Security in smart cities using YOLOv8 to detect lethal weapons

Abstract

The increase in the illegal use of lethal weapons at a global level has reached
worrying figures, resulting in an increase in assaults and armed robberies. Based on the above, closed circuit television (CCTV) surveillance systems emerge as an alternative solution. Therefore, the use of artificial intelligence is explored in order to detect the presence of lethal weapons in images accurately. In this study, a convolutional neural network model YOLOv8 is trained. A database including 4104 images with the presence of lethal weapons is generated. The Google Colab platform is used for the training phase, since it provides us with a free graphic processing unit (GPU), and the YOLOv8x and YOLOv8n models are used for comparison. The results indicate an advantage when using the YOLOv8 models, and when comparing them with similar models already proposed in the studied literature, we can conclude that our model stands out with an accuracy of 89.56% in the detection of lethal weapons. In conclusion, we were able to obtain a model capable of detecting lethal weapons in CCTV images, in addition to being able to be used in applications that require real-time detection.