Yazar "Orak, Ilhami Muharrem" seçeneğine göre listele
Listeleniyor 1 - 10 / 10
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe An Attempt at Automatic Label Generation for Object Entry and Exit on Multimedia with a Semantic Search(Assoc Information Communication Technology Education & Science, 2018) Menemencioglu, Oguzhan; Orak, Ilhami MuharremFilling the semantic gap between low level keywords which are retrieved automatically from multimedia data and human interpretations of data becomes critical. The research aims to handle the issue by using semantic search on multimedia data. A model is proposed with this research which detects enter/exit points and performs automatic label generation instead of hand using labelling. The model is implemented on a popular and commonly used benchmarking dataset. After the reliability of the model is proofed, it is implemented on a test dataset. It is indicated that the multiple interpretation of results can improve the retrieved information and make the model usability possible.Öğe Detection of directional eye movements based on the electrooculogram signals through an artificial neural network(Pergamon-Elsevier Science Ltd, 2015) Erkaymaz, Hande; Ozer, Mahmut; Orak, Ilhami MuharremThe electrooculogram signals are very important at extracting information about detection of directional eye movements. Therefore, in this study, we propose a new intelligent detection model involving an artificial neural network for the eye movements based on the electrooculogram signals. In addition to conventional eye movements, our model also involves the detection of tic and blinking of an eye. We extract only two features from the electrooculogram signals, and use them as inputs for a feed-forwarded artificial neural network. We develop a new approach to compute these two features, which we call it as a movement range. The results suggest that the proposed model have a potential to become a new tool to determine the directional eye movements accurately. (C) 2015 Elsevier Ltd. All rights reserved.Öğe Examination of Speed Contribution of Parallelization for Several Fingerprint Pre-Processing Algorithms(Univ Suceava, Fac Electrical Eng, 2014) Gorgunoglu, Salih; Orak, Ilhami Muharrem; Cavusoglu, Abdullah; Gok, MehmetIn analysis of minutiae based fingerprint systems, fingerprints needs to be pre-processed. The pre-processing is carried out to enhance the quality of the fingerprint and to obtain more accurate minutiae points. Reducing the pre-processing time is important for identification and verification in real time systems and especially for databases holding large fingerprints information. Parallel processing and parallel CPU computing can be considered as distribution of processes over multi core processor. This is done by using parallel programming techniques. Reducing the execution time is the main objective in parallel processing. In this study, pre-processing of minutiae based fingerprint system is implemented by parallel processing on multi core computers using OpenMP and on graphics processor using CUDA to improve execution time. The execution times and speedup ratios are compared with the one that of single core processor. The results show that by using parallel processing, execution time is substantially improved. The improvement ratios obtained for different pre-processing algorithms allowed us to make suggestions on the more suitable approaches for parallelization.Öğe Fast and simple computer aided internal electrical wiring project design & integration of calculations(Elsevier Science Bv, 2012) Orak, Ilhami Muharrem; Akgul, BayramFor drawing electrical wiring diagram of buildings, mostly two dimensional CAD programs are used. Drawing is done starting from the source (utility pole) to main power panel and then it is distributed towards to subpanels. Then, end points such as sockets, lighting armature, etc are defined. In this study several algorithms are developed in order to evaluate different part of drawing project. Tree structure of drawing is analyzed and all materials to be used are decided based on an algorithm. Apart from these, some of the supplies to be used in project but not shown in the drawings are also be chosen with certain criteria. These decision systems will result in efficiency of overall cost of the project. In wire drawing one of the most important point is the line drawing with reference to an object. Apart from light outlet all wires are referenced to walls. Furthermore, all supplies shown in drawings are allocated with reference to walls and with an angle to walls. In wiring diagram drawings, standard CAD programs have no ability for wire drawing with reference to walls. Besides that, suggestion and selection of the most appropriate supplies is not possible. These increase drawing time and result in very high cost calculation errors. These algorithms provide fast and correct implementation. By analyzing drawings in a tree structure on which users have minimal impact, and with automatic selection of supplies all type of calculation mistakes are minimized. (C) 2011 Published by Elsevier Ltd.Öğe A novel method based on deep learning algorithms for material deformation rate detection(Springer, 2024) Ozdem, Selim; Orak, Ilhami MuharremGiven the significant influence of microstructural characteristics on a material's mechanical, physical, and chemical properties, this study posits that the deformation rate of structural steel S235-JR can be precisely determined by analyzing changes in its microstructure. Utilizing advanced artificial intelligence techniques, microstructure images of S235-JR were systematically analyzed to establish a correlation with the material's lifespan. The steel was categorized into five classes and subjected to varying deformation rates through laboratory tensile tests. Post-deformation, the specimens underwent metallographic procedures to obtain microstructure images via an light optical microscope (LOM). A dataset comprising 10000 images was introduced and validated using K-Fold cross-validation. This research utilized deep learning (DL) architectures ResNet50, ResNet101, ResNet152, VGG16, and VGG19 through transfer learning to train and classify images containing deformation information. The effectiveness of these models was meticulously compared using a suite of metrics including Accuracy, F1-score, Recall, and Precision to determine their classification success. The classification accuracy was compared across the test data, with ResNet50 achieving the highest accuracy of 98.45%. This study contributes a five-class dataset of labeled images to the literature, offering a new resource for future research in material science and engineering.Öğe A parallel algorithm for defect detection of rail and profile in the manufacturing(Gazi Univ, Fac Engineering Architecture, 2017) Orak, Ilhami Muharrem; Celik, AhmetObtaining a result by processing an image via an automatic system may be useful in many fields today. Manufacturing a defective product is an undesired case for manufacturers in many fields. Processing images is an efficient method used to detect defects on images to eliminate the defective products. Since image processing is conducted on pixel basis, it entails great workload. In cases where speed is important in processing, parallel image processing might be a solution. Therefore, processing images in the current multi-core computers by paralleling them with additional hardware and software can boost the performance. The performance in parallel image processing is related to relevance of the algorithm to the parallelism and its accurate distribution to the processors. Common use of the resources and excess of data exchange affect the performance directly. In this study, parallel application of COLMSTD algorithm developed to detect the defects on rail and profile surface during rolling in Kardemir Inc. rolling plant was conducted in two different ways. The 1st method was carried out by selecting the CUDA core numbers in GPU structure by software and the 2nd method was conducted by using single CUDA core. The performance of the results obtained on GPU (Graphics Processing Unit) with the support of CUDA (Compute Unified Device Architecture) interface was compared with that of CPU values.Öğe A Review on Semantic Text and Multimedia Retrieval and Recent Trends(Igi Global, 2015) Menemencioglu, Oguzhan; Orak, Ilhami MuharremSemantic web works on producing machine readable data and aims to deal with large amount of data. The most important tool to access the data which exist in web is the search engine. Traditional search engines are insufficient in the face of the amount of data that consists in the existing web pages. Semantic search engines are extensions to traditional engines and overcome the difficulties faced by them. This paper summarizes semantic web, concept of traditional and semantic search engines and infrastructure. Also semantic search approaches are detailed. A summary of the literature is provided by touching on the trends. In this respect, type of applications and the areas worked for are considered. Based on the data for two different years, trend on these points are analyzed and impacts of changes are discussed. It shows that evaluation on the semantic web continues and new applications and areas are also emerging. Multimedia retrieval is a newly scope of semantic. Hence, multimedia retrieval approaches are discussed. Text and multimedia retrieval is analyzed within semantic search.Öğe Semantic Querying on Multimedia Data(Ieee, 2016) Menemencioglu, Oguzhan; Orak, Ilhami MuharremEffective information retrieval is not provided with manual multimedia processing methods against huge amount of multimedia data which is produced in nowadays. Even automatic annotation generation methods are hard to address, they only operate on image processing step. Ultimately, it does not mean having a system that retrieve, relate and interpret the retrieved data. This study is done for filling the gap that is mentioned. In the study, inquiry is done by use of semantic web infrastructure on the produced, processed data of multimedia which is a source of ontologies. It is aimed to build a framework to introduce meaning on result data with semantic search. In this manner, it will be possible to access the knowledge effectively on multimedia datasets.Öğe A Simple Solution to Prevent Parameter Tampering in Web Applications(Igi Global, 2017) Menemencioglu, Oguzhan; Orak, Ilhami MuharremBusiness over the internet such as banking and several online services are growing rapidly. Similarly, social media web portals are also getting more and more involved in our daily life. Since these applications are popular and consist of personal and valuable data, they attract malicious attacks to their vulnerable points. The weakness can also be faced in all businesses and institutions that do not care the necessary security steps. The web parameter tampering is one of the major attacks which is based on the modification of parameters. In order to prevent the parameter tampering, a novel and simple mechanism is implemented by verifying the validity. The mechanism is based on a deterministic finite state machine. Beside this static method, the system also has run time validation which leads for the usage of hybrid analysis approach. As an evaluation, performance assessment of the algorithm is done for real time attacks targeting a web site.Öğe Üretim aşamasında ray ve profilde oluşan kusurların tespitine yönelik bir paralel kusur algılama algoritması(2017) Orak, Ilhami Muharrem; Çelik, AhmetOtomatik bir sistem tarafından görüntü işlenerek sonuç elde etmek günümüzde pek çok alanda gerekliolabilmektedir. Kusurlu ürün üretimi birçok alanda karşılaşılan üreticiler tarafından da istenmeyen birdurumdur. Görüntü işleyerek görüntü üzerindeki kusurların tespit edilmesi bu alanda kullanılan biryöntemdir. Görüntü işleme piksel temelli yapıldığından dolayı çok iş yükü oluşturmaktadır. Hızın işlemsürecinde önem arz ettiği durumlarda paralel görüntü işlemenin yapılması bir çözüm olabilmektedir. Bundandolayı mevcut çok çekirdekli bilgisayarların donanım ya da yazılım yardımıyla paralelleştirilerekgörüntülerin işlenmesi performans sağlayacaktır. Paralel görüntü işlemede elde edilen performans, kullanılanalgoritmanın paralelliğe uygun olması ve işlemcilere doğru bir şekilde dağıtım yapılması ile ilişkilidir.Kaynakların ortak kullanımı ve veri alışverişinin fazla olması performansı doğrudan etkiler. Bu çalışmada;Kardemir A.Ş. haddehanede, haddeleme işlemi sırasında ray ve profil yüzeylerinde meydana gelen kusurların tespit edilmesine yönelik geliştirilen COLMSTD algoritmasının paralel uygulaması, iki farklışekilde gerçekleştirilmiştir. 1. Yöntemde GPU'da çalışacak CUDA çekirdek sayıları yazılımsal olarakdeğiştirilmiş ve 2. Yöntemde ise tek CUDA çekirdeğindeki blok sayıları değiştirilerek, testgerçekleştirilmiştir. GPU (Graphics Processing Unit) üzerinde CUDA (Compute Unified DeviceArchitecture) arayüz desteğinin uygulanması ile elde edilen değerler ile CPU üzerinde elde edilen değerlerkıyaslanmıştır.
