Gunjan Joshi
The University of Tokyo
Gunjan Joshi is a PhD student and MEXT scholar in the department of Electrical Engineering and Information Systems at the University of Tokyo, Japan, under the guidance of Dr. Akira Hirose and Dr. Ryo Natsuaki. She holds an M.Eng in Electrical Engineering and Information Systems from the University of Tokyo, Japan (2021), and a B.Eng in Electronics and Communication Engineering from Visvesvaraya Technological University, India (2016). In addition to her academic pursuits, Gunjan has gained experience as a guest scientist at the German Aerospace Center and has worked as a working professional in India.
Gunjan's research interests research interests focus on the integration of synthetic aperture radar and optical remote sensing data using explainable neural networks. Her primary focus is on utilizing this fusion technique for earthquake damage assessment and glacier mapping. In addition to her research, Gunjan finds solace in reading and exploring the world through her camera.
One of the most effective ways for assessing the impacts of climate change is through the observation of mountain glaciers since these serve as reliable indicators of the changes in environmental dynamics. By conducting detailed and accurate classifications of glaciers, we can gain valuable insights into their overall health. Such information is instrumental in managing water resources effectively and implementing proactive measures to prevent devastating floods triggered by glacial lake outbursts. However, conducting on-site observations of glaciers presents numerous challenges due to their steep terrain and remote locations.
The recent increase in the number of earth observation missions has provided us with powerful tools to overcome these limitations by remotely observing the rugged glacierized terrains. These missions offer high-resolution satellite imagery and data, enabling us to gain detailed insights into the dynamic changes occurring on the Earth's surface. Moreover, these Earth observation missions vary throughout the electromagnetic spectrum and include both active and passive sensors. The combined use of active and passive sensors in Earth observation missions can facilitate in a comprehensive understanding of glaciated areas.
Our work focuses on fusing the Synthetic Aperture Radar sensor data and the optical sensors data using explainable AI (XAI) which can help us in understanding the contributions of the various sensors in accessing climate change.
