Detecção de Armas de Fogo Usando DETR com Múltiplas Redes Neurais Autocoodenadas

Abstract

This paper presents a new strategy that uses multiple neural networks in conjunction with the DEtection TRansformer (DETR) network to detect firearms in surveillance images. The continuous growth of gun violence around the world has forced many agencies, companies and consumers to deploy closed circuit TV (CCTV) surveillance cameras in an attempt to combat this epidemic. However, the large number of cameras to be observed leads to an overload of CCTV operators, generating fatigue and stress and, consequently, a loss of surveillance efficiency. The strategy developed in this work presents a methodology that promotes collaboration and self- coordination of networks in the full connected layers of DETR through the technique of multiple self-coordinating artificial neural networks (MANN), which does not require the use of a coordinator. This self-coordination consists of training the networks one after the other and integrating their outputs without an extra element called a coordinator. In order to improve remote surveillance, the insertion of deep neural networks has proven to be efficient in detecting and identifying objects in videos, and in various situations has produced more accurate and consistent results than human beings. To the best of our knowledge, this work is the first to introduce MANN into transform-based object detection architectures.