Yazar "Koca, Merve" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • [ X ]
    Öğe
    Comparative Analysis of SSD and Faster R-CNN in UAV-Based Vehicle Detection
    (Institute of Electrical and Electronics Engineers Inc., 2024) Hansen, Kristine S.; Bruun, Frederikke M.; Sermsar, Funda; Nygaard, Mette; Koca, Merve
    Artificial intelligence-based methods for monitoring transportation networks and vehicles play a crucial role in enhancing forensic analysis and security applications. Continuous surveillance enabled by object detection algorithms allows real-time monitoring of roads and highways, facilitating tasks such as traffic flow monitoring, accident detection, and identification of suspicious vehicles or behaviors. Integrating these algorithms into surveillance systems supports law enforcement in swiftly locating vehicles of interest and responding effectively to incidents, thereby improving security measures and enhancing forensic investigations through detailed analysis of surveillance footage. Furthermore, object detection aids in optimizing traffic management by identifying congestion points and optimizing traffic signals, thus enhancing road safety and mobility. This study evaluates the performance of SSD and Faster R-CNN in vehicle detection using UAV-based aerial imaging, providing insights into their strengths and limitations for applications such as aerial surveillance and traffic monitoring. By comparing these algorithms comprehensively, this study aims to guide the selection of the most suitable model for effective vehicle detection in diverse operational environments. The findings contribute to advancing AI applications in transportation and security, offering insights into optimizing surveillance systems for enhanced safety, efficiency, and responsiveness in managing urban mobility and security challenges. © 2024 IEEE.
  • [ X ]
    Öğe
    Enhancing Performance in Mobile Ad-Hoc Networks via Expanded View of Network Topology
    (Institute of Electrical and Electronics Engineers Inc., 2024) Zehre, Kerem Mert; Dalay, Mehmet Faruk; Koca, Merve; Alam, Soude Felouz; Yildiz, Mukerrem
    This work discusses the important role of routing protocols in Mobile Ad-Hoc Networks (MANET) and highlights efficient information transfer via dynamic mapping and single-hop transmission services. Fisheye State Routing (FSR), an evolution towards large networks, is being investigated for its simplicity, efficiency, and scalability. The study comprehensively analyzes the behavior of FSR in grid environments by evaluating key performance metrics. The findings have proven FSR to be effective, demonstrating superior performance and bandwidth optimization. The application outputs simulated in the network simulator verify the ability of FSR in packet delivery speed and throughput efficiency, providing decisive insights for MANET optimization. © 2024 IEEE.
  • [ X ]
    Öğe
    Performance Comparison of Faster R-CNN and SSD in Vehicle Detection from Aerial Traffic Videos
    (Institute of Electrical and Electronics Engineers Inc., 2024) Sermsar, Funda; Moghadam, Mahrus Thabit; Koca, Merve
    Vehicle detection is essential for urban planning, traffic management, and various security and surveillance applications. This study compares the performance of two popular deep learning algorithms, Faster R-CNN and SSD, for vehicle detection using traffic video recordings obtained from unmanned aerial vehicles. The dataset, comprising images taken from diverse angles, heights, and lighting conditions, was utilized to assess the effectiveness of both models. Faster R-CNN demonstrated superior accuracy in detecting vehicles and motorcycles under various conditions, including night scenarios and scenes with obstacles such as shadows, traffic signs, and light poles. However, it struggled with detecting very small objects and those with color tones similar to the background. In contrast, while SSD performed well under optimal lighting, it exhibited limitations in detecting smaller vehicles partially obscured by environmental elements. Quantitative analysis revealed that Faster R-CNN had 13.21% higher precision, 12.5% higher recall, and 10.04% higher mAP compared to SSD. Despite its superior detection performance, Faster R-CNN's longer detection time indicates that SSD may be more suitable for real-time applications where speed is crucial. © 2024 IEEE.