Processing Levels
When placing an order for satellite imagery, users can select the desired processing level to best suit their needs.
Level-2A Surface Reflectance
Level-2A products are surface reflectance imagery that have undergone atmospheric correction and reprojection to a projected map coordinate system. These products are designed for users who don't have unique atmospheric correction requirements and prefer ready-to-use data for analysis.
Key Features
Pixel Units and Scale
- Units for surface reflectance are unitless, represented as an empty string (
""
) in the metadata. - Delivered as a raster dataset with 16-bit floating point (uint16) data type.
- Pixel values are scaled integers with a scale factor (i.e., 0.0001) to convert to float reflectance. This approach optimizes file size storage while maintaining precision.
- Example:
{
"nodata": 0,
"sampling": "area",
"data_type": "uint16",
"unit": "",
"scale": 0.0001,
"offset": 0
}
Consistent Reprojection and Pixel Size
- Reprojected to Universal Transverse Mercator (UTM) / WGS84, with the UTM zone and hemisphere determined by the center pixel of the image scene.
- Example:
"proj:epsg": 32611
- Example:
- Output pixel size is 5 m x 5 m in the projected coordinate system.
- All downstream products (imagery, masks, etc.) use the same projection and pixel size for consistency.
Geometric accuracy
- Wyvern's Level-2A products have the same geometric accuracy as Level-1B and are not orthorectified. We plan on improving the geometric accuracy in future releases.
Level-1B Top-of-Atmosphere Radiance
Level-1B products are hyperspectral imagery that have been geometrically corrected to a map-projected north-up Geographic WGS84 (EPSG:4326) coordinate system and radiometrically corrected to top-of-atmosphere radiance. These products are suitable for advanced analytical techniques and industry-standard radiometric and geometric processing.
Key Features
Geometric Correction and Coordinate System
- Imagery is georeferenced to Geographic WGS84 (EPSG:4326).
- Example:
"proj:epsg": 4326
- Example:
- The spatial X & Y cell size of each pixel is defined in angular longitude & latitude degrees.
- Pixel cell size is set to the angular degree equivalents of 5 m x 5 m at the latitude of the image center.
- Pixel cell size in will be non-square and vary based on latitude (e.g., at 40°N: Lon=0.0000586° x Lat=0.0000450°).
Radiometric Correction and Pixel Units
- Delivered as a raster dataset with 32-bit floating point (float32) data type.
- Pixel values represent top-of-atmosphere (TOA) radiance.
- Units are 'W / (m² * sr * µm)' (watts per square meter per steradian per micrometer).
- No scaling factor is required; values are in physical units.
- Example:
{
"nodata": -9999,
"sampling": "area",
"data_type": "float32",
"unit": "W⋅sr-1·m-2·µm-1",
"scale": 1.0,
"offset": 0
}
Metadata and Analytical Applications
- Metadata enables radiometric & geometric data processing and advanced analytics (deep learning, machine learning, spectral indices, classification, anomaly detection, material identification, sub-pixel mixture analysis, spectral target detection).