Yazar "Barstugan, Mucahid" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim Yok
    Öğe
    Classification of Mammogram Images by Dictionary Learning
    (IEEE, 2017) Barstugan, Mucahid; Ceylan, Rahime
    Dictionary Learning is a method used in signal and image processing. In this study, classification of mammogram images were realized by using dictionary learning and sparse representation algorithms. The attributes of the images were detected with Wavelet Transform and PCA, and the new dataset which was created by the obtained attributes were classified by Dictionary Learning. Moreover, the classification performance of the Dictionary Learning algorithm was evaluated by classifying the new dataset with SVM, Rotation Forest and AdaBoost algorithms,. The best classification accuracy was obtained by PCA-Dictionary Learning algorithm as 98.89%.
  • Küçük Resim Yok
    Öğe
    A Discriminative Dictionary Learning-AdaBoost-SVM Classification Method on Imbalanced Datasets
    (IEEE, 2017) Barstugan, Mucahid; Ceylan, Rahime
    Sparse representation is a signal processing method which is mostly used in signal compression, noise reduction, and signal and image restoration fields. In this study, sparse representation was used in a different way from the traditional methods. In the proposed method, a hybrid structure was created by combining dictionary learning and ensemble classifier AdaBoost algorithms. The main idea of this method is to obtain the sparse coefficients from an over-complete dictionary and to use the coefficients in the weight update formula of AdaBoost. Support Vector Machines (SVM) classifier was used as weak classifiers of AdaBoost, and AdaBoost-SVM classifier structure was created. Multiplying the sparse coefficients with weight of weak learners process in weight update formula has given satisfying results on imbalanced datasets during the experiments.
  • Küçük Resim
    Öğe
    Feature Selection using FFS and PCA in biomedical data classification with AdaBoost-SVM
    (2018) Ceylan, Rahime; Barstugan, Mucahid
    Recently, there has been an increasing trend to propose computer aided diagnosis systems for biomedical pattern recognition. A computer aided diagnosis method, which aims higher classification accuracy, is developed to classify the biomedical dataset. This process includes two types of machine learning algorithms: feature selection and classification. In this method, firstly, features were extracted from biomedical dataset, then the extracted features were classified by hybrid AdaBoost-Support Vector Machines (SVM) classifier structure. For feature selection, Forward Feature Selection (FFS) and Principal Component Analysis (PCA) algorithms were used, and the performance of the feature selection algorithms was tested by AdaBoost-SVM classifier. Following it, advantages and disadvantages of these algorithms were evaluated. Wisconsin Breast Cancer (WBC), Pima Diabetes (PD), Heart (Statlog) biomedical datasets were taken from UCI database and Electrocardiogram (ECG) signals were taken from Physionet ECG Database, and were used to test the proposed hybrid structure. The used two hybrid structures and other studies in the literature were compared with our findings. The obtained results show that the proposed hybrid structure has high classification accuracy for biomedical data classification
  • Küçük Resim Yok
    Öğe
    Full-Automatic Liver Segmentation on Abdominal MR Images
    (IEEE, 2018) Barstugan, Mucahid; Ceylan, Rahime; Asoglu, Semih; Cebeci, Hakan; Koplay, Mustafa
    Liver segmentation process is a challenging field in computer-aided medical image analysis. This study implemented liver segmentation on Abdominal MR images. The liver was automatically segmented from images by morphological methods with high performance. Liver segmentation process was implemented on 56 MR images and the segmentation results were examined. In this study, an effective and fast method was proposed. Seven evaluation metrics (sensitivity, specificity, accuracy, precision, Dice coefficient, Jaccard rate, Structural Similarity Index (SSIM)) were used to test the performance of the proposed method. Mean Dice coefficient value was obtained as 91.701%, mean Jaccard rate value was obtained as 85.052% on 56 images. Segmentation duration for an image (T1 and T2 phases) was found as 2.828 seconds with the proposed method.