentation using grey Wolf optimizer based neural network
(Institute of Electrical and Electronics Engineers Inc., 2017) Öztürk, Şaban; Akdemir, Bayram
This study proposes a hybrid neural network model for the segmentation of leaf images with various illumination conditions. Segmentation of images with different illumination conditions is a quite challenging process. In particular, the shadows and dark regions in the image can be quite misleading for traditional segmentation algorithms. Using a single feature or reviewing them in a single colour space may work for some images, but this approach does not work on the entire dataset that have different colour. For this reason, automatic segmentation method is proposed in this study by using components from four different colour spaces. Firstly, the image is converted into RGB, HSV, XYZ and YIQ channels. Then, B, S, Z and I components are used to train hybrid neural network. Grey wolf optimizer is used for neural network optimization. The segmentation results of proposed method are compared with the well-known segmentation algorithms and are more successful. The results of proposed method are that sensitivity is 99.66 %, specificity is 98.42 % and accuracy is 99.31 %. © 2017 IEEE.
Combining Different Image Parts of Instruments with Image Mosaicing
(IEEE, 2017) Ozturk, Saban; Ozkaya, Umut; Akdemir, Bayram; Seyfi, Levent A.; Kulaksiz, Afsin
In this study, the different image parts belonging to one instrument are combined to obtain a high resolution and full-bodied image as a whole. Images of different sizes that represent different regions of the same instrument are properly combined to obtain a full representation of the instrument. First, properties of images are obtained by using the histogram of gradient (HOG) features. Then, features similar to those obtained features are searched in other images. For this, the obtained properties are applied using convolution in all image. The regions having the highest convolution value are selected as junction regions. There are two problems in the image combining process. These are: the dimensions of the instrument parts in the images may be different, and the location where these parts are located may not overlap. An approach for automatic image sliding and automatic scale synchronization has been proposed for these problems. Finally, the merged pixels of the resulting image are softened. So, the contrast differences between the images due to the light is minimized. The proposed method was tested using different instruments in experiments. Successful results have been achieved for all compelling test images.
Comparison of Edge Detection Algorithms for Texture Analysis on Glass Production
(ELSEVIER SCIENCE BV, 2015) Ozturk, Saban; Akdemir, Bayram
The use of technological innovations in production will increase the number of product and quality. With proposed method in this paper, it is aimed to improve the production process of glass which used almost every field. In this study, some of the popular edge detection algorithms (Roberts, Prewitt, Sobel, LoG and Canny) are used for the texture analysis process. It is aimed to determine glass surface defect with the applied of mentioned edge detection operators to same image. The results obtained from application are compared with the reference image and texture analysis performance of edge detection algorithms are evaluated. In this study the used material is glass and it is aimed to determine the glass surface defect such as scratch, crack and bubble with the use of edge detection operators. Glass is a difficult material to examine with cameras because glass has reflection and the transparency features. So, some improvement are applied in the image before edge detection algorithms are applied. Performed controlled experiments showed that LoG edge detection algorithm is better than other edge detection algorithms in determining texture analysis. (C) 2015 The Authors. Published by Elsevier Ltd.
Comparison of HOG, MSER, SIFT, FAST, LBP and CANNY features for cell detection in histopathological images
(BIOAXIS DNA RESEARCH CENTRE PRIVATE LIMITED, 2018) Ozturk, Saban; Akdemir, Bayram
Cell segmentation and counting has a very important role in diagnosing diseases and in the treatment process. But the complexity of the histopathological images and the differences in cell groups make this process very difficult, even for an expert. In order to facilitate this process, analysis of histopathological images is performed by using computer vision methods. This paper presents the use of different feature extraction methods for cell detection in histopathological images and the comparison of the results of these algorithms. For this reason, HOG, MSER, SIFT, FAST, LBP and CANNY feature extraction algorithms are used. The aim of the study is to determine cells using different feature extraction methods and to determine which of these feature extraction algorithms will be more successful. Firstly, image pre-processing has been applied to clear the noises in the histopathological images. Then, feature extraction algorithms are applied to image, respectively. Finally, the successes of different feature extraction algorithms have been compared.
Computer aided diagnosis of ECG data on the least square support vector machine
(ACADEMIC PRESS INC ELSEVIER SCIENCE, 2008) Polat, Kemal; Akdemir, Bayram; Gunes, Salih
In this paper we describe a technique that has successfully classified arrhythmia from an ECG dataset using a least square support vector machine (LSSVM). LSSVM was applied to the ECG dataset to distinguish between healthy persons and diseased persons (arrhythmia). The LSSVM classifier trained with four train-test parts including a training-to-test split of 50-50%, a training-to-test split of 70-30%, and a training-to-test split of 80-20%. We have used the classification accuracy, sensitivity and specificity analysis, and ROC curves to test the performance of LSSVM classifier on the detection of ECG arrhythmia. The classification accuracies obtained are 100% for all the training-to-test splits. These results show that the proposed method is more promising than previously reported classification techniques. The results suggest that the proposed method can be used to enhance the performance of a new intelligent assistance diagnosis system. (C) 2007 Elsevier Inc. All rights reserved.
CONTROL OF A REAL-TIME HUMANOID ROBOT BASED ON GESTURE DETECTION
(ST JOHN PATRICK PUBL, 2017) Ozturk, Saban; Akdemir, Bayram
[Abstract not Available]
Convolution Kernel Size Effect on Convolutional Neural Network in Histopathological Image Processing Applications
(IEEE, 2018) Ozturk, Saban; Ozkaya, Umut; Akdemir, Bayram; Seyfi, Levent
In this study, the change in the classification success of the convolutional neural network (CNN) is investigated when the dimensions of the convolution window are altered. For this purpose, four different linear convolution neural network architectures are constructed. The first architecture includes 4 convolution layers with 3x3 convolution window dimensions. The second architecture includes 4 convolution layers with 5x5 convolution window dimensions. The third architecture includes 4 convolution layers with 7x7 convolution window dimensions. The fourth architecture includes 4 convolution layers with 9x9 convolution window dimensions. A dataset consisting of histopathological image patches is used to test the CNN architects that are created. 2000 training images and 250 validation images on dataset are applied to all architectures with the same order, in order to fair assessment. In conclusion, the effect of convolution dimensions on classification of histopathological images by deep learning methods is determined. The test results of four different linear convolutional neural network architectures are evaluated using sensitivity, specificity and accuracy parameters.
A convolutional neural network model for semantic segmentation of mitotic events in microscopy images
(SPRINGER LONDON LTD, 2019) Orturk, Saban; Akdemir, Bayram
Mitosis, which has important effects such as healing and growing for human body, has attracted considerable attention in recent years. Especially, cell division characteristics contain useful information for regenerative medicine. However, the analysis of this complex structure is very challenging process for experts, because many cells are scattered at random times and at different speeds. Therefore, we propose an automatic mitosis event detection method using convolutional neural network (CNN). In the proposed method, semantic segmentation has been applied with the help of CNN in order to make the complex mitosis images more easily understandable. The CNN structure consists of four convolution layers, four pooling layers, one rectified linear unit layer and softmax layer. Generally, the aim of CNN structure is to reduce the image size, but in this study, the image size is preserved for the semantic segmentation which provides high-level information. For this, the size of the images at each layer output is calculated and updated with the appropriate padding parameters. Thus, real-size images presented at the network output can be easily understood. BAEC and C2C12 phase-contrast microscopy image sequences are used for experiments. The precision, recall and F-score parameters are used for evaluating the success of the proposed method and compared with the other methods using the same datasets.
Correlation- and covariance-supported normalization method for estimating orthodontic trainer treatment for clenching activity
(SAGE PUBLICATIONS LTD, 2009) Akdemir, Bayram; Ökkesim, Şükrü; Kara, Sadık; Güneş, Salih
In this study, electromyography signals sampled from children undergoing orthodontic treatment were used to estimate the effect of an orthodontic trainer on the anterior temporal muscle. A novel data normalization method, called the correlation- and covariance-supported normalization method (CCSNM), based on correlation and covariance between features in a data set, is proposed to provide predictive guidance to the orthodontic technique. The method was tested in two stages: first, data normalization using the CCSNM; second, prediction of normalized values of anterior temporal muscles using an artificial neural network (ANN) with a Levenberg-Marquardt learning algorithm. The data set consists of electromyography signals from right anterior temporal muscles, recorded from 20 children aged 8-13 years with class II malocclusion. The signals were recorded at the start and end of a 6-month treatment. In order to train and test the ANN, two-fold cross-validation was used. The CCSNM was compared with four normalization methods: minimum-maximum normalization, z score, decimal scaling, and line base normalization. In order to demonstrate the performance of the proposed method, prevalent p erformance-measuring methods, and the mean square error and mean absolute error as mathematical methods, the statistical relation factor R-2 and the average deviation have been examined. The results show that the CCSNM was the best normalization method among other normalization methods for estimating the effect of the trainer.
Detection of PCB Soldering Defects using Template Based Image Processing Method
(2017) Öztürk, Şaban; Akdemir, Bayram
In this study, a predefined template-based image processing system is proposed to automatically detect of PCB soldering defects that negatively affect circuit operation. The proposed system consists of a scaled inspection structure, a camera, an image processing algorithm merged with Fuzzy and template guided inspection process. The prototype is produced using a plastic material, depending on the focal length of the camera and the PCB size. Image processing step comprises two steps. Firstly, solder joints are determined and boxed using Fuzzy C-means clustering algorithm. Then, the center of each joint is determined. In the next step, a joint template is created that contains solder joints information. This joint template contains information about the joints that includes possible touching odds to other joints. Template accelerates the algorithm diverting to closest joint that may include defect. Finally, each joint is only inspected regarding template guide that based on neighbor joints. Proposed method includes a scaled inspection structure related to focal length of camera. During the every query, PCB must be located same coordinates via mechanical guiding on the structure to obtain same picture. Thus, taken picture could be same every trying. The proposed method is executed 85 times on same sample PCB in case of any fake output error. In order to obtain commercial success, mechanical structure was improved and for inspected PCB success was obtained 100%
Effective histopathological image area reduction method for real-time applications
(IS&T & SPIE, 2018) Ozturk, Saban; Akdemir, Bayram
Histopathologic images are time consuming for both specialist and machine learning methods with their complex structure and huge dimensions. In these cases, delays in the diagnosis of disease occur, as well as the treatment of fewer patients. When the histopathological images are examined at low resolution for shortening the examination time, it is almost impossible to identify the cancerous regions. When examining high-resolution images, it takes a long time to inspect because the image is divided into patches. Despite the fact that fairly fast machine learning methods are offered for the shortening of the analysis period, the number of patches to be examined has a negative effect on the decision time. For this reason, the area under examination needs to be reduced. For this, first of all, the destruction of cell-free areas and then the destruction of areas containing noncancerous cells must be eliminated. An effective and fast method of area reduction is presented for faster analysis and real-time use of histopathological images by machine learning algorithms. A two-step approach is used in the proposed method. In the first step, 3 x 3 texture properties of images are obtained and discrete wavelet transform is applied. Then, the image is cleaned with simple morphological processes. In the second step, the cleaned image is subjected to a discrete wavelet transform. Thus, the changes in cell-containing regions are captured, and regions that may be dangerous are identified. The proposed method reduced the areas to be examined by 98.5% to 99.5% with 95.33% accuracy. (C) 2018 SPIE and IS&T
Effects of Histopathological Image Pre-processing on Convolutional Neural Networks
(Elsevier B.V., 2018) Öztürk, Şaban; Akdemir, Bayram
In this study, classification performance of histopathological images which are processed by pre-processing algorithms using convolutional neural network structure is examined. The images are divided into four different pre-processing classes with their original state and processed with three different techniques. These classes are; original, normal pre-processing, other normal pre-processing and over pre-processing. Histopathological images of these four classes include cancerous and non-cancerous image patches. For these image classes, cancer patch classification is done using the same convolutional neural network structure. In this view, pre-processing effects on the classification success of the convolutional neural network is examined. For the normal pre-processing algorithm, background noise reduction and cell enhancement are applied. For over pre-processing, thresholding and morphological operations are applied in addition to normal preprocessing operations. At the end of the experiments, the most successful classification results are produced with the normal pre-processing algorithms. This is why the meaningful features of the image are left for the CNN structure that automatically learns the feature. The over pre-processing algorithm removes most of these important features from the image. © 2018 The Authors. Published by Elsevier Ltd.
Elliot Waves Predicting for Stock Marketing Using Euclidean Based Normalization Method Merged with Artificial Neural Network
(IEEE, 2009) Akdemir, Bayram; Yu, Lingwen
Financial Marketing is very common in the world to make money or to control the company strategy. Nearly all events trigger to each other and moreover countries. Some predicting methods, on guessing the marketing depends on natural behavior of the events. When, have a scrutinize to backwards, it can be evaluated that some upfront events occur periodically and trigger to each others and may lead to next known moving. Elliot is one of the famous estimating methods on stock marketing. Elliot waves let to time to think and analyze the next moving not in hurry. The proposed method consist of three stages (i) arranging the real time data (ii) data normalization according to Euclidean distance named Euclidean Based Normalization Method and (iii) performing artificial neural network to predict the next swing of the stock or financial marketing. The results compared raw data results and minimum maximum normalization methods to EBNM. Mean squared error and Average Deviation and R-2 statistical value were used as performance criteria. According to MSE, the obtained results were 0.000484, 0.205069 and 0.003178 minimum maximum normalization, raw data set and EBNM method respectively. The performance of the proposed method has more accurate than the other two methods.
Ensemble adaptive network-based fuzzy inference system with weighted arithmetical mean and application to diagnosis of optic nerve disease from visual-evoked potential signals
(ELSEVIER, 2008) Akdemir, Bayram; Kara, Sadik; Polat, Kemal; Guven, Ayegul; Gunes, Salih
Objective: This paper presents a new method based on combining principal component analysis (PCA) and adaptive network-based fuzzy inference system (ANFIS) to diagnose the optic nerve disease from visual-evoked potential (VEP) signals. The aim of this study is to improve the classification accuracy of ANFIS classifier on diagnosis of optic nerve disease from VEP signals. With this aim, a new classifier ensemble based on ANFIS and PCA is proposed. Methods and material: The VEP signals dataset include 61 healthy subjects and 68 patients suffered from optic nerve disease. First of all, the dimension of VEP signals dataset with 63 features has been reduced to 4 features using PCA. After applying PCA, ANFIS trained using three different training-testing datasets randomly with 50-50% training-testing partition. Results: The obtained classification results from ANFIS trained separately with three different training-testing datasets are 96.87%, 98.43%, and 98.43%, respectively. And then the results of ANFIS trained with three different training-testing datasets randomly with 50-50% training-testing partition have been combined with three different ways including weighted arithmetical mean that proposed firstly by us, arithmetical mean, and geometrical mean. The classification results of ANFIS combined with three different ways are 98.43%, 100%, and 100%, respectively. Also, ensemble ANFIS has been compared with ANN ensemble. ANN ensemble obtained 98.43%, 100%, and 100% prediction accuracy with three different ways including arithmetical mean, geometrical mean and weighted arithmetical mean. Conclusion: These results have shown that the proposed classifier ensemble approach based on ANFIS trained with different train-test datasets and PCA has produced very promising results in the diagnosis of optic nerve disease from VEP signals. (C) 2008 Published by Elsevier B.V.
Facial Expression Recognition with an Optimized Radial Basis Kernel
(IEEE, 2018) Ashir, Abubakar M.; Akdemir, Bayram
In this work, a new approach for facial expression recognition has been proposed. The approach has imbedded in it both new feature extraction technique and classification techniques using automatic auto-tuning of kernel parameter optimization in support vector machines. It generally begins with feature extraction from the input vectors using a combination of arithmetic means difference and rotation invariant Local Binary Pattern. The extracted features are projected into a Gaussian space to match it with the radial basis function kernel used in support vector machines for classification. Prior to classification, an optimized parameter for support vector machines training are automatically determined based on an approach proposed which relies on the receiver operating characteristics of the support vector machine classifier. The results obtained from the experiments were impressive and promising. From the experiments conducted on the two facial expression databases with different cross-validation techniques, the proposed approach outperforms its counterparts under the same database and settings.
Facial expression recognition with dynamic cascaded classifier
(Springer London, 2019) Ashir, Abubakar Muhammad; Eleyan, Alaa; Akdemir, Bayram
In this paper, a new approach for facial expression recognition has been proposed. The approach has imbedded a new feature extraction technique, new multiclass classification approach and a new kernel parameter optimization for support vector machines. The scheme of the approach begins with feature extraction from the input vectors, and the extracted features are transformed into a Gaussian space using compressive sensing techniques. This process ensures feature vector dimensionality reduction and matches the features vectors with radial basis function kernel used in support vector machines for classification. Prior to classification, an optimized parameter for support vector machines training is automatically determined based on an approach proposed which relies on the receiver operating characteristics of the support vector machine classifier. With the optimized kernel parameter, new proposed multiclass classification approach is used to finally classify any vector. In all the experiments conducted on the two facial expression databases with different cross-validation techniques, the proposed approach outperforms its counterparts under the same database and settings. The results further confirmed the validity and advantages of the proposed approach over other approaches currently used in the literature. © 2019, Springer-Verlag London Ltd., part of Springer Nature.

Yazar "Akdemir, Bayram" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri