European cities face increasing difficulties in achieving Vision Zero and ensuring the safety of vulnerable road users (VRUs). The rapid growth of micro-mobility (e-scooters, bicycles) and its interaction with private and delivery vehicles has led to a rise in urban accidents, especially during nighttime and low-light conditions such as dusk and dawn. While human behaviour plays a role in some incidents, a significant proportion could be prevented through improved infrastructure, including safer separation of traffic flows, clearer signage and adequate illumination of pedestrian crossings.

However, evaluating nighttime visibility across entire urban networks remains a major limitation. Current methods rely on visual inspections and manual luxmeter measurements, which are time-consuming, inconsistent and not scalable to city-wide deployment. Repeating these measurements across all intersections and under varying weather and seasonal conditions is practically unfeasible.

In cities such as Las Rozas de Madrid, Brașov and Vitoria-Gasteiz, lighting conditions are highly heterogeneous. Vegetation, construction works and even decorative lighting can reduce visibility or create a false sense of safety. As a result, municipalities lack reliable tools to systematically identify hazardous locations and assess lighting interventions. This creates a clear need for an automated, objective and scalable solution to continuously evaluate night-time visibility and support data-driven safety.