Descartes Labs Showcases Cloud-Based Geospatial Intelligence Platform at GEOINT Symposium

Descartes Labs Inc. announced that it will introduce its global-scale machine learning platform to the Defense and Intelligence community during the 2017 GEOINT Symposium in San Antonio. The company’s cloud-native platform is engineered to perform large-scale geographic and temporal analysis of remote sensing data, enabling object detection, change forecasting, and advanced intelligence production.

Attendees of GEOINT 2017 can explore the platform at booth #1325 in the Exhibit Hall from June 5–7. In addition, the company is participating in GEOINT Forward with a Lightning Talk on June 4 and hosting a technical training workshop on June 6.

A Cloud Platform Built for Global-Scale Analysis

Founded in 2014 as a spin-off from Los Alamos National Laboratory, Descartes Labs developed its secure, cloud-based infrastructure to apply machine learning and proprietary forecasting models to satellite imagery at planetary scale. The system is designed to support government agencies, commercial enterprises, and academic institutions requiring high-performance geospatial intelligence.

The platform continuously ingests and updates an extensive historical archive of Earth observation imagery sourced from multiple satellite missions, including Landsat 8, MODIS, and Sentinel-1, Sentinel-2, and Sentinel-3. This continuously refreshed archive supports local, regional, and global analysis across both space and time.

According to co-founder and CEO Mark Johnson, the platform is capable of processing petabytes of real-time and archived imagery to detect subtle surface changes—such as agricultural stress—that may carry implications for national security or regional stability. By identifying early signals of change and generating predictive insights, organizations can allocate resources proactively to mitigate potential crises.

Advanced Cloud-Based Search and Analysis

The Decartes Labs platform enables complex queries across its imagery repository based on parameters such as acquisition date, cloud coverage thresholds, sensor type, and spectral bands. Raster processing tools allow users to retrieve selected imagery subsets, perform on-demand band mathematics, and generate time stacks for multi-temporal analysis.

These capabilities support GEOINT professionals who require rapid, scalable access to commercial imagery resources with minimal latency and high analytical precision.

GeoVisual Search Demonstration at GEOINT

At the symposium, Descartes Labs will also demonstrate GeoVisual Search (GVS), an application powered by its geospatial analytics platform. Through a browser-based interface, users can select a visible feature—such as a wind turbine or airport—directly on a map. The system then analyzes satellite and aerial imagery in the cloud to identify visually similar features worldwide or within defined geographic or temporal boundaries.

Search results are rendered instantly on a map interface accompanied by thumbnail previews, enabling rapid pattern recognition and feature comparison across global datasets.

Johnson emphasized that the platform equips GEOINT professionals with tools designed to accelerate scientific workflows across defense, intelligence, homeland security, public safety, and commercial sectors.

Conference Presentations and Training

Descartes Labs is participating in multiple GEOINT 2017 program elements:

• Lightning Talk at GEOINT Forward (June 4, 2:40 pm) – Co-founder and Chief Science Advisor Steve Brumby will present practical applications of the platform for GEOINT workflows. Attendance at GEOINT Forward requires separate registration.
• Workshop Training Session (June 6, 7–9 am, River Level 006C) – Dr. Daniela Moody will lead a session titled “Fast Prototyping of Satellite Imagery Analysis Algorithms Using Open Source Software and Cloud Computing Infrastructures.” Workshop participation is available as an add-on to symposium registration.

About Descartes Labs

Headquartered in Santa Fe, New Mexico, Descartes Labs is a technology firm focused on advancing forecasting science through large-scale geospatial analytics. The company’s platform applies machine learning to massive datasets, including satellite imagery, to improve monitoring, historical analysis, and predictive modeling.

Its mission centers on addressing complex forecasting challenges to better understand global environmental and socioeconomic dynamics, ultimately helping organizations prepare for future scenarios with greater accuracy and insight.