Hasimoto-Beltran, Rogelio, Odin F. Eufracio-Vazquez, and Berenice Calderon-Damian. "Deep Neural Networks for Passengers’ Density Estimation and Face Mask Detection for COVID-19 in Public Transportation Services." ICT Applications for Smart Cities. Springer, Cham, 2022. 21-35.
Public Transportation Networks (PTN) are considered as one of the essential commuting systems worldwide. Under good planification strategies, PTN improves population mobility, reduces environmental pollution, traffic congestion, and commuting time. Optimal management, control, and tracking of public transportation systems are desirable for improving service quality and users’ wellness. This work proposes a management and control system for public transportation based on passengers’ density estimation. Our final goal is to develop a communication network architecture and a Deep Neural Network for face detection to quantify the density and number of people being transported as a function of time. This collected information (density and number of passengers) is sent to public transport administrators (municipal authorities and bus owners) to take actions for adding/removing busses to/from specific route circuits and improve the users’ commuting quality of service (that is, reducing traveling times, waiting time on the bus stop, and not oversaturated units). Additionally, to control the spread of COVID-19 on PTN, our algorithm can detect passengers with or without a face mask. We provide an in-depth description of our proposed Deep Neural Network and its implementation in Python programming language.