Title:  Fusion of Remote Sensing with Deep Learning for Earth Observation

Speaker:  Dr. Indrajit Kalita, Boston University

ABSTRACT:

This talk will explore the fusion of remote sensing and Convolutional Neural Networks (CNNs) for advanced Earth observation. Beginning with an overview of remote sensing imagery, it will delve into the diverse applications and image characteristics. Transitioning to deep learning, the focus will shift to CNNs, detailing their significance in extracting insights from remote sensing data. Notable works employing Deep CNNs (DCNNs) will be highlighted, showcasing their effectiveness in addressing Earth observation challenges. Attendees will gain practical insights into leveraging these technologies for improved monitoring of our planet.

BIO:

Postdoctoral Researcher (FORMES research group), Computing and Data Sciences (CDS), Boston University, MA, USA,Dr. Indrajit Kalita is a Postdoctoral researcher affiliated with Boston University's Faculty of Computing & Data Sciences (CDS), where he passionately contributes to advancing environmental science through the application of advanced technology. His current focus involves the development of precise rainfall prediction models utilizing deep learning and computer vision techniques. Prior to joining Boston University, Dr. Kalita held a position as a postdoctoral researcher at CYENS - Centre of Excellence, Nicosia in Cyprus. In this role, he explored Earth's behavior over time and its correlation with climate change. Dr. Kalita's academic journey commenced at the Indian Institute of Information Technology Guwahati, an esteemed Indian Government Institute of National Importance. His doctoral dissertation focused on domain adaptation (DA) in land cover classification, involving extensive analysis of remotely sensed images. His research focuses on exploring shifts in the environment across both time and space through remote sensing data.