Weapons Detection

This website is organized as follows:

  1. Summary
  2. Handgun dataset for the sliding window approach
  3. Handgun dataset for the region proposals approach
  4. Handgun testset
  5. Paper

1. Summary

This project focuses on the detection of both handguns and knives in videos using Deep Learning techniques based on CNN (Convolutional Neural Networks).

2. Handgun dataset for the sliding window approach

The trainning dataset, appropriate for the classification task, consists of 102 classes with a total of 9261 images. The pistol class has 200.

3.Handgun dataset for the region proposals approach

The trainning dataset, appropriate for the detection task, contains 3000 images of guns with rich context.

4. Handgun testset

Test dataset for both classification and detection. A total of 608 images of which 304 are images of pistols.

5. Papers

Olmos, R., Tabik, S., & Herrera, F. (2018).
Automatic handgun detection alarm in videos using deep learning.
Neurocomputing, 275, 66-72.
Link to the article

Castillo, A., Tabik, S., Pérez, F., Olmos, R., Herrera, F. (2019).
Brightness guided preprocessing for automatic cold steel weapon detection in surveillance videos using deep learning.
Neurocomputing, 330, 151-161.
Link to the article

Olmos, R., Tabik, S., Castillo, A., Pérez, F., Herrera, F. (2019).
A binocular image fusion approach for minimizing false positives in handgun detection with deep learning
Information fusion, in press.
Link to the article

 

Page Maintained by Francisco Pérez y Alberto Castillo