Abstract
The digitalization of public transport offers new opportunities to improve service efficiency, particularly in managing peak hours and reducing delays. A key contributor to transport inefficiency is crowd congestion at bus stops. Monitoring bus stops enables a better understanding of passenger flow and waiting times, supporting more effective planning and faster real-time response. However, current data on activity at these locations is scarce due to privacy concerns and the practical difficulties of collecting data in urban settings. To address this gap, this paper proposes a solution to monitor bus stops by equipping them with depth cameras, which capture anonymized data by only recording distance and leveraging computer vision techniques to perform people and bus detection. Results show the viability of the proposed solution to optimize bus networks and its potential replicability in other urban environments.