Our solution aims to address actual and perceived safety in urban settings deemed unsafe, especially at nighttime. We aim to employ state-of-the-art AI CCTV surveillance solutions, mobile (using security drones) or static, in combination with real-time human activity recognition (HAR) algorithms to enable accurate and quick response from the local police. Depending on the urban setting, the solution can be adjusted to the local conditions. For example, the City of Amsterdam is especially concerned about the safety on long, desolate bicycle paths on the city outskirts that connect event venues with residential areas. These paths typically have one entry and one exit point with limited or no surveillance. In this situation, we propose setting up a sensor-activated system that would launch security drones with AI-powered CCTV surveillance capabilities if a person using the path doesn’t reach its endpoint or control point within an estimated time period.
Read more from the attachment.
