Yazar "Talu, M. Fatih" seçeneğine göre listele
Listeleniyor 1 - 14 / 14
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Calculation of melatonin and resveratrol effects on steatosis hepatis using soft computing methods(Elsevier Ireland Ltd, 2013) Talu, M. Fatih; Gul, Mehmet; Alpaslan, Nuh; Yigitcan, BirgulIn this work, beneficial effects of melatonin and resveratrol drugs on liver damage in rats, induced by application of acute and chronic carbon tetrachloride (CCl4) have been examined. The study consists of three main stages: (1) Data acquisition: light microscopic images were obtained from 60 rats separated into 10 groups after the preparation of liver tissue samples for histological examination. Rats in first five experimental groups for the four-day and the other five groups for twenty-day were examined. (2) Data processing: by the help of histograms of oriented gradient (HOG) method, obtaining low-dimensional image features (color, shape and texture) and classifying five different group characteristics by using these features with artificial neural networks (ANNs), and support vector machines (SVMs) have been provided. (3) Calculation of drug effectiveness: firstly to determine the differences between group characteristics of rats, a pilot group has been selected (diseased group-CCl4), and the responses of ANN and SVM trained by HOG features have been calculated. As a result of ANN, it has been seen that melatonin and resveratrol drugs have %65.62-% 75.12 positive effects at the end of the fourth day, %84.12-%98.89 positive effects on healing steatosis hepatis at the end of the twentieth day respectively and as a result of SVM, it has been seen that melatonin and resveratrol drugs have %62.5-%68.75 positive effects at the end of the fourth day, %45.12-%60.89 positive effects on healing steatosis hepatis at the end of the twentieth day respectively. (C) 2013 Elsevier Ireland Ltd. All rights reserved.Öğe Comparison of the Stochastic Gradient Descent Based Optimization Techniques(Ieee, 2017) Yazan, Ersan; Talu, M. FatihThe stochastic gradual descent method (SGD) is a popular optimization technique based on updating each theta(k) parameter in the partial derivative J(theta)/partial derivative theta(k) direction to minimize / maximize the (J theta) cost function. This technique is frequently used in current artificial learning methods such as convolutional learning and automatic encoders. In this study, five different approaches (Momentum, Adagrad, Adadelta, Rmsprop ve Adam) based on SDA used in updating the theta parameters were investigated. By selecting specific test functions, the advantages and disadvantages of each approach are compared with each other in terms of the number of oscillations, the parameter update rate and the minimum cost reached. The comparison results are shown graphically.Öğe Examination of the effect of the basic parameters of the auto-encoder on coding performance(Ieee, 2017) Calisan, Mucahit; Talu, M. FatihIn this study, artificial learning approach which can express high dimensional data in a lower space (autocoding) and known as autoencoder in the literature has been investigated in detail without using a predefined ready mathematical model. The most important feature of this method, which can be used in place of traditional feature extraction methods (HOG, SHIFT, SURF, Wavelet, etc.), is the ability to extract data-specific features. By applying the real (MNIST) and synthetic data, the effects on the success of the parameters of the method are measured and the results are presented in a tabular form.Öğe Flame stability measurement through image moments and texture analysis(Iop Publishing Ltd, 2023) Golgiyaz, Sedat; Cellek, M. Salih; Daskin, Mahmut; Talu, M. Fatih; Onat, CemIn this article, the first two moments of the image, mean and standard deviation, uniform local binary pattern (LBP) texture analysis methods were experimentally investigated in coal-fired boilers to measure flame stability. The first two moments of the flame image were used to evaluate the flame stability in terms of color and brightness (average gray value). Although the radiation signal of the flame is widely obtained by the spectral analysis method, the radiation signal of the flame was obtained by the LBP texture analysis method in this study. The flame stability measurement technique proposed in this study does not require prior knowledge about charged coupling devices camera features. Therefore, it can be easily applied to measure flame stability without expensive and complicated adaptation processes. Flame stability was measured with R = 0.9868 accuracy with the proposed method. The experimental results show that the proposed texture analysis method is more effective than current spectral analysis methods. The results obtained within the scope of this study also show that it can be easily applied to existing closed-loop control systems to monitor flame stability.Öğe Hybrid Lossless Compression Method For Binary Images(Istanbul Univ, Fac Engineering, 2011) Talu, M. Fatih; Tusrkoglu, IbrahimIn this paper, we propose a lossless compression scheme for binary images which consists of a novel encoding algorithm which uses a new edge tracking algorithm. The proposed compression scheme has two substages: (i) encoding binary image data using the proposed encoding method (ii) compression the encoded image data using any well-known image compression method such as Huffman, Run-Length or Lempel-Ziv-Welch (LZW). The proposed encoding method contains two subsequent processes: (i) determining the starting points of independent objects (ii) obtaining their edge points and geometrical shapes information. Experimental results show that using the proposed encoding method together with any traditional compressing method improves the compression performance. Computed mathematical results related to compression performance are represented comparatively in a tabular format.Öğe A hybrid tracking method for scaled and oriented objects in crowded scenes(Pergamon-Elsevier Science Ltd, 2011) Talu, M. Fatih; Turkoglu, Ibrahim; Cebeci, MehmetTraditional kernel based means shift assumes constancy of the object scale and orientation during the course of tracking and uses a symmetric/asymmetric kernel, such as a circle or an ellipse for target representation. In a tracking scenario, it is not uncommon to observe objects with complex shapes whose scale and orientation constantly change due to the camera and object motions. In this paper, we propose a multi object tracking method which tracks the complete object regions, adapts to changing scale and orientation, and assigns consistent labels to each object throughout real world video sequences. Our approach has five major components: (1) dynamic background subtraction, (2) level sets, (3) mean shift convergence, (4) object identification, and (5) occlusion handling. The experimental results show that the proposed method is superior to the traditional mean shift tracking in the following aspects: (1) it provides consistent multi objects tracking instead of single object throughout the video, (2) it is not affected by the scale and orientation changes of the tracked objects, (3) its computational complexity is much less than traditional mean shift due to using level set method instead of probability density. (C) 2011 Elsevier Ltd. All rights reserved.Öğe A NOVEL TEXTURE CLASSIFICATION METHOD BASED ON HESSIAN MATRIX AND PRINCIPAL CURVATURES(Ieee, 2014) Alpaslan, Nuh; Hanbay, Kazim; Hanbay, Davut; Talu, M. FatihIn this study, in order to obtain similar effect with conventional gradient operation and extract more robust feature for texture, we use the principal curvature informations instead of the gradient calculation. Through this methods, sharp and important informations about the texture images were obtained by analyzing images of the second order. Considering the classification results obtained, it is shown that the proposed method improve the performance of original CoHOG and HOG feature extraction methods. As a result of experiments on datasets with different characteristics, it is seen that, the proposed method has higher classification performance.Öğe ORACM: Online region-based active contour model(Pergamon-Elsevier Science Ltd, 2013) Talu, M. FatihA new online region-based active contour model (ORACM) is proposed in this paper. The classical geodesic active contour (GAC) model has only local segmentation property, although the Chan-Vese (C-V) model possesses global. An up-to-date active contour model (ACM with SBGFRLS) proposed in Zhang, Zhang, Song, and Zhou (2010) both has the properties of global/local segmentation and incorporates the GAC and the C-V models to raise active contours' performance on image segmentation. However it has two major disadvantages. First, it deforms the active contour model just using the gradient of current level set iteratively and so works too slowly. Second, it needs a parameter a which plays major impact on the results and to be tuned according to input images. The proposed model ORACM eliminates these two disadvantages by using a new binary level set formula and a new regularization operation such as morphological opening and closing. Without changing segmentation accuracy, ORACM requires no parameter and less time over the traditional ACMs. Experiments on synthetic and real images demonstrate that the computational cost of ORACM with the morphological operations is 3.75 times less than the traditional ACMs on average. (C) 2013 Elsevier Ltd. All rights reserved.Öğe Performance Evaluation of Ego-Motion Methods in Static Environment(Ieee, 2015) Akkaya, A. Erhan; Talu, M. FatihEstimating the observer position in 3D environment using a visual image data is called the ego-motion. This method consists of three steps: in the first step, feature points of image are detected; in the second step, optical flow information is obtained; in the third and final step appropriate transformation parameters are calculated using the optimization method. Especially in enclosed areas where GPS can not be used or high estimation errors occured, in underground systems, visual egomotion is critically important. As well as the precise location estimation, providing a low cost solution is one of the superior properties of egomotion method. In this study, different egomotion methods has been observed, the numerical results in noisy and noiseless environment obtained by comparing the accuracy of algorithms presented in the table.Öğe Segmentation of SAR images using improved artificial bee colony algorithm and neutrosophic set(Elsevier Science Bv, 2014) Hanbay, Kazim; Talu, M. FatihThis paper proposes a novel synthetic aperture radar (SAR) image segmentation algorithm based on the neutrosophic set (NS) and improved artificial bee colony (I-ABC) algorithm. In this algorithm, threshold value estimation is considered as a search procedure that searches for a proper value in a grayscale interval. Therefore, I-ABC optimization algorithm is presented to search for the optimal threshold value. In order to get an efficient and powerful fitness function for I-ABC algorithm, the input SAR image is transformed into the NS domain. Then, a neutrosophic T and I subset images are obtained. A co-occurrence matrix based on the neutrosophic T and I subset images is constructed, and two-dimensional gray entropy function is described to serve as the fitness function of I-ABC algorithm. Finally, the optimal threshold value is quickly explored by the employed, onlookers and scouts bees in I-ABC algorithm. This paper contributes to SAR image segmentation in two aspects: (1) a hybrid model, having two different feature extraction methods, is proposed. (2) An optimal threshold value is automatically selected by maximizing the separability of the classes in gray level image by incorporating a simple and fast search strategy. The effectiveness of the proposed algorithm is demonstrated by application to real SAR images. (C) 2014 Elsevier B.V. All rights reserved.Öğe Spiking Neural Network Applications(Ieee, 2017) Celik, Gaffari; Talu, M. FatihSpiking Neural Network (SNN) are 3rd Generation Artificial Neural Networks (ANN) models. The fact that time information is processed in the form of spikes and there are multiple synapses between cells (neurons) are the most important features that distinguish SNN from previous generations. In this study, artificial learning systems which can learn by using basic logical operators such as AND, OR, XOR have been developed in order to understand SNN structure. In SNN, we tried to find optimal values for these parameters by examining the effect of the number of connections between cells and delays between connections to learning success.Öğe Tracking One Dimension State Space Variables With Particle Filter Method(Ieee, 2015) Dilmen, Haluk; Talu, M. FatihParticle filter is among the commonly used methods aims tracking of lineer and non lineer systems. Particle filter takes important place for accurate modeling of nonlinear dynamic systems. Given that the data becomes available instantly, update of the system according to incoming data offers extra gains on better adaptation of instant response and reduces data storage. In this study particle filter investigated on a one dimensional artificial data for the sake of understand theory and working principle.Öğe Unburnt carbon estimation through flame image and gauss process regression(Taylor & Francis Ltd, 2024) Golgiyaz, Sedat; Demir, Usame; Cellek, Mehmet Salih; Daskin, Mahmut; Talu, M. Fatih; Onat, CemThe presence of unburned carbon in coal-burning systems undoubtedly causes a loss in the amount of energy that can be obtained from the system, and also reveals an inadequacy in terms of the usability of the ashes. The expensiveness of existing unburned carbon prediction methods is one of the reasons why these technologies cannot be used. This situation requires working on alternative non-combustible carbon technologies. In this paper, a new approach is presented for estimating unburned carbon in a small-scale coal burner system using the Gaussian regression model and CCD camera-acquired flame image. The proposed approach evaluates brightness, fluctuation amplitude, area, and radiation signal properties of the flame image. The proposed non-combustible carbon estimation technique does not require prior knowledge of CCD camera features. In the feature acquisition phase, results were obtained for each natural component of the flame image in RGB colour space separately, in pairs, all together and for three artificial colour channels (grey image). With the proposed method, the unburned carbon estimation was obtained with an accuracy of R = 0.9664 when all colour channels of the RGB image were used together. This result shows that unburned carbon can be estimated from the instantaneous flame images obtained with the CCD camera.Öğe Web Based Image Processing Application: Rating Diabetes Intensity(Ieee, 2017) Gunduzalp, Veysel; Talu, M. Fatih; Gul, Semir; Zayman, Emrah; Gul, MehmetIn this study, it is A web-based image processing software which has been introduced to process Immunohistochemical images obtaining in experimentally induced diabetic rats and rank the severity of diabetes between the groups. With the software, specialist physicians can upload Images obtaining in rat groups to the system via web on own account, obtain average color intensity and intensity graphs of groups after Determining the basic colors to be evaluated. The software eliminating the subjective evaluation contains mainly three phases in this study, Evaluation of each image content according to basic axes(Three-dimensional projection), Clustering of colors (Expectation maximization method) and Color-axis determination (Calculation of eigenvectors). As a result of, it can be considered that positive results obtained could stimulate Researchers to generalize of the proposed method.
