Pistol Weapon Detection

Insecurity is a problem that affects all cities around the world to a greater or lesser extent, and some of them make use of video surveillance to combat it, setting up monitoring centres with hundreds of cameras. For the most part, these centres are staffed by personnel responsible for observation and incident response. The advancement of technology in the market offers the possibility to optimise and add value to these processes. This paper will present an approach to a possible solution through video analysis and real-time weapon detection using technologies such as OpenCV, Python and Machine Learning among others. A description of the current context of these technologies in the country and the stages of construction of a prototype capable of detecting firearms through local and online cameras will be described.