İman Askerbeyli | Ankara University (original) (raw)

Papers by İman Askerbeyli

Arabian Journal for Science and Engineering

Recently, subjects such as human recognition (HR), age estimation and gender recognition have bee... more Recently, subjects such as human recognition (HR), age estimation and gender recognition have been among the most investigated human–computer interaction topics, in both the academic and other fields. HR is a process in which a person is detected based on the obtained biometrical features. In this study, a new feature extraction method has been suggested through using the signals received from the sensors of the accelerometer, magnetometer and gyroscope attached to the 5 areas on the human body. The feature extraction from the signals is one of the most crucial stage. The reason behind the success of HR is based on the extracted features. However, the extraction of appropriate features for HR is a challenging issue. Various transformation methods like 1D-LBP and 1D-FbLBP have been applied to the sensor-based signals. Following the transformation process, the statistical features have been acquired from the newly developed signals. The classification processes have been carried out with the distinctive methods concerning machine learning (Knn, RF, A1DE, A2DE and ANN) by using these features. According to these results, 1D-LBP (88.4649%) and 1D-FbLBP (91.8281%) methods have been chosen to provide effective features for HR.

ArXiv, 2018

The purpose of this study is to successfully train our vehicle detector using R-CNN, Faster R-CNN... more The purpose of this study is to successfully train our vehicle detector using R-CNN, Faster R-CNN deep learning methods on a sample vehicle data sets and to optimize the success rate of the trained detector by providing efficient results for vehicle detection by testing the trained vehicle detector on the test data. The working method consists of six main stages. These are respectively; loading the data set, the design of the convolutional neural network, configuration of training options, training of the Faster R-CNN object detector and evaluation of trained detector. In addition, in the scope of the study, Faster R-CNN, R-CNN deep learning methods were mentioned and experimental analysis comparisons were made with the results obtained from vehicle detection.

Artificial Intelligence, which works on the ability to learn in machines, has a widespread field ... more Artificial Intelligence, which works on the ability to learn in machines, has a widespread field of research. One of the most researched topics of artificial intelligence is artificial neural networks. Artificial neural networks are effective today with the solution of complex problems, calculation and processing of information. Seismic method, which is one of the basic applications of geophysical field, is widely used especially for the detection of oil by using seismic waves. With the literature review, it is seen that the types of artificial neural network architectures are used. It has been determined that different methods are used in the processing of seismic data. Using the convolutional neural network (CNN), one of the artificial neural network architectures, it is aimed to achieve success in oil detection by seismic waves.

Social network is an important part of people’s daily life, which essentially can be considered a... more Social network is an important part of people’s daily life, which essentially can be considered as a black box, providing great insight into the current trends of people. It has been proved that data obtained from Social Network platforms, including Facebook, Twitter, Instagram, Baidu etc., can be employed in a variety of different fields. Technically, social networks can be shown as nodes and edges, inspired from graph theory, this theory play an important role to express the structure and relations of the network. Big data mining, natural language processing (NLP), text mining and machine learning techniques are mainly utilized for processing and analysing the data. The analysis of social web can reveal the real trends of the people on any subjects, can detect terror activities before attacks, can help to understand political tendencies, cultural or global believes etc. This study is a review of social web mining that aims to examine the literature of social web mining and help to...

Kayalarin petrografik ozellikleri; olusum sekillerini, olusum kaynaklarini ve hangi mineral toplu... more Kayalarin petrografik ozellikleri; olusum sekillerini, olusum kaynaklarini ve hangi mineral topluluklarindan olustugunu ortaya koyar. Kayalarin icermis olduklari minerallerin durus sekilleri ve bicimleri mikroskop altinda gozlemsel olarak tanimlanarak kayalarin adlandirilmalari yapilmaktadir. Mineral oranlari nokta sayma yontemi ile tane boyutlari ve sekillerinin tanimlanmasi ise gozlemsel olarak yapilmaktadir. Bu islemler sirasinda ayrimlama, tanimlama ve sayma gozlemsel olarak yapildigi icin hatalar yapilmaktadir. Bu calismada; yukarida siralanan islemleri yapma sirasinda meydana gelen hatalari ortadan kaldirmak ve incekesit dijital goruntulerini talep edilen ozelliklere uygun olarak otomatik islemek icin algoritmalar tasarlanmis ve Grafik Kullanici Arayuzlu program paketi hazirlanmistir. Bu amacla, ilk asamada arastirilan araziden ornekler toplanmis, laboratuar ortaminda orneklerden ince kesitler yapilmis, sonra mikroskop yardimi ile dijital goruntuler (dijital incekesitler) elde edilmis ve son olarak bu goruntuler tasarlanan algoritmalar ve gelistirilen yazilim ile bilgisayar ortaminda otomatik olarak islenmistir. Gelistirilen yazilimin ozelliklerinden bazilari sunlardir: 1) olcek yerlestirilebilir; 2) mineraller olusturulabilir; 3) mineral bilesenlerinin sinirlari bagimsiz olarak belirlenebilir; 4) zoom islemleri yapilabilir; 5) tanecik ekleme ve silme islemleri yapilabilir; 6) her bir mineral bilesenine farkli renk atana ve degistirilebilir; 7) bilesenlerin sayisi, alan ve eksen olcumleri yapilabilir; 8) goruntudeki bilesenlere isim verilebilirve mevcut isimler degistirilebilir; 9) kayanin doku katsayisi hesaplanabilir; 10) butun bilesenlere ait bilgilerin birkac formatta (excel, pdf) dokumu alinabilir, vb.SummaryPetrographic properties of rocks reveal their forming patterns, the formation of mineral resources and the communities in which the minerals formed. Naming of the rocks is realized by defining the shapes, forms and rates of minerals contained in the rocks under the microscope. Mineral rates are done by point counting method, but the definition of grain sizes and shapes is processed by observational. During these processes, because the segmentation, identification and counting are processed by observational errors are made. In this study, in order to eliminate the errors that occur during the processes listed above and to process the digital images of thin sections in accordance with features demanded automatically the algorithms were designed and Graphical User Interface were prepared. For this purpose, firstly, the samples were collected from the investigated land, secondly, thin sections of samples were made in the laboratory, then the digital images (digital thin sections) were obtainedwith the aid of a microscope, finally, these images are automatically processed by designed algorithms and improved software in computer.Some features of improved software are: 1) scale placement; 2) creating the minerals; 3) the borders of mineral components were determined independently; 4) zoom operations can be made; 5) grains insertion and deletion operations can be performed; 6) different colors can be appointed and replaced to each mineral component; 7) number of components can be calculated, area and axis can be measured; 8) names can be given to the components in images and existing names can be changed; 9) rock texture coefficient can be calculated; 10)information on all components can be saved in a number of formats (excel, pdf), etc.

Son yillarda bulanik mantik kavrami artik bircok teknoloji alaninda daha iyi, basit ve hizli cozu... more Son yillarda bulanik mantik kavrami artik bircok teknoloji alaninda daha iyi, basit ve hizli cozumler sunmasiyla tercih edilmeye baslanmistir. Bulanik mantigin kullanildigi alanlardan birisi de iletisim ag teknolojileridir. Ancak arastirmalar bulanik mantik iceren yontemlerin agin degisken durumlarina adapte olmada zaman zaman yetersiz kaldigini gostermistir. Bu tez ATM (Eszamansiz Aktarim Modu) aglari uzerindeki trafik kontrolu icin elverissiz hale gelen bulanik yontemlere alternatif bir yol aramak amacini gutmektedir. Yapay sinir aglarinin egitilebilme ozelligini, bulanik mantigin bilgi tabanli sozel ifade edebilme ozelligi ile birlestirerek daha yuksek performansta bir kontrol mekanizmasi tasarlanmistir. Uyarlanabilir sinirsel bulanik modellerini kullanarak oncelikle tek anahtarli (switch) aglar uzerinde kanallara iletilebilecek en uygun bit oranina erisilmistir. Daha sonra ise uyguladigimiz teknigi cok anahtarli uzerinde de uygulayarak gercege yakin bir ATM agi uzerindeki etkisi incelenmistir. Geleneksel bulanik modellere gore daha iyi sonuclar elde edilmistir.

In recent years, fuzzy logic methods has become one of the methods mostly used because of finding... more In recent years, fuzzy logic methods has become one of the methods mostly used because of finding solution fast and low cost. One of these methods, the adaptive fuzzy logic method, adapted to the relevant technology and provide the optimum solutions. In this work a network traffic controller developed for minimizing error rate of multiple switches ATM Network with adaptive neurofuzzy method. This method was compared with a conventional fuzzy method to show the advantages and better results. ABR (Available Bit Rate) service of the ATM technology is used when designing and developing the controller. Results are discussed within the framework of today’s technology requirements and problems.

In this paper, this article is about one of the applications of fuzzy logic and neuro-fuzzy in fi... more In this paper, this article is about one of the applications of fuzzy logic and neuro-fuzzy in field of control, it is how to measuring the percentage of carbon dioxide gas which is contained in sugary solution of Fizzy drink by using computer system depended on the concept of fuzzy logic and neuro-fuzzy mechanism, by these two methods the percentage of carbon dioxide can be measured in sugary solution of fizzy drink, at last the comparison can be done among result in order to know which one of methods has less ratio than another to measuring the real value of carbon dioxide of AL-TAMIM company for food stuff.

Heart disease diagnosis procedure is very vital and critical issue for the patient's health. Furt... more Heart disease diagnosis procedure is very vital and critical issue for the patient's health. Furthermore, it will help to decrease disease to a more specific level. The role of using machine learning techniques and data mining algorithms in diagnosis of heart disease is very considerable. The aim of this study was to develop a method of classifying for heart disease degree of patient based characteristic data using adaptive neuro fuzzy inference system. The data were obtained from the University of California at Irvine (UCI) machine learning repository. Seven variables are used as input of prediction model. To test the ability of the trained anfis models to recognize heart disease diagnosis, we used k-fold cross validation method. The experimental results demonstrate that the model successfully forecasts the patient's heart disease degree with an accuracy rate of 92.30%.

In this paper, traffic control of the communication systems is examined over the ATM (Asynchronou... more In this paper, traffic control of the communication systems is examined over the ATM (Asynchronous transfer mode) networks. Fuzzy mechanism and adaptive neuro fuzzy mechanism are designed to controlling flow rate on an ATM switch using a specific method of the ABR (Available Bit Rate) service of the ATM technology. Two models are compared in various cases to evaluate performance in traffic control over the network. Results are discussed within the framework of today’s technology requirements and problems.

Deu Muhendislik Fakultesi Fen ve Muhendislik

Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya g... more Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya gösterilmesi zor olan zamansal değişimleri ortaya çıkarmak ve daha önce yapılan farklı benzer çalışmaların deneysel olarak analiz etmektir. Bu çalışmada ortaya çıkarılması zor olan değişimleri vurgulamak için iki yaklaşım üzerinde çalışılmıştır. Bunlardan ilki olan Euler (piksel bazlı işleme) yaklaşımında, ilk olarak ana video büyültme metodu çağırılır. Bu metot vasıtasıyla standart video dizileri girdi olarak alınır. İkinci olarak, uzaysal ayrışma işlemi yapılır ve sonrasını takip eden işlemde de her bir kareye zamansal filtreleme uygulanır. Buna mukabil, ikinci yaklaşım olan Lagrangian (nokta izleme) yaklaşımında sırasıyla video kayıt, öznitelik noktaları takibi, yörünge kümelenmesi, yoğun optik akış interpolasyonu, katman segmentasyonu, büyütme, doku doldurma ve uyum çevirimi işlemi sırasıyla yapılarak görüntüler üzerinden veri analizinin yapılması sağlanır. Bu tarama makalesi kapsamında her iki yöntemden bahsedilerek karşılaştırılmaları gerçekleştirilmiştir.

Deu Muhendislik Fakultesi Fen ve Muhendislik, 2017

Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya g... more Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya gösterilmesi zor olan zamansal değişimleri ortaya çıkarmak ve daha önce yapılan farklı benzer çalışmaların deneysel olarak analiz etmektir. Bu çalışmada ortaya çıkarılması zor olan değişimleri vurgulamak için iki yaklaşım üzerinde çalışılmıştır. Bunlardan ilki olan Euler (piksel bazlı işleme) yaklaşımında, ilk olarak ana video büyültme metodu çağırılır. Bu metot vasıtasıyla standart video dizileri girdi olarak alınır. İkinci olarak, uzaysal ayrışma işlemi yapılır ve sonrasını takip eden işlemde de her bir kareye zamansal filtreleme uygulanır. Buna mukabil, ikinci yaklaşım olan Lagrangian (nokta izleme) yaklaşımında sırasıyla video kayıt, öznitelik noktaları takibi, yörünge kümelenmesi, yoğun optik akış interpolasyonu, katman segmentasyonu, büyütme, doku doldurma ve uyum çevirimi işlemi sırasıyla yapılarak görüntüler üzerinden veri analizinin yapılması sağlanır. Bu tarama makalesi kapsamında her iki yöntemden bahsedilerek karşılaştırılmaları gerçekleştirilmiştir.

Social network is an important part of people's daily life, which essentially can be considered a... more Social network is an important part of people's daily life, which essentially can be considered as a black box, providing great insight into the current trends of people. It has been proved that data obtained from Social Network platforms, including Facebook, Twitter, Instagram, Baidu etc., can be employed in a variety of different fields. Technically, social networks can be shown as nodes and edges, inspired from graph theory, this theory play an important role to express the structure and relations of the network. Big data mining, natural language processing (NLP), text mining and machine learning techniques are mainly utilized for processing and analysing the data. The analysis of social web can reveal the real trends of the people on any subjects, can detect terror activities before attacks, can help to understand political tendencies, cultural or global believes etc. This study is a review of social web mining that aims to examine the literature of social web mining and help to understand the techniques. However this study also proposes some novel approaches by taking into accounts data of the images on the web which will enrich the handled information while mining the social web. The study also points to the semantic web for reaching more clear inferences from social networks than at present.

Arabian Journal for Science and Engineering

Recently, subjects such as human recognition (HR), age estimation and gender recognition have bee... more Recently, subjects such as human recognition (HR), age estimation and gender recognition have been among the most investigated human–computer interaction topics, in both the academic and other fields. HR is a process in which a person is detected based on the obtained biometrical features. In this study, a new feature extraction method has been suggested through using the signals received from the sensors of the accelerometer, magnetometer and gyroscope attached to the 5 areas on the human body. The feature extraction from the signals is one of the most crucial stage. The reason behind the success of HR is based on the extracted features. However, the extraction of appropriate features for HR is a challenging issue. Various transformation methods like 1D-LBP and 1D-FbLBP have been applied to the sensor-based signals. Following the transformation process, the statistical features have been acquired from the newly developed signals. The classification processes have been carried out with the distinctive methods concerning machine learning (Knn, RF, A1DE, A2DE and ANN) by using these features. According to these results, 1D-LBP (88.4649%) and 1D-FbLBP (91.8281%) methods have been chosen to provide effective features for HR.

ArXiv, 2018

The purpose of this study is to successfully train our vehicle detector using R-CNN, Faster R-CNN... more The purpose of this study is to successfully train our vehicle detector using R-CNN, Faster R-CNN deep learning methods on a sample vehicle data sets and to optimize the success rate of the trained detector by providing efficient results for vehicle detection by testing the trained vehicle detector on the test data. The working method consists of six main stages. These are respectively; loading the data set, the design of the convolutional neural network, configuration of training options, training of the Faster R-CNN object detector and evaluation of trained detector. In addition, in the scope of the study, Faster R-CNN, R-CNN deep learning methods were mentioned and experimental analysis comparisons were made with the results obtained from vehicle detection.

Artificial Intelligence, which works on the ability to learn in machines, has a widespread field ... more Artificial Intelligence, which works on the ability to learn in machines, has a widespread field of research. One of the most researched topics of artificial intelligence is artificial neural networks. Artificial neural networks are effective today with the solution of complex problems, calculation and processing of information. Seismic method, which is one of the basic applications of geophysical field, is widely used especially for the detection of oil by using seismic waves. With the literature review, it is seen that the types of artificial neural network architectures are used. It has been determined that different methods are used in the processing of seismic data. Using the convolutional neural network (CNN), one of the artificial neural network architectures, it is aimed to achieve success in oil detection by seismic waves.

Social network is an important part of people’s daily life, which essentially can be considered a... more Social network is an important part of people’s daily life, which essentially can be considered as a black box, providing great insight into the current trends of people. It has been proved that data obtained from Social Network platforms, including Facebook, Twitter, Instagram, Baidu etc., can be employed in a variety of different fields. Technically, social networks can be shown as nodes and edges, inspired from graph theory, this theory play an important role to express the structure and relations of the network. Big data mining, natural language processing (NLP), text mining and machine learning techniques are mainly utilized for processing and analysing the data. The analysis of social web can reveal the real trends of the people on any subjects, can detect terror activities before attacks, can help to understand political tendencies, cultural or global believes etc. This study is a review of social web mining that aims to examine the literature of social web mining and help to...

Kayalarin petrografik ozellikleri; olusum sekillerini, olusum kaynaklarini ve hangi mineral toplu... more Kayalarin petrografik ozellikleri; olusum sekillerini, olusum kaynaklarini ve hangi mineral topluluklarindan olustugunu ortaya koyar. Kayalarin icermis olduklari minerallerin durus sekilleri ve bicimleri mikroskop altinda gozlemsel olarak tanimlanarak kayalarin adlandirilmalari yapilmaktadir. Mineral oranlari nokta sayma yontemi ile tane boyutlari ve sekillerinin tanimlanmasi ise gozlemsel olarak yapilmaktadir. Bu islemler sirasinda ayrimlama, tanimlama ve sayma gozlemsel olarak yapildigi icin hatalar yapilmaktadir. Bu calismada; yukarida siralanan islemleri yapma sirasinda meydana gelen hatalari ortadan kaldirmak ve incekesit dijital goruntulerini talep edilen ozelliklere uygun olarak otomatik islemek icin algoritmalar tasarlanmis ve Grafik Kullanici Arayuzlu program paketi hazirlanmistir. Bu amacla, ilk asamada arastirilan araziden ornekler toplanmis, laboratuar ortaminda orneklerden ince kesitler yapilmis, sonra mikroskop yardimi ile dijital goruntuler (dijital incekesitler) elde edilmis ve son olarak bu goruntuler tasarlanan algoritmalar ve gelistirilen yazilim ile bilgisayar ortaminda otomatik olarak islenmistir. Gelistirilen yazilimin ozelliklerinden bazilari sunlardir: 1) olcek yerlestirilebilir; 2) mineraller olusturulabilir; 3) mineral bilesenlerinin sinirlari bagimsiz olarak belirlenebilir; 4) zoom islemleri yapilabilir; 5) tanecik ekleme ve silme islemleri yapilabilir; 6) her bir mineral bilesenine farkli renk atana ve degistirilebilir; 7) bilesenlerin sayisi, alan ve eksen olcumleri yapilabilir; 8) goruntudeki bilesenlere isim verilebilirve mevcut isimler degistirilebilir; 9) kayanin doku katsayisi hesaplanabilir; 10) butun bilesenlere ait bilgilerin birkac formatta (excel, pdf) dokumu alinabilir, vb.SummaryPetrographic properties of rocks reveal their forming patterns, the formation of mineral resources and the communities in which the minerals formed. Naming of the rocks is realized by defining the shapes, forms and rates of minerals contained in the rocks under the microscope. Mineral rates are done by point counting method, but the definition of grain sizes and shapes is processed by observational. During these processes, because the segmentation, identification and counting are processed by observational errors are made. In this study, in order to eliminate the errors that occur during the processes listed above and to process the digital images of thin sections in accordance with features demanded automatically the algorithms were designed and Graphical User Interface were prepared. For this purpose, firstly, the samples were collected from the investigated land, secondly, thin sections of samples were made in the laboratory, then the digital images (digital thin sections) were obtainedwith the aid of a microscope, finally, these images are automatically processed by designed algorithms and improved software in computer.Some features of improved software are: 1) scale placement; 2) creating the minerals; 3) the borders of mineral components were determined independently; 4) zoom operations can be made; 5) grains insertion and deletion operations can be performed; 6) different colors can be appointed and replaced to each mineral component; 7) number of components can be calculated, area and axis can be measured; 8) names can be given to the components in images and existing names can be changed; 9) rock texture coefficient can be calculated; 10)information on all components can be saved in a number of formats (excel, pdf), etc.

Son yillarda bulanik mantik kavrami artik bircok teknoloji alaninda daha iyi, basit ve hizli cozu... more Son yillarda bulanik mantik kavrami artik bircok teknoloji alaninda daha iyi, basit ve hizli cozumler sunmasiyla tercih edilmeye baslanmistir. Bulanik mantigin kullanildigi alanlardan birisi de iletisim ag teknolojileridir. Ancak arastirmalar bulanik mantik iceren yontemlerin agin degisken durumlarina adapte olmada zaman zaman yetersiz kaldigini gostermistir. Bu tez ATM (Eszamansiz Aktarim Modu) aglari uzerindeki trafik kontrolu icin elverissiz hale gelen bulanik yontemlere alternatif bir yol aramak amacini gutmektedir. Yapay sinir aglarinin egitilebilme ozelligini, bulanik mantigin bilgi tabanli sozel ifade edebilme ozelligi ile birlestirerek daha yuksek performansta bir kontrol mekanizmasi tasarlanmistir. Uyarlanabilir sinirsel bulanik modellerini kullanarak oncelikle tek anahtarli (switch) aglar uzerinde kanallara iletilebilecek en uygun bit oranina erisilmistir. Daha sonra ise uyguladigimiz teknigi cok anahtarli uzerinde de uygulayarak gercege yakin bir ATM agi uzerindeki etkisi incelenmistir. Geleneksel bulanik modellere gore daha iyi sonuclar elde edilmistir.

In recent years, fuzzy logic methods has become one of the methods mostly used because of finding... more In recent years, fuzzy logic methods has become one of the methods mostly used because of finding solution fast and low cost. One of these methods, the adaptive fuzzy logic method, adapted to the relevant technology and provide the optimum solutions. In this work a network traffic controller developed for minimizing error rate of multiple switches ATM Network with adaptive neurofuzzy method. This method was compared with a conventional fuzzy method to show the advantages and better results. ABR (Available Bit Rate) service of the ATM technology is used when designing and developing the controller. Results are discussed within the framework of today’s technology requirements and problems.

In this paper, this article is about one of the applications of fuzzy logic and neuro-fuzzy in fi... more In this paper, this article is about one of the applications of fuzzy logic and neuro-fuzzy in field of control, it is how to measuring the percentage of carbon dioxide gas which is contained in sugary solution of Fizzy drink by using computer system depended on the concept of fuzzy logic and neuro-fuzzy mechanism, by these two methods the percentage of carbon dioxide can be measured in sugary solution of fizzy drink, at last the comparison can be done among result in order to know which one of methods has less ratio than another to measuring the real value of carbon dioxide of AL-TAMIM company for food stuff.

Heart disease diagnosis procedure is very vital and critical issue for the patient's health. Furt... more Heart disease diagnosis procedure is very vital and critical issue for the patient's health. Furthermore, it will help to decrease disease to a more specific level. The role of using machine learning techniques and data mining algorithms in diagnosis of heart disease is very considerable. The aim of this study was to develop a method of classifying for heart disease degree of patient based characteristic data using adaptive neuro fuzzy inference system. The data were obtained from the University of California at Irvine (UCI) machine learning repository. Seven variables are used as input of prediction model. To test the ability of the trained anfis models to recognize heart disease diagnosis, we used k-fold cross validation method. The experimental results demonstrate that the model successfully forecasts the patient's heart disease degree with an accuracy rate of 92.30%.

In this paper, traffic control of the communication systems is examined over the ATM (Asynchronou... more In this paper, traffic control of the communication systems is examined over the ATM (Asynchronous transfer mode) networks. Fuzzy mechanism and adaptive neuro fuzzy mechanism are designed to controlling flow rate on an ATM switch using a specific method of the ABR (Available Bit Rate) service of the ATM technology. Two models are compared in various cases to evaluate performance in traffic control over the network. Results are discussed within the framework of today’s technology requirements and problems.

Deu Muhendislik Fakultesi Fen ve Muhendislik

Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya g... more Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya gösterilmesi zor olan zamansal değişimleri ortaya çıkarmak ve daha önce yapılan farklı benzer çalışmaların deneysel olarak analiz etmektir. Bu çalışmada ortaya çıkarılması zor olan değişimleri vurgulamak için iki yaklaşım üzerinde çalışılmıştır. Bunlardan ilki olan Euler (piksel bazlı işleme) yaklaşımında, ilk olarak ana video büyültme metodu çağırılır. Bu metot vasıtasıyla standart video dizileri girdi olarak alınır. İkinci olarak, uzaysal ayrışma işlemi yapılır ve sonrasını takip eden işlemde de her bir kareye zamansal filtreleme uygulanır. Buna mukabil, ikinci yaklaşım olan Lagrangian (nokta izleme) yaklaşımında sırasıyla video kayıt, öznitelik noktaları takibi, yörünge kümelenmesi, yoğun optik akış interpolasyonu, katman segmentasyonu, büyütme, doku doldurma ve uyum çevirimi işlemi sırasıyla yapılarak görüntüler üzerinden veri analizinin yapılması sağlanır. Bu tarama makalesi kapsamında her iki yöntemden bahsedilerek karşılaştırılmaları gerçekleştirilmiştir.

Deu Muhendislik Fakultesi Fen ve Muhendislik, 2017

Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya g... more Bu çalışmanın amacı, videolar içerisinde bulunan çıplak gözle görülmesi, açığa çıkarılması veya gösterilmesi zor olan zamansal değişimleri ortaya çıkarmak ve daha önce yapılan farklı benzer çalışmaların deneysel olarak analiz etmektir. Bu çalışmada ortaya çıkarılması zor olan değişimleri vurgulamak için iki yaklaşım üzerinde çalışılmıştır. Bunlardan ilki olan Euler (piksel bazlı işleme) yaklaşımında, ilk olarak ana video büyültme metodu çağırılır. Bu metot vasıtasıyla standart video dizileri girdi olarak alınır. İkinci olarak, uzaysal ayrışma işlemi yapılır ve sonrasını takip eden işlemde de her bir kareye zamansal filtreleme uygulanır. Buna mukabil, ikinci yaklaşım olan Lagrangian (nokta izleme) yaklaşımında sırasıyla video kayıt, öznitelik noktaları takibi, yörünge kümelenmesi, yoğun optik akış interpolasyonu, katman segmentasyonu, büyütme, doku doldurma ve uyum çevirimi işlemi sırasıyla yapılarak görüntüler üzerinden veri analizinin yapılması sağlanır. Bu tarama makalesi kapsamında her iki yöntemden bahsedilerek karşılaştırılmaları gerçekleştirilmiştir.

Social network is an important part of people's daily life, which essentially can be considered a... more Social network is an important part of people's daily life, which essentially can be considered as a black box, providing great insight into the current trends of people. It has been proved that data obtained from Social Network platforms, including Facebook, Twitter, Instagram, Baidu etc., can be employed in a variety of different fields. Technically, social networks can be shown as nodes and edges, inspired from graph theory, this theory play an important role to express the structure and relations of the network. Big data mining, natural language processing (NLP), text mining and machine learning techniques are mainly utilized for processing and analysing the data. The analysis of social web can reveal the real trends of the people on any subjects, can detect terror activities before attacks, can help to understand political tendencies, cultural or global believes etc. This study is a review of social web mining that aims to examine the literature of social web mining and help to understand the techniques. However this study also proposes some novel approaches by taking into accounts data of the images on the web which will enrich the handled information while mining the social web. The study also points to the semantic web for reaching more clear inferences from social networks than at present.