Profile URL: https://hdl.handle.net/20.500.13091/12013
Job Title:Arş. Gör. Uzm.
Email Address:kbyapar@ktun.edu.tr
Main Affiliation:10.02. Department of Software Engineering
Status: Current Staff
ORCID:
0009-0006-7666-6177
0009-0006-7666-6177YÖK Akademik: F84882EE0453F9A2
Name Variants:
Yapar, K. B. Yapar, K. Bugrahan
3 results
Scholarly Output Search Results
Now showing 1 - 3 of 3
Article Comparative Study on Most Used Modern Symmetric Encryption Algorithms for Breast Cancer Ultrasound Images(2025-12-25) Doğan, Nurettin; Yapar, Kürşad BuğrahanIn this study, several prominent symmetric encryption algorithms, including AES, Blowfish, DES, 3DES, and RC4, were applied to encrypt breast ultrasound images. These algorithms were further compared with RSA, a representative asymmetric encryption method. Three different breast ultrasound images, categorized as benign tumor, malignant tumor, and normal tissue, were used in the encryption process. Each image was provided in two file formats (PNG and BMP) and at two resolutions (256 × 256 and 512 × 512 pixels). The performance of the algorithms was evaluated using metrics such as the Number of Pixel Change Rate (NPCR), Unified Average Changing Intensity (UACI), histogram analysis, and encryption and decryption times. The experimental results demonstrated that symmetric encryption algorithms were more efficient than their asymmetric counterpart. Additionally, a comparative analysis of the symmetric algorithms was conducted, along with a discussion of their relative performance.Article Digital Detox and Physical Activity: Enhancing Sleep Quality in University Student(Association of Mathematicians (MATDER), 2026-02-23) Yapar K.B.; Doğan N.; Yapar, Kürşad Buğrahan; Doğan, NurettinExcessive smartphone use among university students, driven by activities such as social media engagement and online content consumption, has raised concerns about its impact on sleep quality. This study examines how reducing smartphone use and increasing physical activity affect sleep quality and daytime fatigue in this population. A single-group pretest-posttest quasi-experimental approach was utilized, involving a convenience sample of university students (mean age: 19.83±1.39 years, BMI: 22.56±3.29). The data for this study were obtained through the administration of the Personal Information Form, the Smartphone Addiction Scale-Short Version (SAS-SV), the Pittsburgh Sleep Quality Index (PSQI), and the Epworth Sleepiness Scale (ESS). The intervention reduced daily smartphone use by 30 minutes and increased physical activity by five minutes over one week. Following the intervention, notable improvements were observed in all subdimensions of the PSQI (p < 0.001), accompanied by a statistically significant reduction in ESS scores (p < 0.05), indicating enhanced sleep quality and reduced daytime drowsiness. These findings suggest that smartphone addiction negatively affects sleep, while digital detox and physical activity interventions enhance sleep quality. From a human-computer interaction standpoint, this research emphasizes the effectiveness of user-oriented digital solutions in promoting healthier smartphone habits and supporting student well-being. © MatDer.Master Thesis Derin öğrenme modelleri ile portakal meyvesi görüntülerinden hastalık tespiti(2026) Yapar, Kürşad Buğrahan; Öksüz, Özgür; Doğan, NurettinCitrus diseases are recognized as a major issue that threatens fruit quality and agricultural productivity, leading to severe economic losses. Traditional diagnostic methods present significant limitations in practical use due to their labor-intensive nature, high costs, and dependence on expert evaluation. This increases the demand for automated and reliable solutions. In this study, three state-of-the-art deep learning architectures 'EfficientNet-B0, ConvNeXt-Tiny, and Vision Transformer (ViT-B/16)' were systematically compared on the Orange Fruit dataset, which includes healthy, citrus canker, and melanose samples. To enhance the generalization ability of the models and reduce the risk of overfitting, transfer learning was applied using networks pretrained on ImageNet, and advanced data augmentation strategies such as Mixup and Cutout were integrated. Experimental findings indicate that the ConvNeXt-Tiny architecture, supported by pretraining and data augmentation strategies, achieved superior performance compared to the other models. This model yielded the best results with an Accuracy of 98.76%, F1-Score of 99%, ROC-AUC of 99.95%, Precision of 99%, and Recall of 99%. Beyond numerical performance metrics, the transparency of model decisions was also addressed. Within the scope of explainable artificial intelligence (XAI), the widely adopted Grad-CAM method was employed alongside RISE, which is considered one of the first applications in agricultural contexts. RISE provided a more holistic visualization of the regions on the fruit surface influencing the decision process, thereby strengthening interpretability. The results demonstrate that modern deep learning models have significant potential for the early detection of citrus diseases, based on evaluations conducted on orange fruit images. Furthermore, the integration of XAI methods offers a substantial contribution to enhancing the reliability and transparency of digital agricultural decision-support systems.
No research topics data found.
Sustainable Development Goals
3GOOD HEALTH AND WELL-BEING
1
Research Products
This researcher does not have a Scopus ID.
This researcher does not have a WoS ID.
No records found in other affiliations.
| Journal | Count |
|---|---|
| International Advanced Researches and Engineering Journal | 1 |
| Turkish Journal of Mathematics and Computer Science | 1 |
Current Page: 1 / 1
Scholarly Output
3
Articles
2
Views / Downloads
16/0
Supervised MSc Theses
1
Supervised PhD Theses
0
WoS Citation Count
0
Scopus Citation Count
0
Patents
0
Projects
0
WoS Citations per Publication
0.00
Scopus Citations per Publication
0.00
Open Access Source
2
Supervised Theses
1
Scopus Quartile Distribution
Competency Cloud
