top of page
Advancing satellite image processing for natural resources management
Browse our projects:
Atmospheric correction of satellite images
This research project will develop and validate a harmonized Landsat-Sentinel-2 framework for water reflectance retrieval in optically complex waters. This sensor-generic framework will solve complex radiative transfer calculations to derive water reflectance products across Mississippi inland and coastal waters.
Deep learning for land cover
Land Resources Essential Information System (LREIS) provides essential information for assessing and managing land and natural resources. In particular, land cover is one of the most important spatial representations of physical coverage of the Earth's surface to create thematic maps from remotely sensed imagery at a range of spatial and temporal scales. A scalable, AI-driven LREIS is critical to leverage the next generation of very detailed land cover maps in highly complex landscape across the U.S.
Agricultural soil management
Soil sampling is a key practice for gathering field-specific data that supports decisions in soil health assessment and management, and site-specific information is difficult and costly to acquire, and spatially limited over time, especially for small and medium-sized farmers. This research project will develop and validate a new operational large-area Satellite-based Soil Sampling Design Tool (S3DTool) using historical crop variability analysis and auxiliary data at field-scale.
HPC Agriculture Research
The overall objective of the proposed partnership between Mississippi State University (MSU) and Agricultural Research Service (ARS) focuses on the development of new cross-disciplinary analytical approaches for agricultural big data encompassing geospatial, epidemiological, and computational tools that enable data-driven decision-making for agricultural problems that affect field to table.
Satellite virtual buoy for water quality
This project will develop a new support tool (Satellite-based Virtual Buoy Observation Network) to monitor and characterize water quality conditions and trends in oyster reef sites across Mississippi Sound, delivering the results in a data portal for oyster stakeholders, marine agencies, and community organizations. Specifically, this project will integrate satellite and in situ data to build a virtual monitoring network across Mississippi Sound and quantify the impacts of sediment-rich freshwater dynamic (i.e., timing, duration, frequency) caused by Bonnet Carré Spillway openings.
This research supports the Microsoft FarmVibes.AI project in developing an open-source deep learning tool for satellite data analysis in the agricultural context. Crop-type mapping generates baseline information to build other digital technologies in the FarmVibes.AI project, and innovative solutions with AI models for crop type mapping using satellite data will be explored.
bottom of page