As part of our dedication to reducing complexity in the earth observation space, EXPLORE delivers a standard product regardless of the underlying sensor. This ensures that your process can expect a consistent product every Order.

Naming Scheme

All SkyWatch products follow a standardized naming scheme. Here is an example:

SKYWATCH_3S_PS_20200422T1711_Tile_0_0_Cj8RAT52

SKYWATCH - All products start with "SKYWATCH", so they can be identified from any other products you might be using.

Satellite Code - This product has "3S" as the satellite code, indicating that it is a TripleSat product. Here is a table of the most common satellite codes:

Code	Satellite
3S	TripleSat
SS	SkySat
PL	Pleiades
SP	SPOT
S2	Sentinel-2
PNEO	Pleiades-Neo

Processing - The letters "PS" on this image indicate that it has been pansharpened. Currently, pansharpened is the only option for display.
Timestamp - The next set of numbers and letters indicate the time at which the image was collected, in this case on 2020-04-22 at 17:11 UTC.
Tile Numbers - This image is Tile 0,0 of the full image. In cases where the imagery is too large to be manageable (~>50km2), the imagery is tiled before delivery. These tiles are numbered from top left, with the first number indicating the row, and the second indicating the column.
ID - This is an 8-character random ID generated for each output.

Projection

All outputs are provided in EPSG 4326 projection (WGS84 Geographic). This is the default coordinate system for most GIS and other software, and is readily compatible with many basemaps and open data products. Wikipedia Article about WGS84.

Files

There are two main formats that image files are delivered with through EXPLORE - GeoTIFF and PNG. In both cases, if the image would be more than 10,000 pixels in any dimension (e.g. a 5km-wide 50cm image), the results are tiled seamlessly and are delivered together.

GeoTIFF

Georeferenced Tagged Image File Format (GeoTIFF) is an open file format and de facto standard based on the TIFF format and is used as an interchange format for georeferenced raster imagery. GeoTIFF is widely by remote-sensing data experts and scientists.

SkyWatch always delivers a non-BIGTIFF, unsigned INT16, uncompressed GeoTIFF.

For a more complete overview of the GeoTIFF format, we recommend this well-documented article from GeoSpatial World .

How to open GeoTIFF files in SNAP

The European Space Agency offers a free file viewer for remote-sensed data called SNAP. It is available for download here.

How to open GeoTIFF files in Python

OpenEV in the GDAL suite will also display GeoTIFF. GDAL is very helpful in general for GeoTIFF but a bit difficult to install.

For example, the following code reads you the size of the 16-band images:

import gdal
from gdalconst import * img_filename_16bandA = 'data/16band/6100_1_3_A.tif'
img_filename_16bandM = 'data/16band/6100_1_3_M.tif'
img_filename_16bandP = 'data/16band/6100_1_3_P.tif' datasetA = gdal.Open(img_filename_16bandA, GA_ReadOnly )
datasetM = gdal.Open(img_filename_16bandM, GA_ReadOnly )
datasetP = gdal.Open(img_filename_16bandP, GA_ReadOnly ) print 'Size is ',datasetA.RasterXSize,'x',datasetA.RasterYSize, \
 'x',datasetA.RasterCount
print 'Size is ',datasetM.RasterXSize,'x',datasetM.RasterYSize, \
 'x',datasetM.RasterCount
print 'Size is ',datasetP.RasterXSize,'x',datasetP.RasterYSize, \
 'x',datasetP.RasterCount

Outputs:

Size is 136 x 134 x 8
Size is 848 x 837 x 8
Size is 3391 x 3348 x 1

Or you could use tifffile:

import tifffile as tiff
P = tiff.imread(img_filename_16bandP) 
tiff.imshow(P)

How to open GeoTIFF files in R

devtools::install_github("ropensci/geojsonio")
library("geojsonio")
install.packages("rgdal", type = "source")
install.packages("rgeos", type = "source")
library("rgdal")
library("rgeos")
library(ggplot2)
grid_6010_4_4 <-geojson_read("./data/train_geojson/train_geojson/6010_4_4/Grid_6010.geojson", method = local, what= 'sp')
plot(grid_6010_4_4)
ggplot(grid_6010_4_4, aes(long, lat, group = group)) + geom_polygon()

To access images:

library(raster)
raster_6044_4_4 <- raster("./data/three_band/6040_4_4.tif")
plot(raster_6044_4_4)
library(rgdal)
gdal_6044_4_4 <- readGDAL(paste0("./data/three_band/", '6040_4_4', ".tif"))
plot(gdal_6044_4_4)

PNG Files

The visual output for any Order is provided as a rendered PNG file. In cases of RGB, RGB+NIR, or All Optical Bands, this image is made up of the Red, Green, and Blue bands of the image, and is thus always a True Colour image. For processed imagery such as NDVI or NDWI2, this image is a rendered version of that index.