Yüksek Lisans Tezleri

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/15956

Browse

Filters

2002 - 200534

Settings

COAR Access Rights: search.filters.coarAccessRights.metadata only accessDepartment: search.filters.department.Çankaya Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Bölümü

Search Results

Now showing 1 - 10 of 34
  • Master Thesis
    Fingerprint detection in digital images
    (2004) Fakılı, Defne
    öz Dijital resimlerde parmakizi saptanması Fakılı, Defne Yüksek Lisans, Bilgisayar Mühendisliği Tez Yöneticisi: Prof.Dr. Ziya Güvenç Haziran 2004, 65 sayfa Parmak izinden yola çıkarak kimlik belirlemesi, kimlik belirlemek için kullanılan yöntemlerden en güvenilir olanlarındandır. Çünkü her kişinin kendine özgü farklı bir parmak izi vardır ve bu yaş ilerledikçe değişmez. Doğumdan ölüme kadar aynı şekil özelliklerini korur. Parmak iziyle kimlik belirlemesi eski Çin devletinde ilk olarak uygulanmaya başlamıştır. Bu tezin birinci konusu parmak izinden kimlik belirlemede yapılacak gerekli işlemlerden oluşmaktadır. Bu tezde parmak izi kimlik belirlemesinde gerekli olan ilk üç aşama üzerinde durulmaktadır. Bu aşamalar sırasıyla; hatalardan arındırma (filtreleme yoluyla), keskinleştirme, bulanıklaştırarak keskin ve gereksiz detayları yumuşatma, imajı binary imaj formatma çevirme ve parmak izi çizgilerini inceltme aşamalarıdır. Bu tezde parmak izi algılaması için gerekli olan aşamalardan, bu aşamalar bahsedilmektedir.Bu tezin ikinci konusu ise bir dijital imajda parmak izini saptama için gerekli işlemlerden oluşuyor. Bir dijital imajda parmak izinin ve arka planda yazıların olduğu düşünülürse bu parmak izini saptamak ve ayırmak tezin ikinci konusudur.
  • Master Thesis
    Component-oriented software engineering modeling tool
    (2004) Çelik, Sezai Gökçe
    oz BİLEŞEN YÖNELİMLİ YAZILIM MÜHENDİSLİĞİ MODELLEME ARACI Çelik, Sezai Gökçe Yüksek Lisans, Bilgisayar Mühendisliği Bölümü Tez Yöneticisi: Yrd. Doç. Dr. Gülsün TÖRECİ Eylül 2004, 71 sayfa Bileşen Yönelimli geliştirme yeniden kullanımla yalandan ilgilidir. Temel fikir basittir. Yeni sistemler geliştirilirken önceden geliştirilmiş bileşenler kullanılır. Eğer bir sistem için özel bir fonksiyon geliştiriyorsanız, onu daha sonra başka sistemlerde de kullanılacak şekilde geliştirmek gerekmektedir. İlke olarak basit gibi görünse de geliştirmenin gerçekten zor oluğu görünmüştür. Bu çalışmada, bileşen yönelimli yazılım mühendisliğinin geçmişi ve esasları incelenmiştir. Bu esaslar doğrultusunda geliştirilmiş olan COSEML (Bileşen Yönelimli Yazılım Mühendisliği Modelleme Dili) temelleri esas alınarak, eksikliği hissedilen modelleme aracı tasarlanıp, geliştirilmiştir.
  • Master Thesis
    Specifications of object oriented database systems and performance test of an object oriented database engine
    (2002) Can, Barbaros
    1990'ların başından bu yana nesne-yönelimli programlama bilgisayar dünyasında oldukça popüler hale gelmiştir. Bu eğilimin etkisiyle, geleneksel veri tabanı sistemlerinin eksikliklerini gidermek amacıyla nesne-yönelimli veri tabanı sistemleri ortaya çıkmıştır. Bu tez nesne-yönelimli veri tabanı sistemlerinin özelliklerini açıklamakta ve nesne-yönelimli veri tabanı sistemleri ile ilişkisel veri tabanı sistemleri arasında bazı karşılaştırmalar yapmaktadır. Tezin ilk iki bölümü çalışmanın teorik alt yapısını verirken, tezin üçüncü bölümü örnek bir uygulamaya dayanarak sistemleri karşılaştırmaktadır. Bu uygulamada nesne-yönelimli veri tabanı olarak db4o, büyüklük olarak her ikisi benzer ölçülerde olduklarından, ilişkisel veri tabanı olarak da MySql seçilmiştir. Uygulamanın programlanması için ise, tam bir nesne yönelimli programlama dili olan Java kullanılmıştır
  • Master Thesis
    Implementation of DOA by SR
    (2005) Yengel, Emre
    This study describes a new estimation method used for low signal to noise ratio in non-Gaussian noise. A nonlinear estimator is developed by combining a nonlinear processing method called Stochastic Resonance with Wiener filter. In the method, the recieved signal is first quantized by a symmetric 3-level quantizer and processed by the Wiener filter. Simulation results show that under sufficiently low signal to noise ratio conditions, this quantizer-estimator leads an enhancement of the estimation performaned by the Wiener filter. Non-Gaussian noise distributions that are used in simulations are relevant for an underwater acoustic enviorement. This kind of estimators can be used in Direction Of Arrival (DOA) applications
  • Master Thesis
    Comparative study of human face identification methods
    (2005) Karaömeroğlu, Betül
    Face recognition algorithms have gained popularity in the recent years. Nowa days there are a lot of face recognition systems. However, it is known that these systems work properly in ideal environments. Behind of that, these applications provide low performance in some situations such as illumination, occlusion, different facial expression or presence of rotation and tilt conditions. In this thesis, a new approach to face recognition problem has been proposed. The presented system is a specialized version of PCA augmented with Gabor Wavelet Transform. Firstly, 2D Gabor Wavelet Transform is applied to cope with the variations due to illumination and facial expression changes; then the modified PCA approach is used for reducing a large set of correlated variables iii to a small number of uncorrelated components. The performance of the proposed algorithm is compared with the other algo rithms based on the effect of the illumination, facial expression and occluding objects such as eye glasses and facial hair
  • Master Thesis
    GTD diffraction coefficient for higher order edge diffracted fields
    (2005) Küçükoğuz, Celal
    Diffraction is the change in the direction of a wave at the edges of the objects. Multiple diffraction occurs when there are two ör more edges in öne ör more objects in same geometry. If there are öne ör more interaction between these edges, this type of diffraction is called multiple diffraction. There are several methods of calculating diffraction. Öne of them is geometrical theory of diffraction which is the first ray based method which considers diffraction. Rays are diffracted from edges and affect the total field according to this method. Another method is uniform theory of diffraction which removes the inconsistencies of geometrical theory of diffraction in transition regions. iii This method uses Fresnel fünctions to overcome this problem. There is also uniform asymptotic theoıy of difrraction which modifies geometrical theory of dif&action with the same objective and uses Fresnel functions. But, this method uses Fresnel functions in a different manner. Another öne is physical theory of dif&action which corrects physical optics fîeld just like uniform theory of diffraction corrects geometrical optics. in this thesis study, uniform theory of diffraction is used to calculate dif&action, and multiple difrraction is expressed in a pointwise manner which is a new perspective on this subject. This new approach is compared in two different geometries which both contain two perfectly conducting slits and öne aperture between them. Effect of the number of interactions and effect of the aperture length on the total field is discussed in this thesis. Matlab software is used to plot the electromagnetic fields
  • Master Thesis
    Appliying neural networks to character recognition
    (2005) Shertil, Mahmud Saad
    Optical Character Recognition (OCR) is one of the most widely used applications of automatic pattern recognition and it is a very active research field since the 50's. Today there are numerous algorithms that perform this task, each with its own strengths and weaknesses. In this thesis we explore neural networks to find the best topology regarding the number of layers and the number of neurons at each layer by classifying two different datasets using backpropagation networks with two different approaches. We then compare the results between them and show that the obtained optimum set of parameters is valid for both printed letter and the handwritten digits. Moreover the letter dataset consists of sixteen features with 20000 samples and twenty-six classes. Here we see that the number of classes is more than the number of features which affects our results. The second dataset which is handwritten digits consists of sixty-four features with 5620 samples and ten classes. Here we see that the results are better than the first dataset
  • Master Thesis
    Access telecommunication systems
    (2005) Küçükoğlu, Afşin
    This thesis studies the differences between xDSL, LMDS, Fiber and FSO. Cable access system XDSL is compared with the optical fiber access communication method. As wireless access communications, LMDS and FSO systems are compared. Finally, a full comparison of all the key points for all the access systems are made. Issues compared cover the state of the art, information rate, communication distance, security, commercial prices, and the system installation costs. The comparison is supported by a software which is running under Unix systems on the internet. In the second part, atmospheric structure and the atmospheric molecular effects on the light waves in different ranges are studied. Temperature, C02 and water molecular effects in the atmosphere and the effects on the intensity are also studied
  • Master Thesis
    Pattern recognition: Comparison study
    (2005) Salim, Fawzı
    Öne of the most important effects the field of Cognitive Science can have on the field of Computer Science is the development of technologies that make our tools more human. Evidenced by the fact that we are not currently ali using Tablet computers, accurate hand writing recognition is clearly a difficult problem to solve. Neural netvvorks field is interested with applications for many practices such as industrial process, marketing, medicine, business which is also relevant with hand writing recognition (our main study in this thesis). Hand writing recognition is an important field with applications in business 'form-filling1, including handwritten postal addresses, cheques, insurance applications, mail-order forms, tax returns, credit card sales slips, customs declarations and many others. These applications ali generale a handwritten script from an unconstrained population of writers and writing implements, which must subsequently be processed off-line by computer. To consider at the importance of neural networks and its applications, we used it in this research and we applied backpropagation algorithm to give us better results. We will present more details in the thesis especially in chapter four. The process of recognizing of handwriting from pixel information falls into a field of artificial iii l intelligence called pattern ör image recognition. Lots of work has been done in this field recently, and most techniques for pattern and image classifıcation make use of neural networks. This work implements neural networks in order to "learn" to recognize general features of hand vvritten digits using the well know backpropagation algorithm
  • Master Thesis
    Analysis of natural language sentences using natural language processing methods
    (2005) Bolat, Saadettin
    The aim of this study is to develop a Web-based Question Answering System, which primarily focuses on open domain fact-based short answer questions and also some definition questions. The system takes the advantage of Web which contains enormous amount of documents available online and the redundancy (multiple answer occurrences) available in the Web documents. In order to find a correct answer to a question, the first step is to determine the type of answer that the question asks for. The system analyzes and classifies the questions according to the hand-made question analysis templates. These question analysis templates are based on the question words, English grammar and some other keywords in the questions. It first creates possible sentence substrings and then search engine specific queries. To analyze the retrieved search results, the system applies a series of filters to each sentence containing the iii candidate answer and a score is given to each sentence. Among all candidate sentences, top 5 ranked sentences containing the answers are returned as possible answers. The performance of our system has been compared to the AnswerBus question answering system which is an online, publicly accessible system on the Web, and similar to our approach in several aspects. According to the evaluation results, our system appeared better than AnswerBus in terms of accuracy