Yazar "Kutlu, Yakup" seçeneğine göre listele
Listeleniyor 1 - 15 / 15
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Classification of Fish Species With Two Dorsal Fins Using Centroid-Contour Distance(Ieee, 2015) Iscimen, Bilal; Kutlu, Yakup; Uyan, Ali; Turan, CemalColor, texture and shape are generally used features in order to recognise an object from an image. In this study centroid-contour distance method is used in order to classify fish species with two dorsal fins. Therefore, fish images with two dorsal fins were used from fish images database taken under specific conditions. Various image processing methods were applied on images in order to extract centroid-contour distances. These distances were used as features and Nearest Neighbour algorithm was used for classification. 15 species from 427 fish images were classified with 95% general accuracy achievement.Öğe Diagnosis of congestive heart failure using Poincare map plot(2012) Yayik, Apdullah; Kutlu, YakupIn this study, in order to diagnose congestive heart failure patients (CHF), Poincare map obtained from raw ECG data is used. CHF and normal ECG data, which are distributed freely via internet, are analyzed. Poincare map is divided into equal rectangle cells and points in all of the cells are determined. These values are used for knn (k-nearest neighbour) classification. At the result of this study, it is considered that CHF patients and normal people can be separated each other using features obtained from Poincare map of Raw ECG record. © 2012 IEEE.Öğe ECG Based Human Identification Using Logspace Grid Analysis of Second Order Difference Plot(Ieee, 2015) Altan, Gokhan; Kutlu, YakupIn this study, Second Order Difference Plot (SODP) features are used for ECG based human identification. SODP is a method that allows to determine the features with the statistical analysis of the situations obtained from distributions and the distribution of each of successive points on an unstable and linear signals. ECG records of 90 individuals in Physionet ECG-Id database are used in the study. These records are divided into segments with logarithmic grid, number of points in each segment was examined. Extracted features are classified using 2 Fold Cross Validation with kNN classifier, speed and performance of identification system were investigated. As a result, ECG based Human identification using Logspace Grid Analysis of SODP was performed in a very short time with 91.52% success.Öğe ECG based Human identification using Logspace Grid Analysis of Second Order Difference Plot(Institute of Electrical and Electronics Engineers Inc., 2015) Altan, Gokhan; Kutlu, YakupIn this study, Second Order Difference Plot (SODP) features are used for ECG based human identification. SODP is a method that allows to determine the features with the statistical analysis of the situations obtained from distributions and the distribution of each of successive points on an unstable and linear signals. ECG records of 90 individuals in Physionet ECG-Id database are used in the study. These records are divided into segments with logarithmic grid, number of points in each segment was examined. Extracted features are classified using 2 Fold Cross Validation with kNN classifier, speed and performance of identification system were investigated. As a result, ECG based Human identification using Logspace Grid Analysis of SODP was performed in a very short time with 91.52% success. © 2015 IEEE.Öğe Epileptic state detection : Pre-ictal, inter-ictal, ictal(2015) Yayık, Apdullah; Yıldırım, Esen; Kutlu, Yakup; Yıldırım, SerdarEpileptic seizure detection and prediction from electroencephalography (EEG) is a vital area of research. In this study, SecondOrder Difference Plot (SODP) is used to extract features based on consecutive difference of time domain values from three states of EEG (pre-ictal, ictal and inter-ictal), and Multi-Layer Neural Network classifier is used to classify these three classes. The proposed technique is tested on a publicly available EEG database and classified with Naive Bayes and k-nearest neighbor classifiers. As a result, it is shown that overall accuracy of 98.70% can be achieved by using the proposed system with Neural Network classifier.Öğe Feature extraction for ECG heartbeats using higher order statistics of WPD coefficients(Elsevier Ireland Ltd, 2012) Kutlu, Yakup; Kuntalp, DamlaThis paper describes feature extraction methods using higher order statistics (HOS) of wavelet packet decomposition (WPD) coefficients for the purpose of automatic heartbeat recognition. The method consists of three stages. First, the wavelet package coefficients (WPC) are calculated for each different type of ECG beat. Then, higher order statistics of WPC are derived. Finally, the obtained feature set is used as input to a classifier, which is based on k-NN algorithm. The MIT-BIH arrhythmia database is used to obtain the ECG records used in this study. All heartbeats in the arrhythmia database are grouped into five main heartbeat classes. The classification accuracy of the proposed system is measured by average sensitivity of 90%, average selectivity of 92% and average specificity of 98%. The results show that HOS of WPC as features are highly discriminative for the classification of different arrhythmic ECG beats. (C) 2011 Elsevier Ireland Ltd. All rights reserved.Öğe Frekans seçici yüzeyler, teknikleri ve uygulamaları(Hatay Mustafa Kemal Üniversitesi, 2004) Kutlu, Yakup; Ünal, EminBu çalışmada, belirli bir rezonans frekansına sahip periyodik kare halka ve dairesel halka Frekans Seçici Yüzeylerin, farklı dielektrik sabiti ve geliş açılarda iletim yansıma katsayıları hesaplanıp, kalkanlama etkinlikleri incelenmiş ve birbirleriyle karşılaştırılması amaçlanmıştır. Periyodik yarıklar içeren iletken levhalar veya periyodik yamalar içeren dielektrik yapılara frekans seçici yüzeyler (FSY) denir. Bu çalışmada, dielektrik katman üzerine kare halka ve dairesel halka geometrili iletken yapılar periyodik olarak yerleştirilmiştir. Bu yapıların periyodikliğinden dolayı elektrik ve manyetik alanlar, floquet modları cinsinden ifade edilmiştir. Sınır şartları, indüksiyon teoremi ve moment metodu kullanılarak, gelen düzlem dalganın bir iletken eleman üzerinde oluşturduğu bilinmeyen akım yoğunluğu, bulunmuştur. Bilinmeyen akım katsayıları elde edildikten sonra iletilen ve yansıyan alanlar bulunmuştur. İletilen alan ile gelen alan arasındaki ilişkiyle kalkanlama etkinliği hesaplanmıştır. Kare halka ve dairesel halka frekans seçici yüzeylerin, geliş açısı ve dielektrik sabitinin farklı değerleri için kalkanlama etkinliği incelenip birbirleriyle karşılaştırılmıştır. Elde edilen veriler sonucunda; her iki FSY yapının da kalkanlama etkinliği ve açısal kararlılıkları iyi denebilecek düzeyde ve dairesel halkanın kare halkaya oranla daha kararlı olduğu görülmüştür. 2004, 85 sayfaÖğe Image analysis methods on fish recognition(Ieee, 2014) Iscimen, Bilal; Kutlu, Yakup; Reyhaniye, Asil Nadir; Turan, CemalThe aim of study is creating a new database which contains fish species and classifing this fish species. A new fish database was created by using the fish photos in seas of Turkey. The new feature set are obtained by marking the biometric points on fish to identify family and species of fishes. The features were obtained by using the three different biometric measurement techniques (Euclidean network technique, quadratic network technique, triangulation technique). A classificatios system was created by using Naive Bayesian classifier. The obtained accuracy is 93.10% for 7 families on family classification and 75.71% for 15 species on species classification.Öğe Improving Pseudo Random Number Generator Using Artificial Neural Networks(Ieee, 2013) Yayik, Apdullah; Kutlu, YakupPseudo-random number generators generate sequent of digits that cannot be expected before. Random number generators are used in lots of studies especially physical and statical implementations. In this paper; by using Multi-Layer Perceptron Neural Network, a traditional random number generator is strengthened. In the end of the study; both of random number generators are tested by some randomness tests of National Institute of Standard Technology test suite. As a result, it is learned that Neural Networks can generate good random numbers.Öğe Konjestif kalp yetmezliğinin Hilbert-Huang dönüşüm ile analizi(2014) Altan, Gökhan; Yayık, Abdullah; Kutlu, Yakup; Yıldırım, Serdar; Yıldırım, EsenHilbert-Huang Dönüşümü (HHD) liner olmayan ve sabit olmayan sinyaller üzerinde öznitelik çıkartma, filtreleme gibi işlemlerde sıkça kullanılan bir yöntemdir. Bu çalışmada, HHD yönteminin kalp sinyallerine uygulanması sonucu özniteliklerin belirlenmesi ve belirlenen bu özniteliklerin Kongestif Kalp Yetmezliği (KKY) olan hastaların kontrol grubundan ayırt edilerek sınıflama yapılması üzerine bir çalışma yapılmıştır. Kalp hızı değişkenlerinden elde edilen RR sinyalleri, HHD işleminden geçirilerek içsel mod fonksiyonları (İMF) bileşenleri elde edilmiş, dönüşüm sonrası elde edilen sinyallerin istatistiksel bilgileri öznitelik olarak belirlenmiştir. Elde edilen öznitelikler, Yapay Sinir Ağları (YSA) kullanılarak sınıflandırma başarımı incelenmiştir. Sonuç olarak, RR sinyallerden elde edilen İMF bileşenlerin istatistiksel öznitelikleri kullanılarak sınıflama işleminde iyi sonuçlar alınabileceğini göstermiştir.Öğe MULTI-STAGE FISH CLASSIFICATION SYSTEM USING MORPHOMETRY(Parlar Scientific Publications (P S P), 2017) Kutlu, Yakup; Iscimen, Bilal; Turan, CemalThe aim of this study is to create a multi-stage fish classification system with high accuracy rate. Classifications are based on biometric points of the fishes that consists of three main phases, data acquisition, feature extraction and classification. In the first phase, fish image database was collected, then features were extracted using morphometry and classified with three stage classifier model. Nearest Neighbor algorithm was used as classifier, and 25 fish species were classified with accuracy of about 99%.Öğe NEURAL NETWORK BASED CRYPTOGRAPHY(Acad Sciences Czech Republic, Inst Computer Science, 2014) Yayik, Apdullah; Kutlu, YakupIn this paper, neural network based cryptology is performed. The system consists of two stages. In the first stage, neural network-based pseudo-random numbers (NPRNGs) are generated and the results are tested for randomness using National Institute of Standard Technology (NIST) randomness tests. In the second stage, a neural network-based cryptosystem is designed using NPRNGs. In this cryptosystem, data, which is encrypted by non-linear techniques, is subject to decryption attempts by means of two identical artificial neural networks (ANNs). With the first neural network, non-linear encryption is modeled using relation-building functionality. The encrypted data is decrypted with the second neural network using decision-making functionality.Öğe A new approach to early diagnosis of congestive heart failure disease by using Hilbert-Huang transform(Elsevier Ireland Ltd, 2016) Altan, Gokhan; Kutlu, Yakup; Allahverdi, NovruzCongestive heart failure (CHF) is a degree of cardiac disease occurring as a result of the heart's inability to pump enough blood for the human body. In recent studies, coronary artery disease (CAD) is accepted as the most important cause of CHF. This study focuses on the diagnosis of both the CHF and the CAD. The Hilbert-Huang transform (HHT), which is effective on nonlinear and non-stationary signals, is used to extract the features from R-R intervals obtained from the raw electrocardiogram data. The statistical features are extracted from instinct mode functions that are obtained applying the HHT to R-R intervals. Classification performance is examined with extracted statistical features using a multilayer perceptron neural network. The designed model classified the CHF, the CAD patients and a normal control group with rates of 97.83%, 93.79% and 100%, accuracy, specificity and sensitivity, respectively. Also, early diagnosis of the CHF was performed by interpretation of the CAD with a classification accuracy rate of 97.53%, specificity of 98.18% and sensitivity of 97.13%. As a result, a single system having the ability of both diagnosis and early diagnosis of CHF is performed by integrating the CAD diagnosis method to the CHF diagnosis method. (C) 2016 Elsevier Ireland Ltd. All rights reserved.Öğe A Review on Respiratory Sound Analysis using Machine Learning(Ieee, 2016) Altan, Gokhan; Kutlu, YakupAuscultation of the respiratory sounds is an inexpensive and effective method for diagnosing cardiopulmonary disorders using lung sounds from chest and back. Nowadays, high system performances in the management of robust processes that require great attention were increased using the computer-aided analysis methods and the developments of the diagnosis system. Analysis of the respiratory sounds with computer-aided systems allows objective and useful assessments. In this study, a brief description of the abnormal respiratory sounds was presented. The main aims of the study are performing a systematic review about methods and the machine learning algorithms that are used to classify the abnormal respiratory sounds for diagnosis of cardio-pulmonary disorders and evaluating the development of possible methods on respiratory sounds in the future.Öğe Topographic Analysis and Diagnosis of Congestive Heart Failure Using Second-Order Difference Plot(Ieee, 2014) Yayik, Apdullah; Kutlu, YakupIn this study, in order to diagnose congestive heart failure (CHI) patients, second-order difference map (SDOP) features obtained from raw electrocardiogram (EEG) data is used CHF and normal patients' electrocardiogram data which are distributed freely via internet, is analyzed By U-matrix presentation of Self-Organized Maps, distances between groups and clusters are examined 1x16 Axed SODP features of raw EEG data obtained from normal and CHF patients can successfully be classified using U-matrix presentation of SOM. As a classifier Neural Network based classifier is applied and %99.9 over all accuracy performance is reached
