Edge to Cloud End to End Solution of Visual Based Gun Detection
Qi Yao, Weijun Tan, Jingfeng Liu, Delong Qi · Journal of Physics Conference Series · 2020
Abstract There are over a half million cases of gun violence globally every year. State of the art video surveillance requires human to watch the video footage to monitor gun violence. In this paper, we present an AI based edge to cloud solution appropriate for smart surveillance and control. Firstly, we build the key training dataset by harvesting the publicly available videos and movies. Then we design three deep convolution neural networks (CNNs) models, there are two models running on the edge device, which is a video camera, and the third one runs on the cloud. By distributing these three models into edge and cloud, we are able to achieve a good gun detection system that can be deployed in real life.