Yıldız Technical University Research Information System
Not a member yet
91324 research outputs found
Sort by
Contactless biometric verification from in-air signatures using deep siamese networks
No full textIn-air signature is a behavioral biometric trait that has gained increasing attention in recent years due to its contactless nature and potential for secure, hygienic, and remote authentication. Unlike traditional pen-and-paper or tablet-based systems, in-air signature methods capture signing gestures in three-dimensional space, typically using fingertip tracking or depth sensing, offering greater flexibility and accessibility in various application contexts. In this study, we developed a deep learning-based biometric verification model using in-air signature data collected from 25 participants. The collected dataset was structured into 200 positive (same-person) and negative (different-person) signature pairs, capturing both inter-person and intra-person variability. A Siamese Neural Network architecture based on Bidirectional LSTM layers and contrastive loss was used to learn a discriminative embedding space for signature verification. To rigorously evaluate generalization capability across users, we employed a customized cross-validation protocol based on the Leave Two Sample Out (LTSO) approach, a more stringent variation of the traditional Leave One Sample Out (LOSO) method, resulting in 300 unique train-test splits. The proposed system achieved strong overall performance, with an average accuracy of 85%, F1-score of 85%, and recall of 91%, indicating its effectiveness even with limited training data. These results demonstrate the feasibility of using in-air signatures as a practical, contactless biometric modality and support the viability of Siamese neural networks for learning person-specific patterns in motion-based verification tasks.</p
Use of waste animal blood as a bio-based air-entraining additive in the production of sustainable hydraulic lime-based grouts
No full textIdentification of biobutanol-associated genes and pathway limitations in non-virulent<i> Bacillus</i><i> anthracis</i> 3B1 through functional genome analysis and fermentation optimization
No full textBiobutanol is a promising biofuel alternative that addresses the energy crisis and reduces the environmental impact of fossil fuels. This study investigates the metabolic potential of Bacillus anthracis 3B1, isolated from rice cultivated under the system of rice intensification (SRI). As a facultative anaerobe, Bacillus anthracis 3B1 may offer greater metabolic flexibility and higher tolerance to butanol compared to strictly anaerobic Clostridium. Further gene annotation revealed that the genome of B. anthracis 3B1 lacks virulence genes such as pag, cya, and lef, indicating that the strain is non-virulent. The study integrates descriptive, exploratory, and experimental approaches by combining whole genome sequencing with the screening of various fermentation factors to optimize biobutanol yields and fermentation efficiency, supporting its application in sustainable bioenergy solutions. Functional genome analysis revealed key genes and enzymes involved in butanol biosynthesis. Annotation using the Rapid Annotations using Subsystems Technology (RAST) platform identified a butanol biosynthesis subsystem. Further functional annotation through Clusters of Orthologous Groups (COG), Gene Ontology (GO), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) via eggNOG-mapper indicated the presence of genes encoding butanol-related enzymes, although KEGG analysis suggested an incomplete pathway. Despite these genomic indicators, no butanol was detected under the tested fermentation conditions. However, the strain produced metabolites such as propanol, ethanol, acetoin, carbon dioxide, and acetic acid. Fermentation experiments showed up to 72.57 % glucose consumption and a 0.5 pH drop, indicating active metabolism. These findings suggest that optimizing fermentation or metabolic engineering may be needed to realize B. anthracis 3B1 butanol potential
Hot-Pressing Process of 7075 Aluminum Alloy: Microstructure, Corrosion and Wear Properties
No full textIn this study, 7075 aluminum alloy was fabricated by hot-pressing process with various production temperatures and times. It was also aimed to reveal the effects of hot-pressing process parameters on microstructure, corrosion and wear properties of the 7075 aluminum. Samples were produced at 450 °C, 475 °C and 500 °C with 30-min and 60-min process duration under 200 MPa pressure and 0.1-bar vacuum. Microstructural analysis revealed that the magnesium- and zinc-containing secondary phases decomposed under the enhanced steady cooling conditions, which facilitated sufficient diffusion. It was observed that continuous grain boundary phases transformed into a discontinuous structure via the diffusion of magnesium with increasing process temperature. The corrosion behavior of the samples was dependent on the microstructure, and also the discontinuous morphology of the magnesium- and copper-rich secondary phases decreased the corrosion rate down to 0.1746 mm/year. On the other hand, the formation of secondary phases negatively affected the wear resistance, and lower surface damage was obtained in samples produced at 450 °C for 30 minutes and 60 minutes
Real-time performance monitoring of microwave line-of-sight links in mobile communication using fiber Bragg grating sensors for vibration and acceleration detection at mobile base stations
No full textThis study presents a real-time performance monitoring approach for microwave line-of-sight links in mobile base stations, utilizing triaxial fiber Bragg grating (FBG) accelerometers to detect and analyze vibration- and acceleration-induced disturbances that affect the radio link system. A total of seven experimental test scenarios were conducted under controlled external excitations. Key performance indicators, including background block error rate (BBER), errored seconds (ES), errored seconds rate (ESR), severely errored seconds (SES), and severely errored seconds rate (SESR), were analyzed to quantify the effects of mechanical disturbances on radio link stability. Experimental results show that high-amplitude accelerations cause temporary degradation in channel quality, leading the system to downgrade its modulation scheme from high levels, such as 4096 quadrature amplitude modulation (QAM), to lower levels, like 4 QAM. These degradations were reversible, as modulation levels returned to higher states once channel conditions improved. Additionally, performance degradations were found to be associated with physical factors such as structural contact issues at the radio link pole, external interferences, or wind-induced oscillations. Unlike conventional methods, the proposed system enables remote and real-time detection of environmental mechanical anomalies. This work introduces a novel framework integrating acceleration measurements with link performance metrics, establishing a foundation for predictive maintenance and autonomous fault detection in wireless communication infrastructure
Extension of the VIKOR method in a decomposed fuzzy environment for addressing inconsistencies in decision-making
No full textThis paper introduces a novel extension of the VIseKriterijumska Optimizacija I Kompromisno Resenje extended with Decomposed Fuzzy Sets, denoted as “Decomposed Fuzzy VIKOR,” which aims to address the challenges of inconsistent responses and indecision among decision-makers. As a recent contribution to fuzzy set extensions, Decomposed Fuzzy Set differ from other extensions by employing bidirectional questioning. The aim of bidirectional questioning is to allow decision-makers to provide responses by posing questions that are equivalent in meaning but represent two opposing perspectives: functional (optimistic) and dysfunctional (pessimistic). In this study, six criteria obtained from a comprehensive literature review are considered, and the Decomposed Fuzzy VIseKriterijumska Optimizacija I Kompromisno Resenje method is applied to a cold food supply chain to identify the most suitable alternatives based these criteria. The results are compared with those obtained using the Intuitionistic Fuzzy VIseKriterijumska Optimizacija I Kompromisno Resenje method, revealing the impact of different scenarios on optimal alternatives. The paper concludes by emphasizing the significance of integrating various multi-criteria decision-making methods with Decomposed Fuzzy Set for real-life problem-solving
KENT KİMLİĞİNİ YANSITACAK VE İKLİM DEĞİŞİKLİĞİNE UYUM SAĞLAYABİLECEK ODUNSU BİTKİLER: BARTIN
No full textThe Early Reception of Russell’s Scientific Philosophy in Turkey
No full textThis paper presents the early reception of Bertrand Russell’s scientific philosophy in 1930s Turkey, which reflects a strong interest for philosophical currents related to the sciences, such as pragmatism and logical empiricism. It is exemplified by the first translations, one of them by the historian ofscience Abdülhak Adnan Adıvar (1936), but also by the first presentations of Russell’s philosophy by Mehmet Saffet (1933b), a former student of John Dewey, and Hilmi Ziya Ülken (1936), who worked with Hans Reichenbach during his stay at Istanbul University between 1933 and 1938. The studyof this early reception of Russell’s scientific philosophy, which should be situated in the context of the Turkish pedagogical policies in the 1930s, questions the significance it had for early developments in the field of philosophy of science in Turkey but also contributes to writing the early history of analytic philosophy in Europe from a transnational perspective.</p
Inhibition effects of amino acids on polyphenol oxidase activity isolated from medlar fruit
No full textIn this study, medlar polyphenol oxidase (PPO) was partially purified by (NH4)2SO4 precipitation and dialysis, respectively. The aim of the study was to investigate the inhibition effects of amino acids, which are candidate PPO ligands, on the activity of the medlar PPO enzyme for advanced biochemical purification techniques and to create a usage field for the enzyme inhibitors in different industrial sectors. No any inhibition studies of amino acids have been investigated on medlar PPO in literature yet. Inactivation of PPO is preferred to be prevention of decreasing of nutritional quality and shelf life of foods. Two bands were determined in electrophoresis analyses. Following, amino acids effects were studied on medlar PPO activity to investigate the potentials of Glycine (Gly), L-Phenylalanine (L-Phe), L-Tyrosine (L-Tyr), L-Cysteine (L-Cys), L-Serine (L-Ser), L-Aspartic acid (L-Asp), L-Histidine (L-His), L-Lysine (L-Lys), L-Proline (L-Pro), and L-Methionine (L-Met) whether acting as natural PPO inhibitors. Inhibition types were determined for catechol and L-Cys was found as a potent competitive inhibitor of medlar PPO. While Gly, L-Phe, L-Pro, L-Ser, L-His, and L-Lys showed uncompetitive inhibition; L-Tyr, L-Asp, and L-Met showed mixed-type inhibition. Statistical analysis was performed to understand whether the chemical structure or concentration of inhibitors showing the same type of inhibition made a statistically significant difference on the enzyme activity %. The results showed that the structure of inhibitors did not make a statistically significant difference on the enzyme activity % while inhibitor concentration created significant difference
Ecofriendly high-performance low-cost Cu<sub>2</sub>ZnSnSe<sub>4</sub> solar cells: Experimental characterization and SCAPS-1D simulation
No full textCu2ZnSnSe4 (CZTSe) is a promising low-cost and ecofriendly p-type semiconductor for solar cells. However, CZTSe solar cells are have shortcomings, as they typically involve hazardous CdS-buffer layers. Costly dopant elements Ag and/or Ge are also normally used. This study aims at producing efficient, ecofriendly and low-cost solar cells. The Cd-, Ag- and Ge-free configuration metal/MoSe2/CZTSe/ZnSe/i-ZnO/ZnO-Al/metal is proposed. CZTSe film is prepared by the facile and low-cost sol-gel method, and characterized by elemental analysis, optical-absorption spectra, surface morphology, surface profiling and wettability. The cell is simulated by SCAPS1D. Optimal CZTSe-layer thickness is 2.5 mu m with optimized doping concentration 5 x 1016 cm- 3. With these parameters, the cell exhibits an open-circuit potential 0.56 V, a short-circuit potential 47.33 mA/cm2 and a fill factor 73.82 %. With a cell conversion efficiency 19.5 %, the proposed cell outperforms earlier CZTSe cells in terms of cost and environmental friendliness. This opens new research inroads toward improved CZTSe-based solar cells