Analyse thermique d'images satellite / Thermal Satellite Image Analysis

Constellr fournit des donnees globales de temperature de surface pour l'agriculture durable et la surveillance des ressources. En s'appuyant sur l'imagerie thermique satellite, l'entreprise produit des informations fines et regulieres sur la sante des cultures et le stress hydrique, afin d'aider agriculteurs et gestionnaires de terrain a optimiser l'irrigation, les rendements et l'usage des ressources.

Constellr provides global land surface temperature data for sustainable agriculture and resource monitoring. Utilizing thermal satellite technology, they offer precise, daily insights into plant health and water stress. This enables farmers and land managers to optimize irrigation, improve yields, and reduce resource consumption. Constellr's data supports proactive decision-making, aiming to enhance food security and environmental sustainability through advanced thermal imaging.

Deeper Analytics a joue un role cle dans l'amelioration des analyses de temperature de surface de Constellr en traitant plusieurs problemes d'integration de donnees. Nous avons developpe des algorithmes capables d'harmoniser les differences de resolution spatiale entre Landsat, MODIS et Sentinel pour produire des jeux de donnees unifies et a plus haute resolution. Nous avons egalement aligne les resolutions temporelles afin de rendre comparables et combinables des images acquises a des instants differents. Enfin, nous avons mis en place une approche de comblement des zones masquees par les nuages en fusionnant plusieurs sources spatio-temporelles, ce qui a nettement renforce la continuite, la precision et la fiabilite des cartes thermiques.

Deeper Analytics played a crucial role in enhancing Constellr's land surface temperature analysis by addressing critical data integration challenges. We developed sophisticated algorithms to harmonize the varying spatial resolutions of satellite data from sources like Landsat, MODIS, and Sentinel. This upsampling ensured a unified, high-resolution dataset. Furthermore, we implemented techniques to align temporal resolutions, enabling seamless comparison and combination of images captured at different timestamps. Finally, another algorithm effectively filled data gaps caused by cloud cover by combining multiple data sources with their respective spatio-temporal resolution , ensuring continuous and comprehensive coverage. These contributions significantly improved the accuracy and reliability of Constellr's thermal data, empowering users with more precise insights.