17 results
Search Results
Now showing 1 - 10 of 17
Conference Object Citation - Scopus: 2Real Time Infrared Image Enhancement;(2012) Akdeniz,N.; Tora,H.This study evaluates the implementation of Balanced Contrast Limited Adaptive Histogram Equalization (BCLAHE) for infrared images (IR) on an embedded platform. The aim was to achieve real time performance for the operator display target application. The system configured for this aim is a dual processor media application device OMAP3530, which consists of an ARM and a DSP processor. System is configured so that hardware sources are used efficiently and various performance improvement techniques are investigated. Performance analysis is done over IR images with different dynamic range. © 2012 IEEE.Article Citation - Scopus: 1Vowel Classification Based on Waveform Shapes(ASTES Publishers, 2019) Tora,H.; Karacor,G.; Uslu,B.Vowel classification is an essential part of speech recognition. In classical studies, this problem is mostly handled by using spectral domain features. In this study, a novel approach is proposed for vowel classification based on the visual features of speech waveforms. In sound vocalizing, the position of certain organs of the human vocal system such as tongue, lips and jaw is very effective on the waveform shapes of the produced sound. The motivation to employ visual features instead of classical frequency domain features is its potential usage in specific applications like language education. Even though this study is confined to Turkish vowels, the developed method can be applied to other languages as well since the shapes of the vowels show similar patterns. Turkish vowels are grouped into five categories. For each vowel group, a time domain speech waveform with an interval of two pitch periods is handled as an image. A series of morphological operations is performed on this speech waveform image to obtain the geometric characteristics representing the shape of each class. The extracted visual features are then fed into three different classifiers. The classification performances of these features are compared with classical methods. It is observed that the proposed visual features achieve promising classification rates. © 2019 Advances in Science, Technology and Engineering Systems.All rights reserved.Conference Object Producing Synthetic Speech From Turkish Text Via a Single Sound Synthesizer Ic;(2010) Tora,H.; Cengizler,Ç.In this study, new speech sounds were created for Turkish letters from the allophones listed in SpeakJet Magnevation complex sound synthesizer IC, which is intended for English, by selecting and paring the most similiar phonems. Consequently, Turkish Text to Speech synthesizer was implemented with minimum enviroment elements and no physical modification on IC. Paring based on SAMPA showed that all Turkish phonems can be obtained from English allophones with high compatibility. The text to speech and paring algorithm presented herein is developed to be embedded in a microcontroller and capable to vocalize a random Turkish text with SpeakJet IC.Conference Object Citation - Scopus: 1Performance Evaluation of Self Organizing Neural Networks for Clustering in Esm Systems;(IEEE Computer Society, 2014) Gencol,K.; Tora,H.Electronic Support Measures (ESM) system is an important function of electronic warfare which provides the real time projection of radar activities. Such systems may encounter with very high density pulse sequences and it is the main task of an ESM system to deinterleave these mixed pulse trains with high accuracy and minimum computation time. These systems heavily depend on time of arrival analysis and need efficient clustering algorithms to assist deinterleaving process in modern evolving environments. On the other hand, self organizing neural networks stand very promising for this type of radar pulse clustering. In this study, performances of self organizing neural networks that meet such clustering criteria are evaluated in detail and the results are presented. © 2014 IEEE.Conference Object Segmentation of Isolated Words Into Voiced-Unvoiced Sound Components by Kurtosis;(Institute of Electrical and Electronics Engineers Inc., 2015) Uslu,B.; Tora, Hakan; Tora,H.; Tora, Hakan; Airframe and Powerplant Maintenance; Airframe and Powerplant MaintenanceThis study presents a new approach to the segmentation of isolated words into their voiced/ unvoiced parts. It is well known that voiced/ unvoiced discrimination has an important role in speech synthesis and coding applications. The offered method makes this discrimination using the kurtosis values of the words. The performance of the proposed approach was tested on Turkish digit recordings from zero to nine. It has been observed that this approach segments the parts successfully in not only clean speech but also in noisy speech. © 2015 IEEE.Conference Object Recognition of Characters on Vehicle License Plates;(2010) Tora,H.; Bora,K.In this study, a simple and effective method is proposed for segmenting alphanumeric and numeric characters on vehicle license plates and recognizing the segmented characters.The proposed approach is basically based on template matching technique. Features used for matching are obtained by scanning the segmented characters from left-to-right, right-to-left, top-to-bottom, and bottom-to-top. The features extracted in this way reveals the fact that how a character is moving and changing along its four-side.The character recognition is accomplished by using this information of the character.Experiments done show that successful results are obtained. ©2010 IEEE.Conference Object Design and Implementation of an Expressive Talking Mobile Robot: Toztorus(Institute of Electrical and Electronics Engineers Inc., 2018) Tozan,O.; Tora,H.; Uslu,B.; Una,B.; Ceylan,E.This paper is about a brand new robot and all its development stages from the design to the show time. As an undergraduate research project (the LAP program at Atilim University), the robot TozTorUs is the outcome of the dense efforts of a team. With the sensors equipped, it navigates autonomously in the environment in which it is located by avoiding the obstacles. It can understand your questions and answer them using Google's speech technologies. Although it is not a humanoid robot, with eyes and mouth simulator LED displays, it is as friendly as a human. We can also control TozTorUs using a mobile phone. Apart from these, it is able to adjust its height with respect to the visitor's, thus allowing it to make an eye contact with the person. Although TozTorUs is designed for welcoming, it may also be employed for consulting, security and elderly assistance. © 2018 IEEE.Conference Object Citation - Scopus: 2The Use of Cumulants for Voiced-Unvoiced Segments Identification in Speech Signals;(IEEE Computer Society, 2014) Uslu,B.; Tora,H.In this study, voiced-unvoiced classification performance of Turkish sounds using skewness and kurtosis is examined. The analyses show that higher order cumulants can be employed as a feature in voiced-unvoiced classification that is vital in speech processing applications. Furthermore, it has been shown that cumulants are also useful for identifying voiced and unvoiced segments in noisy speech signals. © 2014 IEEE.Conference Object Citation - Scopus: 1Higher Order Statistical Analysis of Turkish Phones;(IEEE Computer Society, 2014) Tora,H.; Uslu,B.In this study, histograms of Turkish phones were examined using higher order cumulants. As is known, phones constituting a language, are composed of letters as vowels and consonants. These letters can also be grouped as voiced and unvoiced phones. It is observed that unvoiced letters show a Gaussian-like distribution and result in small values of skewness and kurtosis. On the other hand, vowels and voiced consonants lead to a non-Gaussian distribution. Voiced and unvoiced phones are related with their skewness and kurtosis values. It is empirically shown that higher order cumulants are likely to be a feature in describing Turkish phones. © 2014 IEEE.Conference Object Citation - Scopus: 3Emotion classification using hidden layer outputs(2012) Günler,M.A.; Tora,H.Neural network (NN) with Multi-Layer Perceptron (MLP) is a supervised learning algorithm composed of artificial neurons. Multilayer NN is capable of solving nonlinear classification problems such as emotion identification by using facial expressions that is presented in this paper. Hidden layer outputs of NN provide useful information about facial appearance. This study addresses that without fully training NN hidden layer outputs can be used as feature. It is shown that an acceptable recognition rate is obtained by means of hidden layer outputs. © 2012 IEEE.
