Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/202
Browse
8 results
Search Results
Conference Object Real-Time Robotic Car Control Using Brainwaves and Head Movement(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 2018) Ozturk, Nedime; Yilmaz, Bulent; Onver, Ahmet YasinEmotiv Epoc Headset is a portable and low-cost device. In this study, Emotiv Epoc headset was used in order to obtain real-time gyro and EEG signals. The aim of this study was to control a robotic car in real-time by using head movement and opening and closing of the eyes. The maximum and minimum amplitude of the gyro signal, and the ratios of the beta waves of O1 and O2 channel to alpha waves of the same channels were used as threshold values. These threshold values were used to determine the direction of the robotic car. Because of its low-cost and easy implementation, Arduino Uno was used to manage the robotic car. This study has shown that brain waves and head movements can control a device in real time. This system has the potential to be used in neurofeedback and brain-computer interface applications.Conference Object Effect of Bilinear Interpolation on the Texture Analysis of Colonoscopy Images(IEEE345 E 47TH ST, NEW YORK, NY 10017 USA, 2017) Kacmaz, Rukiye Nur; Yilmaz, BulentInterpolation is a method that is used to obtain unknown intensities with the help of known intensities on an image. This method is frequently used in the literature to eliminate light reflection on colonoscopy images. Texture features are the most important characteristics used to describe the region or objects of interest in the image. They are the measures of intensity variation of a surface that determine properties such as smoothness, roughness, and regularity. The aim of this study is to find out the how bilinear interpolation applied on colonoscopy images with reflection impact texture features obtained from the same images. A research carried out to make reasonable comparison between a texture feature from an image with no reflection and the same feature obtained from the same image with synthetically added reflections with various percentages. Using the approaches like gray level co-occurence matrix (GLCM), gray level run length matrix (GLRLM), neighborhood gray tone difference matrix (NGTDM) 126 features were extracted from each 32x32 sub-images coming from 610 colonoscopy images. Several of the features extracted from sub-images with no reflection and reflection were not statistically significantly different, while majority of them were affected from the reflections.Conference Object Detection of Variation Instances on Colonoscopy Videos using Structural Similarity Index(IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA, 2018) Kacmaz, Rukiye Nur; Yilmaz, BulentThe aim of this study is to reduce the number of images extracted from the videos recorded by the specialists during the colonoscopy process for further examination, thereby enabling the specialist to deal with fewer images. Since the images obtained from the videos are very similar, the main assumption of this study is that the whole video can be represented by fewer images. The approach used in this study is the structural similarity index. Totally, images were obtained from 4 different videos coming from healthy, ulcerative colitis, Crohn's, and polyp patients. The noisy images in these videos were eliminated manually. When the structural similarity index between two consecutive clear images was less than 0.83, the second image was selected and shown to the specialist for his/her examination. By this way, the frames carrying significantly new information from the videos were defined as the variation instances. The tests on healthy or diseased colon videos showed that only 5-10% of the clear images provide significantly new information.Conference Object Prognostic Significance of the Texture Features Determined Using Three Dimensional 18F-FDG PET Images: New Potential Biomarkers(Soc Nuclear Medicine inc, 2016) Karacavus, Seyhan; Yilmaz, Bulent; Kayaalti, Omer; Tasdemir, Arzu; Kaya, Eser; Icer, Semra; Asyali, MusaConference Object Fusion and Analysis of PET and CT Images of Patients With Non-Small Cell Lung Cancer(Springer, 2016) Ayyildiz, O.; Yilmaz, B.; Karacavus, S.; Kayaalti, O.; Eset, K.; Gazeloglu, C.; Kaya, E.Conference Object Machine Learning Assisted Particle Size and Type Classification Using Wavelength-Dependent Scattering Patterns(Wroclaw University of Science and Technology (WUST), 2021) Sinan Genc; Kutay Icoz; Talha ErdemThe presence of microplastics in oceans and water supplies have increased to critical levels within the last decade [1]. In addition to the huge mass of plastics in the seas, additional contribution to water pollution comes from our chemical wastes including toothpaste, detergents, cosmetics etc. All these pollutants end up in seas or clean water sources, which eventually affects the sea life but also the human health via the water consumption and the food chain [2]. Slowing down the microparticle pollution in water first relies on identifying and tracking these particles in a cost-effective manner so that the microparticles can be easily detected before they accumulate. To address this challenge, in this study, we investigated the scattering patterns of different microplastic samples at different concentrations in aqueous samples. By analyzing these scattering patterns obtained using blue, green, and red low-power lasers, we show that it is possible to classify the microparticles particles in terms of their size, concentration, and first time for the material type in a liquid sample thanks to random forest algorithm that accomplish the limited theoretical calculations. Figure 1: (a) Scattering behavior of 8 um Me samples under green laser with increasing concentration (b) scattering behavior of three different microplastics under red laser with 1.50 fM concentration. The aim of all these experiments was to show that the scattering patterns change for different type and size of the microplastics in liquid samples. As presented in Fig. 1(a), for the same excitation wavelength and material at the same size (green laser operating at 520 nm and 8 um-sized Melamine microparticles) the distance of the observed peaks and valleys from the center does not have any concentration dependence; nevertheless, the scattering intensity increases very strongly as the particle concentration increases. Especially the average intensities of the pixels, farthest away from the center turn out to be clear indicators of the microparticle concentration in water. In Fig. 1(b), we employed a red laser operating at 650 nm to record the scattering patterns of these particles all at 1.50 fM. The comparison between the melamine and polystyrene particles having the same size reveals peaks at different distances from the center and at different intensities originating from the different refractive indices of these particles.Conference Object Citation - Scopus: 5Emotion Detection Using Multivariate Synchrosqueezing Transform via 2D Circumplex Model(Institute of Electrical and Electronics Engineers Inc., 2018) Ozel, Pinar; Akan, Aydin; Yilmaz, Bulent; Özel, Pınar; Akan, Aydin I.; Yilmaz, BulentEmotion detection by utilizing signal processing methods is a challenging area. An open issue in emotional modeling is to obtain an optimum feature set to use for the classification process. This study proposes an approach for emotional state classification by the investigation of EEG signals via multivariate synchrosqueezing transform (MSST). MSST is a post-processing technique to compose a localized time-frequency representation yielding multivariate syncyrosqueezing coefficients. After obtaining these coefficients from EEG signals for 18 subjects from DEAP dataset, coefficients and self-assessment-mannequins (SAM) labels of those subjects are used for emotional state classification by using support vector machines (SVM) nearest neighbor, decision tree, and ensemble methods. The accuracy rate is 70.6% for high valence high arousal (HVHA), 75.4% for low valence high arousal (LVHA), 77.8% for high valence low arousal (HVLA), and 77.2% for low valence low arousal (LVLA) cases using SVM. © 2019 Elsevier B.V., All rights reserved.Conference Object Redesign of commercial color filters for color enriched LCD displays(Gdansk University of Technology (GUT),, 2018) Genç, Sinan; Uran, Can; Mutlugün, EvrenHaving as much as different colors on displays is the main aim for a high color gamut LCD. Using conventional backlight systems, a blue LED with a YAG phosphor layer implemented onto it, a high portion of CIE 1931 color space is missed [1,2]. Not only broad emission spectrum of Yttrium Aluminum Garnet (YAG) for yellow light, but also crosstalk of commercial RGB color filters have huge impact of that result. Using quantum dots (QDs) which are promising backlight agents in terms of color quality can increase the number of different colors on displays thanks to their narrow emission spectra, ease in controllability of optical properties and high photoluminescence efficiency [3:5]. However, when it comes to the color filters, broad transmission spectra and crosstalk between those spectra reduces the quality [6]. In this study, we design, simulate, analyze a QD based backlighting system and compare it with conventional phosphor based white light. Simulating both yellow phosphor based LED and QD based LED in software, we engineer spectral parameters i.e. full width at half maximum, peak emission wavelength and intensities of emitters. Furthermore, we investigate the effect of commercial color filters on those two systems and propose a new, industrially appropriate color filter spectra. Using QD based backlight increases the NTSC color gamut area from 65-70% to 127% with more than 99.8% coverage and the negative effect of commercial color filters, around 15% that reduced the gamut ratio to 109%, is balanced with suggested spectral transmission parameters of RGB color filters for QD based backlighting systems.