Murat Uyar

1D-local binary pattern based feature extraction for classification of epileptic EEG signals

In this paper, an effective approach for the feature extraction of raw Electroencephalogram (EEG) signals by means of one-dimensional local binary pattern (1D-LBP) was presented. For the importance of making the right decision, the proposed method was performed to be able to get better features of the EEG signals. The proposed method was consisted of two stages: feature extraction by 1D-LBP and classification by classifier algorithms with features extracted. On the classification stage, the several machine learning methods were employed to uniform and non-uniform 1D-LBP features. The proposed method was also compared with other existing techniques in the literature to find out benchmark for an epileptic data set. The implementation results showed that the proposed technique could acquire high accuracy in classification of epileptic EEG signals. Also, the present paper is an attempt to develop a general-purpose feature extraction scheme, which can be utilized to extract features from different categories of EEG signals.

A hybrid decision support system based on rough set and extreme learning machine for diagnosis of hepatitis disease

Abstract Hepatitis is a disease which is seen at all levels of age. Hepatitis disease solely does not have a lethal effect, but the early diagnosis and treatment of hepatitis is crucial as it triggers other diseases. In this study, a new hybrid medical decision support system based on rough set (RS) and extreme learning machine (ELM) has been proposed for the diagnosis of hepatitis disease. RS-ELM consists of two stages. In the first one, redundant features have been removed from the data set through RS approach. In the second one, classification process has been implemented through ELM by using remaining features. Hepatitis data set, taken from UCI machine learning repository has been used to test the proposed hybrid model. A major part of the data set (48.3%) includes missing values. As removal of missing values from the data set leads to data loss, feature selection has been done in the first stage without deleting missing values. In the second stage, the classification process has been performed through ELM after the removal of missing values from sub-featured data sets that were reduced in different dimensions. The results showed that the highest 100.00% classification accuracy has been achieved through RS-ELM and it has been observed that RS-ELM model has been considerably successful compared to the other methods in the literature. Furthermore in this study, the most significant features have been determined for the diagnosis of the hepatitis. It is considered that proposed method is to be useful in similar medical applications.

Automatic identification of butterfly species based on local binary patterns and artificial neural network

A computer vision method was proposed for automatically identifying butterfly species.To our knowledge, it was the first study in identifying the butterfly species with computer vision.The method is based on local binary patterns and artificial neural network.Results demonstrated that the proposed method has achieved well recognition accuracy rates. Butterflies are classified firstly according to their outer morphological qualities. It is required to analyze genital characters of them when classification according to outer morphological qualities is not possible. Genital characteristics of a butterfly can be determined by using various chemical substances and methods. Currently, these processes are carried out manually by preparing genital slides of the collected butterfly through some certain processes. For some groups of butterflies molecular techniques should be applied for identification which is expensive to use. In this study, a computer vision method is proposed for automatically identifying butterfly species as an alternative to conventional identification methods. The method is based on local binary pattern (LBP) and artificial neural network (ANN). A total of 50 butterfly images of five species were used for evaluating the effectiveness of the proposed method. Experimental results demonstrated that the proposed method has achieved well recognition in terms of accuracy rates for butterfly species identification.

Classification of power quality disturbances based on S-transform and image processing techniques

This paper presents a method that combines discrete S-transform (DST) time-frequency distribution (TFD) and local binary pattern (LBP) based image analysis technique for classifying power quality (PQ) disturbances. The purpose of this combination is to extract discriminative features by utilizing from both capability of generating the compact TFD of a non-stationary signal and the efficient image representation capability of LBP. In the proposed method, DST based TFDs of PQ disturbance signals are considered as 2-D images. LBP histograms are used to extract the features from TF images. Initially, the uniform patterns in TF images are obtained by the LBP operator. Next, the occurrence histograms of these patterns are used to produce representative feature vectors that can capture the unique and salient characteristics of each disturbance. The classification performance of the proposed method is evaluated through total 2400 disturbance signals. The experimental results have shown that the rate of correct classification is about 98 % for the different PQ disturbances.

An efficient rotation invariant feature extraction method based on ring projection technique

This study presents an efficient rotation-invariant feature extraction method based on ring projection technique. The main advantage of this method is to reduce the number of sampling frequency of standard ring projection method. The proposed method is compared with the ring projection and local binary patterns according to the computational speed of the feature extraction and classification accuracy. By incrementally rotating first image of each texture class by 30 and 45 degrees enrich the dataset and yield two texture datasets having totally 1332 and 888 samples from the original Brodatz texture image dataset, respectively. Throughout the study Weka machine learning and data mining tool is utilized. As a classifier Naive Bayes, Bagging and J48 decision tree are used due to their simplicity and speed. Classification performance is evaluated via 10 fold cross validation technique. It is observed that, the proposed method outperforms other alternatives in terms of classification accuracy and feature extraction speed.