SavedAdd your journal

Journals / Uluslararası Teknolojik Bilimler Dergisi / 2026 / Cilt: 18 Sayı: 2026

Comparative analysis of deep learning models for real-time knife detection

Gerçek zamanlı bıçak tespiti için derin öğrenme modellerinin karşılaştırmalı analizi

Mukaddes Karabıyık(ISPARTA UYGULAMALI BİLİMLER ÜNİVERSİTESİ)0009-0002-5835-6651
Kıyas Kayaalp(ISPARTA UNIVERSITY OF APPLIED SCIENCES)0000-0002-6483-1124
Pages
—
Publication Date
2026-09-24
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The overall findings provide a holistic experimental framework for selecting the most appropriate model based on the specific requirements of the application scenario.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study investigates the effectiveness of contemporary deep learning architectures for the automated detection of bladed weapons, which represents one of the most critical requirements of modern security systems.
📊
Key Findings & Data:Within the scope of the research, a comparative analysis was conducted using a comprehensive dataset representing diverse environmental conditions, evaluating YOLOv8, RT-DETR, RetinaNet, and SSD + MobileNetV3 models.
💡
Conclusion & Impact:The overall findings provide a holistic experimental framework for selecting the most appropriate model based on the specific requirements of the application scenario.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Derin Öğrenme · Nesne Tespiti · Bıçak Tespiti

Abstract

This study investigates the effectiveness of contemporary deep learning architectures for the automated detection of bladed weapons, which represents one of the most critical requirements of modern security systems. Within the scope of the research, a comparative analysis was conducted using a comprehensive dataset representing diverse environmental conditions, evaluating YOLOv8, RT-DETR, RetinaNet, and SSD + MobileNetV3 models. The performance of these models was assessed not only through static imagery but also via real-time video streams captured by a computer camera to evaluate their practical application capabilities. Through this study, the trade-offs between speed and accuracy of different architectures were analyzed, and their performance characteristics under varying environmental conditions were determined. The overall findings provide a holistic experimental framework for selecting the most appropriate model based on the specific requirements of the application scenario.

Özet

Bu çalışma, modern güvenlik sistemlerinin en kritik ihtiyaçlarından biri olan kesici aletlerin otomatik tespiti için güncel derin öğrenme mimarilerinin etkinliğini araştırmaktadır. Araştırma kapsamında, farklı çevresel koşulları temsil eden geniş kapsamlı bir veri seti kullanılarak; YOLOv8, RT-DETR, RetinaNet ve SSD + MobileNetV3 modelleri üzerinden karşılaştırmalı bir analiz yürütülmüştür. Modellerin performansı sadece statik görseller üzerinden değil, aynı zamanda bilgisayar kamerası aracılığıyla alınan gerçek zamanlı video akışları üzerinde de test edilerek pratik uygulama kabiliyetleri değerlendirilmiştir. Çalışma ile farklı mimarilerin hız ve doğruluk dengeleri analiz edilerek, değişken çevresel koşullardaki performans karakteristikleri belirlenmiştir. Elde edilen genel bulgular, uygulama senaryosunun gereksinimlerine göre en uygun modelin seçilmesine yönelik bütüncül bir deneysel çerçeve sunmaktadır.

🔖

Cite This Article

Karabıyık, M., Kayaalp, K. (2026). Comparative analysis of deep learning models for real-time knife detection. Uluslararası Teknolojik Bilimler Dergisi, Cilt: 18 Sayı: 2026. https://doi.org/10.55974/utbd.1881140

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (16)

  1. United Nations Office on Drugs and Crime. Global Study on Homicide 2023. https://www.unodc.org/unodc/en/data-and-analysis/global-study-on-homicide.html (Erişim Tarihi: 17.01.2026).
  2. Zhao ZQ, Zheng P, Xu ST, Wu X. Object detection with deep learning: A review. IEEE Transactions on Neural Networks and Learning Systems, 30(11), 3212-3232, 2020.
  3. Pérez-Hernández F, Tabik S, Lamas A, Olmos R, Fujita H, Herrera F. Object detection binary classifiers methodology based on deep learning to identify small objects handled similarly: Application in video surveillance. Knowledge-Based Systems, 194, 105590, 2020.
  4. Zulfqar S, Saeed S, Zia MA, Ali A, Mehmood F, Ali A. A comprehensive overview of deep learning models for object detection from videos/images. International Journal of Artificial Intelligence and Soft Computing, 2026.
  5. Amador-Salgado Y, Padilla-Medina JA, Pérez-Pinal FJ, Barranco-Gutiérrez AI, Rodríguez-Licea MA, Martínez-Nolasco JJ. Knife detection using indoor surveillance camera. 2021 IEEE World AI IoT Congress (AIIoT), Texas, USA, 10-13 May 2021.
  6. Berardini D, Migliorelli L, Galdelli A, Frontoni E, Mancini A, Moccia S. A deep-learning framework running on edge devices for handgun and knife detection from indoor video-surveillance cameras. Multimedia Tools and Applications, 83(7), 19109-19127, 2024.
  7. Ağdaş MT, Gülseçen S. Güvenlik kameralarında otomatik silah ve bıçak tespit sistemi: Karşılaştırmalı YOLO modelleri. Avrupa Bilim ve Teknoloji Dergisi, (41), 16-22, 2022.
  8. Huynh SPT, Dang TQ, Nguyen HD, Tran PNH, Nguyen NQV. Knife detection using YOLOv5: A deep learning approach. Proceedings of the 2024 9th International Conference on Intelligent Information Technology, Da Nang, Vietnam, 23-25 February 2024.
  9. Ku FS, Suguna M. Subduing crime and threat in real-time by detecting weapons using YOLOv8. 2023 International Conference on Circuit Power and Computing Technologies (ICCPCT), Kollam, India, 10-11 August 2023.
  10. Roboflow. Knife Detection Merge Dataset. https://app.roboflow.com/mayk/knifesdetectionmerge (Erişim Tarihi: 18.01.2026).
  11. Roboflow. Knife Detection (knife-yc1b3). https://app.roboflow.com/mayk/knife-yc1b3-lhgqv/1 (Erişim Tarihi: 18.01.2026).
  12. Roboflow. Knife Detection (knife-detection-lpiw1). https://app.roboflow.com/mayk/knife-detection-lpiw1-7pekn/1 (Erişim Tarihi: 18.01.2026).
  13. Sohan M, Sai Ram T, Rami Reddy CV. A review on YOLOv8 and its advancements. International Conference on Data Intelligence and Cognitive Informatics, Singapore, 18-19 December 2024.
  14. Adnan MA, Ahammed F, Rahman MF, Alam N, Prince IA, Ahamed MT. Raspberry Pi-powered IoT device for blind navigation using MobileNetV3-SSD image processing. 2023 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD), Dhaka, Bangladesh, 22-23 September 2023.
  15. Zhao Y, Lv W, Xu S, Wei J, Wang G, Dang Q, Chen J. DETRs beat YOLOs on real-time object detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA, 16-22 June 2024.
  16. Tan L, Huangfu T, Wu L, Chen W. Comparison of RetinaNet, SSD, and YOLO v3 for real-time pill identification. BMC Medical Informatics and Decision Making, 21(1), 324, 2021.