On the mathematical modelling of visual sensors when computing coverage metrics in camera-based sensing applications

When designing and implementing visual sensor networks, mathematical models are frequently used to support better deployment strategies and quality assessment metrics in applications such as Smart Cities and Industry 4.0. However, the literature in this subject is not uniform and there are more than one way to model a visual sensor. This paper presents existing mathematical models and useful equations that can be employed when computing coverage, organizing and comparing them for different scenarios when analyzing coverage performance and computational costs. Moreover, a simulation scenario is defined considering a typical visual monitoring application in an industrial plant, presenting coverage results for different visual sensing models. Analyzing their mathematical costs, researchers can better decide which model and configuration parameters are more suitable for their needs, benefiting new works in this area.