DESIGN OF A DATASET FOR WEAPON DETECTION-BASED SECURITY SYSTEMS
Keywords:
Dataset, computer vision, firearms, video surveillance, security, training, model, YOLOAbstract
DOI: https://doi.org/10.46296/ig.v9i18.0354
Abstract
Automatic threat detection through computer vision has become an increasingly important approach for enhancing video surveillance systems. However, the performance of these detection models largely depends on the quality of the data used for training. In this context, the present research aims to design a specialized dataset for the detection of armed and unarmed individuals, with the objective of serving as a foundation for the development of intelligent monitoring systems. To achieve this, the CRISP-ML(Q) methodology was employed, dividing the dataset creation process into the stages of data collection, selection, preparation, and validation. The dataset was constructed from videos obtained from publicly available sources across Ecuador and subsequently annotated for use in model training. Finally, to validate the proposed dataset, a YOLO-based object detection model was trained. The obtained results demonstrated satisfactory performance, indicating that the developed dataset possesses suitable characteristics for training automatic threat detection models in video surveillance environments.
Keywords: Dataset, computer vision, firearms, video surveillance, security, training, model, YOLO.
Downloads
References
Ahmed, S., Bhatti, M. T., Khan, M. G., Lövström, B., & Shahid, M. (2022). Development and Optimization of Deep Learning Models for Weapon Detection in Surveillance Videos. Applied Sciences, 12(12), 5772. https://doi.org/10.3390/app12125772
Amado-Garfias, A. J., Conant-Pablos, S. E., Ortiz-Bayliss, J. C., & Terashima-Marín, H. (2024). Improving Armed People Detection on Video Surveillance Through Heuristics and Machine Learning Models. IEEE Access, 12, 111818-111831. https://doi.org/10.1109/ACCESS.2024.3442728
Carmona Gálvez, J. (2025). Especialización de un modelo LLM en una disciplina específica. https://hdl.handle.net/10630/46582
Carvajal Gutiérrez, H. A., & Lozano Zamora, J. S. (2026). Diseño e implementación de modelos predictivos y analíticos entrenados con datos históricos empresariales para evaluar riesgo, desempeño y anomalías en empresas futuras bajo el enfoque CRISP-ML. https://hdl.handle.net/20.500.12494/64341
Castillo, A., Tabik, S., Pérez, F., Olmos, R., & Herrera, F. (2019). Brightness guided preprocessing for automatic cold steel weapon detection in surveillance videos with deep learning. Neurocomputing, 330, 151-161. https://doi.org/10.1016/j.neucom.2018.10.076
Debnath, R., & Bhowmik, M. K. (2021). A comprehensive survey on computer vision based concepts, methodologies, analysis and applications for automatic gun/knife detection. Journal of Visual Communication and Image Representation, 78, 103165. https://doi.org/10.1016/j.jvcir.2021.103165
Detector gente armada Model > Overview. (s. f.). Roboflow. https://universe.roboflow.com/trabajo-articulo/detector-gente-armada
Hnoohom, N., Chotivatunyu, P., & Jitpattanakul, A. (2022). ACF: An Armed CCTV Footage Dataset for Enhancing Weapon Detection. Sensors, 22(19), 7158. https://doi.org/10.3390/s22197158
Jorge, M., Aragon, J., & Gonzales, C. (2023). Uso de herramientas tecnológicas para la prevención del crimen. Revista Escpogra PNP, 3, 34-43. https://doi.org/10.59956/escpograpnpv3n1.3
juanmolina15. (s. f.). Script/script.py at main · juanmolina15/Script. GitHub. https://github.com/juanmolina15/Script/blob/main/script.py
L.Priyanka, & Sahithi, D. (2025). AUTONOMOUS FIREARMS DETECTION AND THREAT MONITORING USING DEEP VISION MODELS IN SMART CITIES. International Journal of Data Science and IoT Management System, 4(4), 181-186. https://doi.org/10.64751/
Matczak, P., Wójtowicz, A., Dąbrowski, A., & Mączka, K. (2023). Cost-Effectiveness of CCTV Surveillance Systems: Evidence from a Polish City. European Journal on Criminal Policy and Research, 29(4), 555-577. https://doi.org/10.1007/s10610-022-09527-5
Molina, J. (2026). Videos dataset. https://doi.org/10.5281/zenodo.20724901
Salazar Franco, M. A. (2023). Plataforma para desarrollo y evaluación de modelos de reconocimiento de gestos de la mano: Modelo de reconocimiento de 11 gestos de la mano usando deep learning y celdas de memoria desarrollado en Python. https://bibdigital.epn.edu.ec/handle/15000/25083
Studer, S., Bui, T. B., Drescher, C., Hanuschkin, A., Winkler, L., Peters, S., & Müller, K.-R. (2021). Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology. Machine Learning and Knowledge Extraction, 3(2), 392-413. https://doi.org/10.3390/make3020020
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Scientific Journal INGENIAR: Engineering, Technology and Research
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
