Design and Proposal of a Database for Firearms Detection

  • David RomeroEmail author
  • Christian Salamea
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1066)


Closed circuit television (CCTV) surveillance systems that implement monitoring operators have multiple human limitations, these systems usually don’t provide an immediate response in different situations of danger like an armed robbery. To address this security gap, a firearms detection system has been developed through convolutional neural networks (CNNs). For its development a large database of images is necessary. This article presents the creation and characteristics of this database, which is made up of 247,576 images obtained from the web. This article addresses the application of different techniques for the creation of new images from the initial ones to increase the database, obtaining up to 22.7% relative improvement in the accuracy of the network after increasing the database. The database is structured into two classes. The first class is made up of people that have a gun and the second class of people not carrying a gun. The use of this database in the development of the detection system obtained up to 90% in “Precision” and “Recall” metrics in a convolutional neural network configuration based on “VGG net”, through the use of grayscale images.


Convolutional neural network Database Detection Firearm 


  1. 1.
    Deisman, W.: CCTV: Literature review and bibliography. Royal Canadian Mounted Police, Ottawa (2003)Google Scholar
  2. 2.
    Tiwari, R., Verma, G.: A computer vision based framework for visual gun detection using SURF. In: 2015 International Conference on Electrical, Electronics, Signals, Communications and Optimization (EESCO), pp. 1–5, Visakhapatnam, India. IEEE (2015)Google Scholar
  3. 3.
    Tiwari, R., Verma, G.: A computer vision based framework for visual gun detection using Harris interest point detector. Procedia Comput. Sci. 54, 703–712 (2015)CrossRefGoogle Scholar
  4. 4.
    Halima, N., Hosam, O.: Bag of words based surveillance system using support vector machines. Int. J. Secur. Appl. 10(4), 331–346 (2016)Google Scholar
  5. 5.
    Gaga, M., Lach, S., Sieradzki, R.: Automated recognition of firearms in surveillance video. In: 2013 IEEE International Multi-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support (CogSIMA), San Diego, CA, pp. 45–50. IEEE (2013)Google Scholar
  6. 6.
    Olmols, R., Tabik, S., Herrera, F.: Automatic handgun detection alarm in videos using deep learning. Neurocomputing 275, 66–72 (2018)CrossRefGoogle Scholar
  7. 7.
    Olmos, R., Tabik, S., Lamas, A., Pérez-Hernández, F., Herrera, F.: A binocular image fusion approach for minimizing false positives in handgun detection with deep learning. Inf. Fusion 49, 271–280 (2019)CrossRefGoogle Scholar
  8. 8.
    Reolink, Accessed 26 Apr 2019
  9. 9.
    Redmon, J., Farhadi, A.: Yolov3: an incremental improvement. arXiv (2018)Google Scholar
  10. 10.
    Mosselman, F., Weenink, D., Lindegaard, M.: Weapons, body postures, and the quest for dominance in robberies: a qualitative analysis of video footage. J. Res. Crime Delinq. 55(1), 3–26 (2018)CrossRefGoogle Scholar
  11. 11.
    Simonyan, K., Zisserman, A.: Very deep convolutional neural networks for large-scale image recognition. arXiv (2014)Google Scholar
  12. 12.
    Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional neural networks. In: Lecture Notes in Computer Science, vol. 8689, pp. 818–833 (2014)Google Scholar
  13. 13.
    Menditto, A., Patriarca, M., Magnusson, B.: Understanding the meaning of accuracy, trueness and precision. Accredit. Qual. Assur. 12, 45–47 (2006)CrossRefGoogle Scholar
  14. 14.
    Galdi, P., Tagliaferri, R.: Data mining: accuracy and error measures for classification and prediction. Encycl. Bioinform. Comput. Biol. 1, 431–436 (2018)Google Scholar
  15. 15.
    Hammad, A.M., Elhadary, R.S., Elkhateed, A.O.: Multimodal biometric personal identification system based on Iris & Fingerprint. Int. J. Comput. Sci. Commun. Netw. 3, 226–230 (2013)Google Scholar
  16. 16.
    David, J., Goadrich, M.: The relationship between precision-recall and ROC curves. In: Proceedings of the 23rd International Conference on Machine Learning, vol. 06, pp. 233–240 (2006)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Grupo de Investigación en Interacción Robótica y Automática (GIIRA)Universidad Politécnica SalesianaCuencaEcuador
  2. 2.Speech Technology GroupUniversidad Politécnica de Madrid, Ciudad UniversitariaMadridSpain

Personalised recommendations