The satellite systems we use to capture, analyze, and distribute data about the Earth are improving every day, creating bold new opportunities for impact in how we manage our crops, protect our forests, plan our infrastructure, and more. From capturing imagery and developing processing pipelines to developing data products thar are easily viewable across any screen, we help intepret your issue from the sky.
A typical serviced satellite image (bought or downloaded) includes several processing steps. From the raw image (Level 0), data are calibrated into units of physical reflectance (called Level 1 processing), and the image is geolocated and ortho-rectified following an elevation model of the terrain or Ground Control Points (GCP) of a certain Accuracy, under a geodetic reference frame (e.g. Mercator). Over large areas, several captures may need to be stitched together. T his product is usually referred as Level 2 or similar. In many cases, the end user will only see the final image, either in a report or as an interactive map on the web, such as the maps in this report.
Satellites capture light in wavelengths that are outside the range of human vision, such as infrared or ultraviolet light, that can help to understand the surface characteristics of the reflecting substances. Different objects reflect these light frequencies in different ways, and common satellite analysis techniques combine human-visible and invisible images (called “spectral bands”) to characterize their subjects. For example, spectral bands covering frequencies in the middle-infrared wavelength regions are responsive to moisture content in vegetation, forest canopy and soil, while near-infrared wavelength regions tend to emphasize vegetation health and— at a coarser scale— overall biomass.Taken together, the spectral response of each band creantes a unique signature, known as a spectral curve, that can help communicate information about conditions within a single type of land cover. This includes calculating well-known indices such as NDVI to assess vegetation health or false coloring to isolate urbanized areas.
Enterprise grade services from Digital Globe, BlackBridge,
GeoEye, Planet offer faster access to new imagery,
archival images and derived data products. Publicly available sources from LandSat and Sentinel also offer archival images,
but in raw format, requiring image processing using tools such as GDAL and ImageMagick.
Conventional models of identifying, buying and delivering imagery are giving way to new cloud based selection and delivery models. Users can search through archival content from multiple suppliers to find the best suited imagery, buy the imagery through an enterprise subscription, and have it delivered online. As the value of raw pixels declines, there is greater innovation and competition to provide data products derived from imagery. At Vizonomy, we work with various partners to ensure that all our projects use the most optimal image possible, taking into account its cost, resolution, and timestamp.
|Satellite||Open Data||Spacial resolution(m)||Resivit rate (days)||Cost ($ per Km2)|
|Airbus SPOT 6/7||No||1.5||1||5.15|
|EU Sentinel 2||Yes||10||5||0|
|NASA/USGS LandSat 1-3||Yes||60||18||0|
|NASA/USGS LandSat 4-5||Yes||60||18||0|
|NASA/USGS LandSat 7-8||Yes||15||16||0|
Spatial resolution refers to the size of one pixel on the ground. For example 15 meters means that one pixel
on the image corresponds to a square of 15 by 15 meters on the ground. This is also sometimes referred to as
Ground Sample Distance (GSD). Temporal resolution refers to the how often data of the same area is collected.
This is typically referred to as Revisit Time.
Freely available imagery (e.g., Landsat, Sentinel, MODIS) tends to either have a revisit time measured in days (1-4 days) with resolution in the hundreds of meters (300m-500m), or a revisit time measured in weeks (10-20 days) with resolutions in the tens of meters (10m-30m).
High resolution commercial imagery is available up to .3m resolution, with revisit times varying quite a bit. Some sensors are tasked, or pointed to collect specific areas rather than always just collecting the area directly below. As a result, some areas may not be covered at all by tasked satellites. While there is still a premium for the highest resolution imagery (0.50m), medium to low resolution is suitable for many applications, and increasingly affordable or available at no cost.
Spectral bands can be combined mathematically
to emphasize a particular set of characteristics:
vegetation, water, urbanized areas, forest fires, and more.
In addition to false color composition, spectral bands can be combined mathematically to emphasize a particular set of characteristics. These techniques may draw from all relevant bands, rather than the three band limit set by human vision, to draw out very specific characteristics.
Synthesizing data from multiple spectral bands, through ratio or coefficient-based transformations, can produce indices that can be used to compare every point in an image on the same scale. The most common indices, such as Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), measure vegetation health and distinguish between types of plants, but other indices have been developed to measure burn severity, geologic qualities such as the presence of certain minerals, water turbidity, mud, snow, and more.
Urbanization is arguably one of the most significant landuse/landcover change occurring in the world today. Conversion to urban land - losses of agricultural and forested lands, coupled with increasing impervious surface cover - has direct effects on natural temperature regulation and alters the hydrologic cycle. Such changes to the hydrologic cycle represents the most significant urban water quality issue present today because stormwater runoff from impervious surfaces creates water quality problems including higher water temperatures and elevated levels of contaminants in surface waters. A simple form of satellite or drone image interpretation involves assigning invisible light bands to Red, Green, and Blue channels to create a false color image, in order to highlight hidden characteristics related to those bands. False color images can distinguish muddy water from muddy land, pinpoint fires, or clearly show the extent of a growing city.
Advanced analytical techniques go beyond inter-band math, and may classify data into predefined fuzzy or map-like categories using machine learning techniques, compare imagery taken at different times to detect change, compar imagery taken at different angles to estimate elevation, or combine multiple source datasets to exploit the best qualities of each.
Elevation and Surface models are constructed by calculating the offset in radar or image-based data acquired from different angles.
Analysis of imagery from different time periods to detect and understand changes. Examples include urban growth, deforestation, ice melting, and landslide detection.
Analysis of imagery to extract and identify buildings, roads, land covers, water extents, forests, etc. Once extracted, these items can be sorted, counted, and analyzed.