Deciphering Earth Intelligence

The distance between pixel centers measured on the ground. A GSD of 30cm means one pixel represents 30cm of the earth's surface.

An active radar system that can 'see' through clouds, smoke, and darkness by emitting its own energy pulses and measuring the return.

A series of aerial photos stitched together and geometrically corrected to be uniform, removing distortion so it can be used as an accurate map.

Imagery captured at an angle (usually 40–45 degrees) rather than straight down (nadir), allowing views of building facades.

Imagery captured looking straight down (90 degrees) at the ground.

Sensors that capture light across hundreds of narrow spectral bands beyond the visible spectrum to identify materials by their chemical signature.

The season of capture. 'Leaf-Off' is winter/spring (no leaves); 'Leaf-On' is summer (full canopy).

Light Detection and Ranging. A remote sensing method using pulsed lasers to measure variable distances to the Earth, creating precise 3D maps.

A rendering technique representing 3D scenes as millions of overlapping 3D 'blobs' (Gaussians) rather than geometric meshes.

DSM (Digital Surface Model) captures everything including trees/cars. DTM (Digital Terrain Model) captures only the bare earth.

A set of data points in a 3D coordinate system (X, Y, Z), usually generated by LiDAR or photogrammetry.

A wireframe structure draped with imagery to create a photorealistic 3D model of a city or object.

Interferometric Synthetic Aperture Radar. A technique for mapping ground deformation using radar images collected from orbiting satellites.

Geospatial Artificial Intelligence. The application of computer vision and machine learning to spatial data to extract insights.

AI models that don't just analyze maps but generate *new* spatial data, such as synthetic imagery, in-painting clouds, or simulation.

Massive AI models pre-trained on petabytes of satellite imagery that understand Earth features generally and can be fine-tuned.

Converting locations and attributes into numerical vectors to capture semantic meaning (e.g., 'suburban vibe').

An AI process that classifies every pixel in an image into a category (e.g., 'This pixel is roof,' 'This pixel is tree').

Satellite data processed to a minimum set of requirements allowing for immediate analysis without additional user effort.

Algorithmic comparison of two images of the same location from different dates to highlight what has changed.

How frequently a satellite or aircraft images the same location (Daily, Monthly, Annually).

How close a feature on the map is to its actual location in the real world.

A file format allowing users to stream just the part of the image they need without downloading the entire file.

Raster is pixels (the picture); Vector is lines/polygons (the data derived from the picture).

An API that allows software to programmatically request a satellite to capture a specific location at a specific time.

Total Addressable Market, Serviceable Available Market, and Serviceable Obtainable Market.