The Pandemic View from Space with Geo-spatial Data

A first Webinar of Geo-informatics series titled “Application of Artificial Intelligence on Geospatial Data” was organized by the Department of Information Technology of Maulana Abul Kalam Azad University of Technology, West Bengal (MAKAUT, WB) on 30th May, 2020. It was an effort to focus on Geospatial Artificial Intelligence, the intersection of geospatial data and artificial intelligence with special emphasis on its applications on monitoring the Pandemic View from Space. The Webinar was coordinated by Dr. Somdatta Chakravortty, Head of Information Technology, MAKAUT, WB.


The participants were not only from the State of West Bengal but also from other States in India and countries such as Spain, United Kingdom, United States, Iran, Pakistan and Germany.

The event was inaugurated by the Hon’ble Vice Chancellor of the University, Prof.(Dr.) Saikat Maitra. In his address he elaborately discussed about the applications of Geospatial data in Environment, Society and in our day to day life. It was a privilege to have two very eminent researchers and experts in the field of Geo-informatics: Prof. Antonio Plaza, Professor in the Department of Technology of Computers and Communications, University of Extremedura, Spain and Prof. Sucharita Gopal Professor in the Department of Earth and Environment & Center for Remote Sensing, Boston University, Boston.
Prof. Gopal deliberated on the subject - Spatializing the Incidence and Mortality of Coronavirus (COVID-19) Pandemic using AI and Big Data Geo-informatics in which she talked about the COVID-19 Pandemic Emerging Research, GIS and Spatial Analysis and AI-Big Data Informatics. She discussed the role of GIS and Big Data in the study of COVID-19 like visualization of disease information, tracking of confirmed cases, AI based prediction and Spatial Models, spatial behavior and decision making like social distancing etc. She pointed out the application of remote sensing sensors and drones for COVID-19, how satellite sensors can detect the change in air pollution level pre and post lock down, Prof. Plaza deliberated on the subject - "Recent Advances in Spectral Unmixing of Remotely Sensed Hyperspectral Data" in which he elaborated on Spectral unmixing as an important technique for the analysis of remotely sensed hyperspectral data.

In this talk, Prof. Plaza discussed several models for spectral unmixing (including linear and nonlinear models) as well as the most recent trends and techniques for unmixing hyperspectral data. This included techniques that assume the presence of pure pixels in the data, techniques that generate virtual endmembers (in order to unmix images with no pure signatures) and deep learning-based approaches for spectral unmixing. He also discussed computationally efficient implementations of hyperspectral unmixing techniques in high performance computing architectures such as graphics processing units (GPUs). Practical examples were provided with different data sets collected on different locations around the world. A very fruitful discussion session followed the deliberations and interesting questions were posted by the participants. A discussion blog has been created in which participants are positing their queries even after the webinar, thus taking the discussion forward. The webinar ended successfully with a short quiz for the participants after which the Certificates were distributed. Geospatial AI, the intersection of geospatial data and artificial intelligence, is thus the new frontier of technological innovation that promises to transform entire business industries. For decades, geographic information systems (GIS) have been used widely to present a view of our world based on geographic and geospatial data. This is a significant revolution, expanding the use of geographic data in a way that promises to disrupt entire industries.