Gökçen Çetinel

Blind equalization of single-input single-output fir channels for chaotic communication systems

Recently we have developed a simplified recursive adaptive blind channel equalization method for Single-Input Single-Output (SISO) chaotic communication systems. Even though the simplified recursive algorithm gives superior results compared to the state of the art chaotic blind channel equalization algorithms, it has a very important limitation: convergence of the adaptive algorithm is ensured for only Strictly Positive Real (SPR) channels. In this study, we propose a non-recursive chaotic blind channel equalization algorithm that works regardless of whether the channel is SPR or not. First, a statistically optimum fixed filter is designed assuming that the channel is known. Then, it is shown via computer simulations that its performance is very close to that of the statistically optimum fixed filter. Furthermore, it gives better results especially for non-SPR channels compared to the well-known minimum nonlinear prediction error method and the simplified recursive algorithm developed in our previous work. The method is computationally simple and does not impose any restrictions on the channel other than being a finite impulse response filter. Since the instantaneous gradient is used to derive the adaptive algorithm, the proposed method works for slowly and smoothly varying linear channels as well.

Chaotic digital image watermarking scheme based on DWT and SVD

In this study chaos based digital watermarking scheme together with Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) is proposed. In the proposed watermarking scheme, the cover image is decomposed to its sub-bands (LL, LH, HL and HH) by a common used frequency domain transform: DWT. Then, the SVD is directly applied to the all sub-bands of the decomposed cover image. The watermark is shuffled with Arnolds Cat Map (ACM) to generate a chaotic watermark. By this way, the robustness and perceptual invisibility of the scheme is improved. In order to evaluate the robustness of the proposed scheme, several image processing and geometric attacks are applied to the scheme. The Normalized correlation (NC) and peak signal-to-noise ratio (PSNR) measures are used to show the performance of the proposed method in terms of robustness and perceptual invisibility. The proposed algorithm gives the promising results and meets the security requirements.

Meme Manyetik Rezonans Görüntülemede Lezyon Sınırlarının ve Alanının Tespit Edilmesi

In this study, we have developed a system for determining the boundaries of the lesions which come into existence in the breast and calculating the lesion area by using images obtained from the MRI system, which is one of the modalities widely used in diagnosis of the breast cancer. The developed system is designed with an interface that provides great convenience to the radiologists and offers many interchangeable options. In order to determine the boundary of the lesion and to calculate the area optimally, in this study four different methods are utilized. These methods are thresholding based (Otsu thresholding method), fuzzy logic based (fuzzy cmeans, FCM), region growing based (Region Growing, RG) and cluster-based (k-means) segmentation methods. The Otsu, FCM and RG methods are single-channel gray-level segmentation methods. In case, the k-means method is a method of segmentation that can be used directly 1 gulsewda@hotmail.com, 2 gcetinel@sakarya.edu.tr * Corresponding Author 0108 Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 22 (1), 108~127, 2018

Identification and grading of spasticity by using AdaBoost and k-NN techniques

Spasticity is a disorder that causes movement constraints in daily life and induced by upper neuron syndrome. Although spasticity is a well-known motor disorder, there are related diagnostic and assessment problems. The lack of common rating in assessment of spasticity causes significant differences between the interpretations of the specialists. In this study, it is aimed to conduct a highly accurate evaluation based on the Ashworth scale, which is one of the most accepted scales for assessment and grading of the spasticity. In this study, Electromyogram (EMG) signals obtained in Rectus Femoris and Biceps Femoris muscles are used to trigger Patela-T reflexes in spastic patients. 26 features are derived by calculating five features in both time and frequency domains from measured EMG signals and six features of Pendulum movement to recognize the normal person and patients with Ashworth 1-Ashworth 2 spasticity levels. Finally, spasticity is graded by using k-nearest neighbor (k-NN) algorithm and AdaBoost method. The performance of the proposed study is evaluated by calculating accuracy, sensitivity, precision and specificity criteria. Simulation results are demonstrated that the study has crucial importance to assess and grade the spasticity.

An EOG based communication channel for paralyzed patients

In this study, Electrooculogram (EOG) based Human-Machine Interface (HMI) application is proposed for partial or completely paralyzed and physically limited people as Amyotrophic Lateral Sclerosis (ALS) patients. In the designed system, EOG signals consisted of vertical and horizontal eye movements are detected by using 6 Ag-AgCl electrodes which placed around the eye. Then, by applying amplifying and filtering processes to the detected signals, EOG signals at 0-+5V amplitude levels are obtained at the output of analog stage. In consequence of digital processing of analog EOG data with microcontroller unit, control signals for HMI applications are acquired. At signal processing stage, after preprocessing step for eye movements (vertical, horizontal and blink) maximum and minimum voltage amplitude values are detected. These values are directly determine the performance of the classification process. K-Nearest Neighbor (k-NN) classifier and Support Vector Machines (SVM) are used for classification. According to the results, k-NN and SVM perform the classification task with %90.3 and %92.6 accuracy, respectively. Simulation results show that with the proposed EOG based HMI system, physically limited patients can communicate with their environments in a successful manner.

RDWT and SVD based secure digital image watermarking using ACM

In this study, digital image watermarking method based on Redundant Discrete Wavelet Transform (RDWT) and Singular value Decomposition (SVD) is proposed. By exploiting the complexity property of the chaotic signals, which is the main reason of using them in the security applications, the robustness and the invisibility of the proposed method is improved. In the proposed watermarking scheme, RDWT is performed to decompose the cover image into four sub-bands (LL, LH, HL and HH). Then SVD is applied to the LL sub-band of the cover image. The chaotic watermark is generated by applying Arnold Cat Map (ACM) to the original watermark. Finally, singular values of the LL sub-band of the cover image are modified with the singular values of the chaotic watermark according to the embedding process. It is demonstrated via computer simulations that the proposed RDWT, SVD and ACM based digital image watermarking scheme provides better watermark concealment and high robustness against both geometric and image processing attacks. Furthermore, the robustness of the proposed scheme is investigated for different dimensions of the binary watermark logo. It is shown that the robustness of our method is independent of the watermark size.

Hybrid Biometric System Using Iris and Speaker Recognition

In this study, a hybrid security system is proposed. The proposed system is composed of two subsystems namely iris recognition system (IRS) and speaker recognition system (SRS). Pre-processing, feature extraction and feature matching are the main steps of these systems. In IRS subsystem, Gaussian filter, Canny edge detector, Hough transform, and histogram equalization is performed for pre-processing, respectively. After that, by applying 4-level Discrete Wavelet Transform (DWT) to pure iris image, the iris image is decomposed into four sub-bands (LL4, LH4, HL4 and HH4). In order to extract the feature vector from iris pattern, the LH4, HL4 and HH4 sub-bands (matrices) are merged into one matrix. Finally the matrix is transformed in vector to obtain the feature vector of iris image. For SRS subsystem, the pre-processing step includes spectral arrangement, silence part removing and band limitation operations. After pre-processing, frame blocking and windowing are applied to the long-term speech samples and then Fast Fourier Transform (FFT) is performed for the each short-term speech segments (frames). Finally, the Mel Frequency Cepstral Coefficients (MFCC) technique is performed in order to obtain feature vector of the speech. The feature matching step of both IRS and SRS is implemented with Dynamic Time Warping (DTW) which is an efficient algorithm to measure the distance between two vectors. According to the DTW results, the false acceptance rate (FAR) is zero and false rejecting rate (FRR) is about 4 % for the proposed hybrid system.

Detection and Counting of Embryonic Stem Cells in Fluorescence Microscopy Images by a Fully Automatic Method

In this paper, an automatic cell counting method under microscopy is proposed. The cell counting process can be performed in two ways: The manual counting in which a specialist counts the cells with naked eye, and the automatic counting that utilizes the computer-based techniques. In manual counting, there are several techniques for dying the cells to turn them visible with naked eye. However, if the concentration is more than normal the cells can overlap. Overlap and incorrect adjusted microscopy parameters are the main factors that cause inaccurate counting results. Furthermore, in manual counting inter-observer variability is high. Even though the same cell image is taken into account by the different specialist, different counting results can be obtained. Because of the above mentioned problems, the cell counting process must be performed automatically. The proposed automatic stem cell counting process is based on image processing techniques that appropriate the frame of method. At first, stem cell sections were obtained under the fluorescence microscopy. In the following pre-processing step Gaussian filtering and background extraction are performed. Before applying watershed algorithm histogram of the image is partitioned in to four parts and the best combination is determined to obtain the most exact counting results. The aim of using watershed algorithm is to make the boundaries and maximum points of the cells more clear. Finally, spherical contours corresponding to the stem cells are counted. The effectiveness of the proposed method is evaluated by performing numerous computer simulations. It is shown that the proposed method gives promising results and can eliminate the subjectivity originated from the manual counting. The method is tested on a database contains two image groups at different noise levels validated by the specialists.

Blind channel equalization of MIMO FIR channels for chaotic communication systems

Recently various blind channel equalization algorithms have been developed for chaotic communication systems, by exploiting some properties specific to chaotic signals. However, communication channel is assumed to be a single-input single-output (SISO) system in most of these studies. There is no blind channel equalization method for multiple-input multiple-output (MIMO) unknown channel case. In this study, we propose an adaptive blind channel equalization algorithm for MIMO chaotic communication systems. In MIMO communication system, multiuser interference (MUI) is an important factor as well as intersymbol interference (ISI) that hinders receiver performance. To increase receiver performance and achieve reliable, high-speed communication a novel method that overcome the effect of ISI and MUI is developed. The effectiveness of proposed algorithm is demonstrated by computer simulations.

Blind channel equalization of multiple-input/single-output FIR systems for chaotic digital communication systems

Recently, Various blind channel equalization techniques for chaotic communication systems have been developed. However, in most of these studies communication channel is assumed to be single-input single-output (SISO) system. To the best of our knowledge, there is no study about blind equalization of multiple-input multiple-output (MIMO) channel case. In this study, an adaptive blind channel equalization algorithm for multiple-input single-output (MISO) chaotic communication system has been proposed. The channel is modelled as a MIMO finite impulse response (FIR) system and the equalizer is designed as a MISO FIR filter. The nonlinear predictability of a chaotic signal is exploited to derive the algorithm and the equalizer coefficients are updated by the proposed algorithm.