A proof of concept for providing traffic data by AI based computer vision as a basis for smarter industrial areas
Abdullah Shams, André Schekelmann, Wilhelm Mülder · Procedia Computer Science · 2022
Algorithms and data serve as the foundation for any SMART approach, whether smart cities or smart urban environments. Availability of near real-time data for such smart applications in a wider geographic area is an expensive undertaking on both the technical and financial front. We introduce an approach to reduce technical and financial barriers to enable near real-time data collection. Additionally, our approach can be scaled with no significant technical depth. Our proposed approach was applied to meet the challenge of underutilized collaboration and data in a semi-closed harbor situated near Neuss and Düsseldorf, Germany. Specifically, we targeted the traffic problems causing adverse economic and environmental effects within the harbor area. By utilizing open-source projects and economically viable hardware, we offer a dashboard with information sourced from multiple data streams in an ad-hoc manner. We leverage crowd solving by providing ready-made solutions that each partner can acquire without technical expertise and in return provide near-real-time traffic data from which all parties can benefit. The work is part of the research project Logistics.NRW funded by Europäischer Fonds für regionale Entwicklung, Leitmarkt Mobilität & Logistik.