Data-driven visual analytics of Human Mobility data and green cover using Image Processing for Smart Cities

This research paper presents an approach to data-driven visual analytics of human mobility data using Kernel Density Estimation visualized through heatmaps, highlighting the need for exploration of forecasting methods and intuitive visualizations using the ARIMA model. A specific geographic area is chosen for the demonstration of the scope of the proposed system. The system is a web application developed using Streamlit, an open-source python framework. To effectively implement the smart city concept, it is crucial to integrate diverse IoT systems, data sources, data streams, and analytical tools into a unified and seamless platform that facilitates the collection, analysis and presentation of information related to urban systems and subsystems. We propose taking into consideration several attributes when analyzing human mobility patterns, such as vehicle/traffic density, modes of transport, transport data, demographics, latitude, and longitude. Additionally, the system utilizes image processing as an efficient method for calculating urban green cover using morphological operations that are computationally cheaper and easy to use as compared to traditional surveys that are time and resource intensive. This information is used to develop smart plans to sustain or increase green cover in select areas, leading to the creation of sustainable and green smart cities.