Karasu, Seçkin | Altan, Aytaç | Saraç, Zehra | Hacıoğlu, Rıfat

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

In this study, the wind speed prediction model is created that gives a minimum error for different hidden layer neuron numbers and delay step numbers. Using the one-minute time series, the prediction of the next wind speed is performed with the NAR neural network model. The predicted values of wind speed obtained are compared with predicted values of wind speed obtained with filter methods. For different window functions and lengths, wind speed prediction is made using filters with different weight coefficients. For the number of hidden layer neurons is 14 and the number of delay steps is 10, MAE, MSE and RMSE values are calculated . . .as 0.0315, 0.0019, 0.0445, respectively, with NAR neural network. It is seen that the proposed method for the wind speed dataset has a higher prediction performance than thefilter methods. © 2017 IEEE Daha fazlası Daha az

Onur, Tuğba Özge | Hacıoğlu, Rıfat

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Reverberation and multipath propagation afîect the accuracy of time of flight (TOF) estimation in confined underwater medium. In this paper, time delay has been estimated by obtaining the transmitted signal and first echo signal from measured signal acquised by using a wideband transducer with 5 MHz central frequency. Cross-correlation (CC) and generalized cross-correlation (GCC) methods have been applied to the obtained signals and their performances have been compared. The results show that GCC method has better performance for time delay estimation in confined underwater medium where multipath propagation and reverberation are se . . .vere. © 2017 IEEE Daha fazlası Daha az

Altan, Aytaç | Hacıoğlu, Rıfat

Conference Object | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

This study focuses on the modelling of 3 axis gimbal system with the RRR joint structure on the Unmanned Aerial Vehicle (UAV), which is autonomously moving for the target tracking, based on experimental input (motor velocities) and output (end effector position) data. The fact that UAVs move in a certain direction and that the camera on the end effector of the gimbal system on it is adhere to the correct target attracts many researchers. The transfer function of the 3 axis gimbal system is obtained by linearly structured OE-Output Error model using experimentally obtained data under different external disturbance effects. Model degr . . .ee is determined and data set based verification is applied. Also, the performance is compared by examining the effect of external disturbance in the transfer function obtained. © 2017 IEEE Daha fazlası Daha az

Narin, Ali | Özer, Mahmut | İşler, Yalçın

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Paroxysmal Atrial Fibrillation (PAF) is a very common rhythm disorder that causes rapid and irregular impulses in the heart. In this study, it is aimed to determine whether patients can be warned before PAF events. 30-minute HRV data used in this study. Each piece of data was divided into 10 pieces of 5-minute parts. Time domain measurements from linear measurements of HRV and Poincare measurements from nonlinear measurements of HRV were used for each segment. Detecting performances were measured for each segment using k-nearest neighbor classifier. Particularly linear measurements have been shown to achieve up to 82% success in pre . . .dicting PAF attack and was observed that PAF attack could be detected 12,5 minutes earlier. © 2017 IEEE Daha fazlası Daha az

Ağaoğlu, Şükrüye Nihal | Özer, Mahmut | Çalım, Ali | Uzuntarla, Muhammet

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

In this paper, the phenomena of Vibrational Resonance is investigated in an excitable system which consists of FitzHugh-Nagumo neurons with electrical coupling. Weak signal detection performance of excitable system is examined in scale-free network (unweighted or weighted) topology. The simulation results show that; weighting the scale-free network, average connectivity degree, amplitude and frequency of weak signal play an active role to determine the data carrying performance of neurons based on Vibrational Resonance. It is determined that, the amount of required energy for creating resonance peaks of excitable system is decreased . . . significantly by choosing the correct value of weight control parameter in a weighted network especially. © 2017 IEEE Daha fazlası Daha az

Çalım, Ali | Özer, Mahmut | Uzuntarla, Muhammet

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Nowadays, neurodegenerative diseases which affect human life quite negatively with motor, cognitive and psychiatric disorders are becoming widespread. One of the most common neurodegenerative disorder is Parkinson's disease. Recent electrophysiological experiments have shown that Basal Ganglia, a special region in the midbrain, is related to Parkinsonism. Beta frequency oscillations, which are important symptoms of Parkinson's disease, emerge intensively in Globus Pallidus and Subtalamus nuclei. In this study, anatomical connections of Globus Pallidus and Subtalamus are constructed computationally, and the cellular properties that g . . .ive rise to emergence of beta oscillations are investigated. © 2017 IEEE Daha fazlası Daha az

Aydemir, Zübeyr | Erkaymaz, Okan | Ferah, Meryem Akpolat

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

In this study, the features of the seminiferous tubule sections were extracted and the presence of the cells and cell stain types detected with the help of the feed forward artificial neural network. By looking at the section view with a small window, 78 features were extracted from the pixels seen by the window and used as an input to the artificial neural network. Artificial neural network outputs are decides presence of the cell and the staining of the cell. The results obtained with the artificial neural network were determined by using the connected component labeling method. The results obtained with the help of the user and t . . .he results obtained with the artificial neural network were compared. It has been shown that the proposed ANN model performs cell counting process comparable to the literature (%76 accuracy). © 2017 IEEE Daha fazlası Daha az

Çalım, Ali | Ağaoğlu, Şükriye Nihal | Uzuntarla, Muhammet

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Vital functions which take place in the brain are accomplished through synchronized neural activities between interconnected neuronal populations. In healthy and unhealthy nerve system, rhythms in different frequencies can be recognized as they are responsible for neural synchronization. Despite this, in the case of global synchronization where oscillation power is the maximum, as observed in epileptic seizures, neural activity can terminate. In this study, effects of neural network characteristics synchronization emerges from on activity termination are computationally investigated. Stochastic Hodgkin-Huxley (H-H) equations are use . . .d in the simulations. The results show that synchronization increases when neurons have high synaptic conductance and due to this, strong synaptic current neurons exposed to terminates the firings. © 2017 IEEE Daha fazlası Daha az

Uzun, Rukiye | Özer, Mahmut

Proceedings | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

We investigate how firing dynamics of neuron changes with the autpse's coupling strength for different ratio of the blockage of potassium and sodium ion channels embedded in membranes affects, with using a biophysically more realistic Hodgkin-Huxley neuron model. In the study, we assume that neuron has an excitatory chemical autapse. It is found that the neuron dynamics does not change for too small coupling strength whether it increases after a certain coupling strength' value in case of different ion channel blockage case. Besides, it is determined that potassium ion channel blockage has a healing effect whereas sodium ion channel . . . blockage has a destructive effect on the neuron dynamics increasing with the autaptic coupling strength. © 2017 IEEE Daha fazlası Daha az

Tanyeri U. | Incetas M.O. | Aydemir Z.

Conference Object | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

Interpolation techniques are used to obtain values that are unknown from known values in functions whose variation is within a certain range. Image processing methods use the same approaches when enlarging and reducing the image. However, some corruptions can be seen in the result image. Likewise, there are degradations due to environmental factors or sensors in the process of acquiring or moving images. This study investigates the filtering performance in interpolated noisy images where both states are active and discusses the situation in interpolated distorted pictures. © 2017 IEEE.

Abdikan S. | Ustuner M. | Sanli F.B. | Bilgin G.

Conference Object | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

In this study, L-band ALOS PALSAR radar satellite image and Landsat TM optical satellite image were used to investigate the contribution of radar satellite image to optical satellite image for land cover mapping. Dual-polarimetric data of ALOS satellite and also normalized difference vegetation index (NDVl) generated from Landsat image were used for the analysis. In addition, different classification techniques were taken into consideration and forest dominated land cover maps were produced and the results were compared. Random Forest (RF), k-Nearest Neighbors (k-NN) and Support Vector Machines (SVM) approaches were applied as image . . . classification techniques. While the best result among the methods is DVM, the data set in which combined data are used gives the best general accuracy result. © 2017 IEEE Daha fazlası Daha az

Ustuner M. | Sanli F.B. | Bilgin G. | Abdikan S.

Conference Object | 2017 | 2017 25th Signal Processing and Communications Applications Conference, SIU 2017

In this study, Sentinel-1A SAR imagery for land use/cover classification and its impacts on classification algorithms were addressed. Sentinel-1A imagery has dual polarization (VV and VH) and freely available from ESA. Istanbul was selected as the study region. After the pre-processing steps including the applying the precise orbit file, calibration, multilooking, speckle filtering and terrain correction, the imagery was classified as the following step. Three classification algorithms (SVM, RF and K-NN) were implemented and the impacts of additional bands (VV-VH, VV+VH etc.) were investigated. Results demonstrated that highest clas . . .sification accuracy of this study was obtained by SVM classification with the original bands (VV and VH) of Sentinel-1A imagery. Moreover, it was concluded that additional bands had different impacts on each classifier within accuracy. © 2017 IEEE Daha fazlası Daha az