Yazar "Karakaya, Fuat" seçeneğine göre listele
Listeleniyor 1 - 10 / 10
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe An efficient color detection in RGB space using hierarchical neural network structure(2011) Altun, Halis; Sinekli, Recai; Tekbas, Ugur; Karakaya, Fuat; Peker, MuratColor detection is generally a primary stage in most of the image processing application, if the application is based on the color information, such as road sign detection, face detection, skin color detection, object detection and object tracking etc. As the performance of subsequent modules in an image processing application is adversely affected by the previous modules, the accuracy of color detection with a high performance inevitably becomes crucial in some applications. This paper introduces a method for an efficient color detection in RGB space using an ensemble of experts in hierarchical structure. In this structure, a set of experts is assigned to evaluate R, G, B components of a pixel and then constructs a degree of membership to the set of predefined class of colors for the given pixel. Then a master neural network constructs its final decision based on the membership probabilities provided by the set of experts. Qualitative and quantitative evaluations of the results show that the proposed hierarchical structure of neural networks is superior over the conventional neural network classifier in color detection. © 2011 IEEE.Öğe Effect of the duty cycle on the spark-plug plasma synthetic jet actuator(E D P SCIENCES, 2016) Seyhan, Mehmet; Akansu, Yahya Erkan; Karakaya, Fuat; Yesildag, Cihan; Akbiyik, Hurrem; Dancova, P; Vesely, MA promising novel actuator called Spark-Plug Plasma Synthetic Jet (SPSJ) has been developed in Atmospheric Plasma Research Laboratory at Nigde University. It generates electrothermally high synthetic jet velocity by using high voltage. SPSJ actuator can be utilized to be an active flow control device having some advantages such as no moving parts, low energy consumption and easy to integrate the system. This actuator consists of two main components: semi-surface spark plug (NGK BUHW) as an anode electrode and a cap having an orifice as a cathode electrode. The cap, having a jet exit orifice diameter of 2 mm, has diameter of 4.4 mm and height of 4.65 mm. This study presents the characteristics of SPSJ actuator by using the hot wire anemometer in order to approximately determine jet velocity in quiescent air. Peak velocity as high as 180 m/s was obtained for f(e) = 100 and duty cycle 50%. The flow visualization indicated that the actuator's jet velocity is enough to penetrate the developed boundary layer.Öğe FPGA implementation of neuro-fuzzy system with improved PSO learning(PERGAMON-ELSEVIER SCIENCE LTD, 2016) Karakuzu, Cihan; Karakaya, Fuat; Cavuslu, Mehmet AliThis paper presents the first hardware implementation of neuro-fuzzy system (NFS) with its metaheuristic learning ability on field programmable gate array (FPGA). Metaheuristic learning of NFS for all of its parameters is accomplished by using the improved particle swarm optimization (iPSO). As a second novelty, a new functional approach, which does not require any memory and multiplier usage, is proposed for the Gaussian membership functions of NFS. NFS and its learning using iPSO are implemented on Xilinx Virtex5 xc5vlx110-3ff1153 and efficiency of the proposed implementation tested on two dynamic system identification problems and licence plate detection problem as a practical application. Results indicate that proposed NFS implementation and membership function approximation is as effective as the other approaches available in the literature but requires less hardware resources. (C) 2016 Elsevier Ltd. All rights reserved.Öğe Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEY, 2016) Peker, Murat; Altun, Halis; Karakaya, FuatA hardware implementation of a computationally light, scale, and rotation invariant method for shape detection on FPGA is devised. The method is based on histogram of oriented gradients (HOG) and average magnitude difference function (AMDF). AMDF is used as a decision module that measures the similarity/dissimilarity between HOG vectors of an image in order to classify the object. In addition, a simulation environment implemented on MATLAB is developed in order to overcome the time-consuming and tedious process of hardware verification on the FPGA platform. The simulation environment provides specific tools to quickly implement the proposed methods. It is shown that the simulator is able to produce exactly the same results as those obtained from FPGA implementation. The results indicate that the proposed approach leads to a shape detection method that is computationally light, scale, and rotation invariant, and, therefore, suitable for real-time industrial and robotic vision applications.Öğe Hardware implementation of discrete wavelet transform and inverse discrete wavelet transform on FPGA(2010) Çavuşlu, Mehmet Ali; Karakaya, FuatIn this paper, hardware implementation of the Discrete Wavelet Transform (DWT) and Inverse Discrete Wavelet Transform (IDWT) based on FPGA is explained. DWT and IDWT algorithms are implemented on the Altera Cyclone-II FPGA. Filtering processes of rows and columns are seriatim applied as in level-by-level architecture. But both addressing for read/write and DWT/IDWT processes are implemented via only one filter by checking kind of filter to be applied. This usage has got advantages of both elapsed times for read/write processes and cost of hardware area. Implementation DWT and IDWT on the hardware is required only 2% hardware area with this approximation. ©2010 IEEE.Öğe Implementation of Edge Dependent Interpolation Based De-Interlacer on FPGA(IEEE, 2016) Cavuslu, Mehmet Ali; Karakaya, Fuat; Balkoca, AlisanInterlacing technique aims to lower the costs of high definition video systems by reducing the data amount sent to receiver unit. Regeneration of image at the receiver unit is an important point of interlacing method. In this study, regeneration (de-interlacing) of frames that are sent to receiver unit is implemented by using edge dependent interpolation method. The method is implemented using VHDL on Altera Cyclone-II FPGA. The method avoids reading of redundant data which yields to reduced operation time. Implementation occupies only %3 of the FPGA that is used in this study.Öğe Implementation of HOG algorithm for Real Time Object Recognition Applications on FPGA based Embedded System(IEEE, 2009) Karakaya, Fuat; Altun, Halis; Cavuslu, Mehmet AliRecent years HOG algorithm has been used to recognize objects in images, with complex content, with a very high success rate. Hardware implementation of this algorithm is very important because of the fact that it can be used in many object recognition applications. In this work HOG algorithm is implemented on FPGA to recognize different geometrical figures with a very high success rate. Objects vertical and horizontal edges have been sharpened using edge detection algorithms to calculate magnitude and angle of the local gradients. Obtained result are used to construct the histograms of gradient orientation. It is observed that each constructed histogram have distinctive features for every object. Rule based classifiers has been used to implement a successful real time object recognition approach on embedded system.Öğe Neural identification of dynamic systems on FPGA with improved PSO learning(ELSEVIER SCIENCE BV, 2012) Cavuslu, Mehmet Ali; Karakuzu, Cihan; Karakaya, FuatThis work introduces hardware implementation of artificial neural networks (ANNs) with learning ability on field programmable gate array (FPGA) for dynamic system identification. The learning phase is accomplished by using the improved particle swarm optimization (PSO). The improved PSO is obtained by modifying the velocity update function. Adding an extra term to the velocity update function reduced the possibility of stucking in a local minimum. The results indicates that ANN, trained using improved PSO algorithm, converges faster and produces more accurate results with a little extra hardware utilization cost. (C) 2012 Elsevier B.V. All rights reserved.Öğe SIH: segmented intensity histogram for orientation estimation in image matching(SPRINGER LONDON LTD, 2016) Peker, Murat; Karakaya, FuatIn this paper, we propose a fast and effective new method to reduce the overhead cost of orientation estimation. The proposed method uses the summation of intensity values from segments of image patches and forms a histogram based on those values. As a result, it is faster than SIFT-like algorithms because it does not require computation of gradient orientations and magnitudes. Also, it is as fast as other intensity-based algorithms with better image matching performance. Proposed method could be easily integrated to any image matching algorithms. Test results indicate that SIFT integrated with proposed orientation estimation method produces comparable results as the original multi-angle SIFT algorithm with less execution time.Öğe The Effect of Genetic Algorithm Parameters on the Solution of Plate Location Detection(IEEE, 2008) Peker, Murat; Altun, Halis; Karakaya, FuatIn this study, a new method based on genetic algorithm and neural networks for determining licence plate location is proposed. The effect of genetic algorithm parameters on the quality of solutions is investigated. The method is able to successfully locate a licence plate in avearge 40 msn, on the image of 768x288 size. This score is 200 times quicker compared to sequential search method. Futhermore the method is able to find multiple plates in an image.
