http://hdl.handle.net/11508/11923 Mon, 27 Apr 2026 13:40:33 GMT 2026-04-27T13:40:33Z http://hdl.handle.net/11508/21060 Derin öğrenme ile beyin MRI görüntülerinde süper çözünürlük: SRCNN, SRGAN ve ESRGAN yaklaşımları Ünlü, Elif Işılay; Çınar, Ahmet Tıbbi görüntüleme teknolojileri, hastalıkların doğru teşhis ve tedavisinde hayati bir öneme sahiptir. Beyin MRI görüntüleri, nörolojik hastalıkların tanı ve takibinde detaylı anatomik bilgi sunar. Ancak, teknik sınırlamalar ve donanım kapasiteleri nedeniyle elde edilen MRI görüntüleri çoğu zaman düşük çözünürlüklü olmaktadır. Bu çalışmanın amacı, düşük çözünürlüklü beyin MRI görüntülerini iyileştirmek için SRCNN (Super-Resolution Convolutional Neural Network), SRGAN (Super-Resolution Generative Adversarial Network) ve ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) yöntemlerinin uygulanması ve karşılaştırılmasıdır. Modellerin performansları, görsel kalite, yapısal benzerlik indeksi (SSIM) ve tepe sinyal-gürültü oranı (PSNR) gibi ölçütlerle değerlendirilmiştir. Elde edilen bulgular, ESRGAN modelinin daha keskin detaylar ve daha gerçekçi görüntüler ürettiğini ortaya koymaktadır. Bu çalışma, derin öğrenme tabanlı süper çözünürlük tekniklerinin tıbbi görüntülemede sunduğu yenilikçi olanaklara dikkat çekmektedir. Mon, 30 Dec 2024 00:00:00 GMT http://hdl.handle.net/11508/21060 2024-12-30T00:00:00Z http://hdl.handle.net/11508/20928 Classification of skin images with respect to melanoma and nonmelanoma using the deep neural network Ünlü, Elif Işılay; Çınar, Ahmet Melanoma is the most common type of skin cancer. At first, for the diagnosis of melanoma, clinical screening is performed and then diagnosis is made by clinical imaging. It is followed up by dermoscopic analysis, biopsy and histopathological examination. Early diagnosis is important in the treatment of melanoma. Automatic recognition of melanoma from dermoscopy images is a difficult task. Therefore, computer aided systems are recommended to reduce time ,cost and accuracy diagnosis. In this paper, a deep learning-based system is used to classify melanoma in color images taken from dermoscopy devices. With this system, differentiation from previous studies can be done with good accuracy without segmentation step and feature extraction. This system provides a significant advantage in hardware implementation. Because there are no preprocessing and segmentation steps. The International Skin Imaging Collaboration database for the designed system is used and includes 1483 training, 517 test data(ISIC). As a result of the classification of these data, the success rate is reached 86-85%. Fri, 28 Dec 2018 00:00:00 GMT http://hdl.handle.net/11508/20928 2018-12-28T00:00:00Z http://hdl.handle.net/11508/12166 Efficient android electronic nose design for recognition and perception of fruit odors using Kernel Extreme Learning Machines Uçar, Ayşegül; Özalp, Recep This study presents a novel android electronic nose construction using Kernel Extreme Learning Machines (KELMs). The construction consists of two parts. In the first part, an android electronic nose with fast and accurate detection and low cost are designed using Metal Oxide Semiconductor (MOS) gas sensors. In the second part, the KELMs are implemented to get the electronic nose to achieve fast and high accuracy recognition. The proposed algorithm is designed to recognize the odor of six fruits. Fruits at two concentration levels are placed to the sample chamber of the electronic nose to ensure the features invariant with the concentration. Odor samples in the form of time series are collected and preprocessed. This is a newly introduced simple feature extraction step that does not use any dimension reduction method. The obtained salient features are imported to the inputs of the KELMs. Additionally, K-Nearest Neighbor (K-NN) classifiers, the Support Vector Machines (SVMs), Least-Squares Support Vector Machines (LSSVMs), and Extreme Learning Machines (ELMs) are used for comparison. According to the comparative results for the proposed experimental setup, the KELMs produced good odor recognition performance in terms of the high test accuracy and fast response. In addition, odor concentration level was visualized on an android platform. Thu, 01 Jun 2017 00:00:00 GMT http://hdl.handle.net/11508/12166 2017-06-01T00:00:00Z