Gun Detection with Model and Type Recognition using Haar Cascade Classifier
Aarchi Jain, Aishwarya, Gaurav Garg · 2020 Third International Conference on Smart Systems and Inventive Technology (ICSSIT) · 2020
The crime percentages brought about by weapons are very concerning in numerous places on the planet, particularly in nations where the ownership of weapons is legitimate or was lawful for a while. In this advanced time of observation and security the quantity of Closed-Circuit Television (CCTV) conveyed out in the open and private places, for example, Cinemas, Malls has expanded exponentially. As of now, there are a huge number of CCTV cameras in operation. In this way, the expanding density of investigation camera recording makes it a challenge for a human administrator to inspect, analyze whether a conceivably risky circumstance is going to occur. Object detection dependent on digital image processing on weapons is very important for establishing security systems. This paper presents a real-time framework for gun detection with model and type recognition. The method uses real-time video as an input and the method used is Haar Cascade Classifier, Open-CV library. This method is mainly designed for security and safety management purposes. This method was built to identify and classify guns in a video and a high accuracy rate is obtained which reaches 95% in Submachine Guns. Software required for the framework is effectively accessible because of which it turns out to be a more affordable framework.