Planet FAQs

We currently make two classes of PlanetScope imagery available to our customers, standard and test.

To qualify for "standard" image quality an image must meet the following criteria: sun altitude greater than or equal to 10 degrees, off nadir view angle less than 20 degrees, and saturated pixels fewer than 20%. If the image does not meet these criteria it is considered "test" quality.

Each individual PlanetScope scenes metadata contains animage_statistics.image_quality property that indicates the quality of that particular image.

Planet is also continuously improving the performance of the telescopes that we put into space, and there is a significant improvement in the quality of newer images. You might see reference to PlanetScope 1 (PS1) and PlanetScope 2 (PS2) imagery within PlanetScope scenes metadata instrumentpropertywhichindicates the generation of telescope that was used to capture an image. All new Planet satellites will host a PS2 sensor.

Once an order has been placed in Planet Explorer, it is typically ready for download in five to ten minutes. This can vary depending on the size of the order.

However, sometimes orders may get stuck in a processing loop - if your order has been in progress for more than a few hours, just place the order again. Orders that are placed but not downloaded will not affect your quota.

Planet quota policy works on the following rule: Every Pixel is Charged Once. Under this, we have come up with some thumb rules that will help you understand it better:

Metadata assets (_xml, _udm) do NOT count against your quota, in other words, only raster files count against your quota

Landsat and Sentinel datasets do not count against your quota

Activating an asset does not charge quota, only downloading does

Every item will count once regardless of how many times it is downloaded or how many assets are downloaded.

Multiple people can download a scene, the quota will only counted once.

It is also possible to download the visual, analytic etc asset of one scene, the quota will still only be counted once.

You can also download a particular asset multiple times, the quota is still only deducted once

If you still have any doubts about this, please reach out to us at [email protected] and we would be happy to clarify them!

Next-Generation PlanetScope imagery refers to improvements made to our flagship PlanetScope imagery on the sensor and hardware. These enhancements include:

Color and vibrancy is improved, and imagery will look sharper and richer due to narrower spectral response

Surface reflectance values are more accurate, improving quantitative analysis

The footprint of a scene will double as the scene height will double.

For more details, please read this page

Are you looking to order imagery from Planet, but you are not sure what the minimum order value/size is? Please look at the prices below.

Minimum order size as stated in the price list from Oct. 17, 2018:

Monitoring- Minimum order size 5,000 USD OR 4,500 EUR

Archive- Minimum order size 5,000 USD OR 4,500 EUR

Discrete imagery- Minimum order size 5,000 USD OR 4,500 EUR- Maximum 5 Polygons, or 1 single contiguous polygon, should be at least 500 km (minimum corridor width 10 km)

Basemaps- Minimum order size 5,000 USD OR 4,500 EUR

SkySat tasking- Minimum order size 5,000 USD OR 4,500 EUR- Minimum AOI size 25 sqkm - Minimum corridor width 2 km

RapidEye tasking, archive- Minimum order size 5,000 USD OR 4,500 EUR- Tasking min. AOI size to order - 1 single contiguous polygon should be at least 3500 km (minimum corridor width 10 km)

Your quota determines the volume (in square kilometers) of imagery you are able to download with your subscription. The following numbers are approximate. Each image will vary in its exact size.

PlanetScope

PlanetScope Scene (old generation) = 180 - 220 km2

PlanetScope Scene (next generation) - 400 - 600 km2

PlanetScope OrthoTile - up to 625km2

SkySat

SkySat Collect - 40 to 528 km2 (6 to 80km x 6.6km)

SkySat Scene - 2 to 5km2

RapidEye

RapidEye Scene - up to 8,000km2

RapidEye OrthoTile - up to 625km2

Download full specification at link below.

Download full guide at link below.

If you believe that you have found a bug in our API, our platform or one of our tools, please submit a new support ticket.

Basemaps are available as downloadable geotiffs or as web services. For web services, Planet provides both WMTS protocol and XYZ tile services as web service protocols.

Checkout CLI commands for basemaps/mosaics here: https://github.com/planetlabs/planet-client-python/blob/master/docs/source/cli/examples.rst

At present, Planet is accepting applications from students, researchers, technical staff and faculty at accredited, degree-granting universities. Nonprofit and government employees are not eligible. In accordance with Planets licenses, applications from certain countries may not be accepted.

Depending upon the country of origin, users that are approved will have access to view and download up to 10,000 square kilometers of Dove or RapidEye imagery per month for non-commercial research purposes. Downloaded images can be published for illustrative purposes with attribution, such as in figures in peer-reviewed manuscripts, blogs and similar content. Planet data may not be sold or used in the development of value-added products or tools that might be sold.

Planets Building Detection leverages our monthly Basemap product to extract buildings. Due to variations or gaps in the mosaic, we may not be able to detect portions of a building. In addition, buildings smaller than 10 m may be missed.

These Analytic Feeds work by detecting different features on the vessel lines, edges, corners, shapes, etc. As the feeds analyzes the image, it may detect these features and determine a box that may only fit over part of the actual vessel.

A Planet API key allows you to access data programmatically using the Planet API so that you can build your own applications using our imagery.

Every Planet account has an API key and you can find yours by logging into the Planet Apps page. The API key is highlighted at the top of the page.

You can also find your API key listed alongside your name and email address on your account page.

Please protect your API key. If anyone else gains access to it, they will be able to make requests that will use up your quota.

QGIS is a free and open source GIS software which you can install on Windows, MacOS, Linux, and BSD. You can add Planets GeoTIFFs by dragging them into QGIS layer or map view. QGIS will automatically recognize the imagery and locate them properly without any additional input.

documentation

You can open the image properties by double-click on the layer and here are some examples of what you can do within layer properties:

View Information such as the coordinate reference system (CRS), extent, unit, compression method, etc.

Under Symbology tab, you can change the band configuration to suit your needs.

You can view the distribution of the bands or colors in your raster in Histogram tab. The histogram is generated when you click the Compute Histogram button.

For more QGIS related, you can explore their .

Planets Road Detection leverages our monthly Basemap product to extract roads. Due to image variability cloud cover, sun angle, registration issues, etc the registration of our basemaps may shift slightly month over month. As a result, our road detections also slightly shift to match the underlying imagery.

All feedback, requests, bug reports, and questions can be directed to [email protected]

There are a range of applications that can be built from our analytic outputs. Here are just a few to help get you started:

Road Detection and Building Detection

Measure the number of pixels classified as road or building over time within your area of interest (AOI) to develop a proxy for development.

Vessel Detection

Build out a time series analysis to develop a pattern of life of vessels detected per day for your AOI.

View a docked vessel by querying the Analytics API for a specific geo-coordinated bounding box. When the result changes, it can indicate the vessel has left the port.

For more inspiration and guidance on the types of applications you can build, please see our Jupyter notebooks: https://github.com/planetlabs/notebooks/tree/analytics/jupyter-notebooks/analytics

Road Detection and Building Detection outputs are available through a WMTS protocol, which can be used within standard GIS tools, like ArcGIS or QGIS. Learn how to leverage the WMTS protocol here. Vessel Detection outputs can be leveraged within standard GIS software by uploading the GeoJSON feature collections provided through the Analytics API.

Each detection comes with a confidence score, which is an indication of how likely it is that the detection is accurate. Scores range from 0.3 to 1, with 1 being the highest likelihood that the detection is a true positive.

Cloud cover metadata represents the percentage of area covered by clouds within the PlanetScope scene from which the detection was made. This information is gathered from Planets Data API and is available for you to query as well.

For Vessel Detection, within each feature collection, there are links to the Planet Data API to get source imagery. Use the activate link to download the source imagery.

For Road Detection and Building Detection, there are links to the source quad used to derive the output as well as the target quad which is the output itself.

Currently, tip and cue capability consists of detecting signals from the imagery or analytics and manually tasking a satellite by working closely with the Planet team.

Planet can detect roads greater than 10 m (~3 pixels). We define a road as anything that a vehicle would be able to drive on. Road Detection performs best in fully developed or undeveloped urban areas, but also performs very well in rural locations as well. Dry riverbeds, mountainous regions, and forests can cause some false positives, but overall the Analytic Feed has been generalized to work at planetary scale.

Planets Road Detection leverages our monthly Basemap product to extract roads. Due to variations or gaps in the mosaic, we may not be able to detect portions of a road. In addition, roads smaller than 10 m, unpaved roads, tracks, and some snowy conditions will not always be captured.

Planet can detect buildings greater than 10 m (~3 pixels). We define a building as any structure that a person can stand beneath, thus structures such as tents would be detected. Our building detection performs best in fully developed or undeveloped urban areas, but also performs very well in rural locations as well.

At this point, Planet does not classify building types (ie, commercial building, residential house, etc). However, we are interested in the use cases we could tackle together if building classification were made possible.

Planets Building Detection leverages our monthly Basemap product to segment buildings. Due to image variability sun angle, registration issues, etc certain buildings may look different sizes month over month.

Planets Building Detection leverages our monthly Basemap product to segment buildings. Due to image variability sun angle, registration issues, etc the registration position of our basemaps may shift slightly month over month. As a result, our building detections also may slightly shift to match the underlying imagery.

Planet can detect vessels down to about 10 m in size (~3 pixels). Small islands, waves, and clouds could easily be interpreted as small boats (even by the human eye) and particularly on cloudy days or rough sea states. The bigger the vessel is, the higher the likelihood that it will be accurately detected. Vessels in the 10-30 m range will be detected with varying accuracy, while vessels greater than 50m will return higher performing results.

There are many reasons this could be the case a few most likely scenarios would be:

Your AOI was partly cloudy this day. If clouds blocked our view of the ground where detections would have taken place, our feeds will not be able to detect anything

Your AOI was completely cloudy. If clouds completely blocked out your area of interest, Planet might not publish the imagery. If there is no published imagery, there can be no detections.

There was nothing to detect. It could be that although we had a clear shot of the area, there were no objects to detect this day

The Analytic Feeds have a number of layers that ultimately result in a bounding box output. Perhaps the vessel is a slightly different color, there is a haze and cloud over the vessel, and the sea state is slightly different. All these factors contribute to the output.

Some days there may be little to no detections due to the factors listed above. Other days, we may have two separate satellites cross your AOI at different times of the morning. In this case, we may detect the same object multiple times in a day.

Planet is supporting Vessel Detection at the moment, though other features are being considered for the product roadmap. Wed be very interested in what classifications would be useful for you and what use cases you have for them.

Our current capability is optimized for ports and shorelines only. Ports generally have stationary vessels, calmer sea states, and docked vessels. To run analytics over open water, the Analytic Feed will need to be trained with a different dataset (eg, varying wake sizes, wave types, sea states, and other maritime objects.) Planet is exploring this opportunity further.

On some days, we may have two separate satellites cross your AOI at different times of the morning. In this case, we may detect the same object multiple times in a day. These are separate detections, but of the same vessel.

As the Vessel Detection Feed analyzes the image, it may detect waves, sea texture, islands, clouds, and other image artifacts and determine that a vessel is present. Some of these detections may have lower confidence scores and can be filtered out.

Planet can demonstrate how to match AIS data with imagery and has working relationships with existing AIS providers. Planet does not provide the AIS data or implement the integration. However, on request, Planet can work with the user to demonstrate how this is done.

Planets Road Detection leverages our monthly Basemap product to extract roads. Due to image variability cloud cover, sun angle, registration issues, etc certain roads may look different month over month.

If you are interested in gaining access to Planets imagery, please contact sales or email us at [email protected].

Yes, please find a list of RSRs for various satellite constellations below in CSV format. A single archive with all CSV files is also available: RSRs.zip.

PlanetScope

PlanetScope RSRs are grouped based on the ID of the satellite. All images have a four-charactersatellite_idin their metadata, such as0c08. Use the first two characters of thesatellite_idto find the appropriate CSV file. The next generation, PS2.SD, instruments are indicated by "PlanetScopeSD" with the same satellite_id structure as 10xx.

0cxx

0dxx

0exx

0fxx

10xx

SD_10xx

RapidEye

RapidEye RSRs are all found in a single CSV file: RapidEye_RSR.csv.

SkySat

There are individual RSR files for each SkySat. Use thesatellite_id identified found in the image metadata.

SSC1

SSC2

SSC3

SSC4

SSC5

SSC6

SSC7

SSC8

SSC9

SSC10

SSC11

SSC12

Planet aspires to reach upto 70% geometric coverage, it is not 100% because we do not image oceans, cannot reach the poles due to orbital inclination and cannot image 100% of land as there will always be issues related to imperfect scheduling/orbits etc.

Out of the 70% that is covered geometrically there is an certain amount of imagery that makes it to the platform (this is the imagery that users actually see). The reasons for not all imagery to make it to the platform include cloud coverage, rectification problems, orbit anomalies and enough GCPs need to be collected for the imagery to be published.

You can access this imagery in the same manner as you access all PlanetScope imagery.

On Tues 6th August, this new imagery will start to be released with the label test under the quality_category field. From Tues 13th August, imagery will start being published as normal under standard and test in the quality_category field.

Next-Generation PlanetScope imagery will have an additional metadata instrument field that will allow specific filtering.

In the API, you can search by instrument filter for PS2.SD for Next-Generation PlanetScope imagery.

In Planet Explorer, you will first need to enable the instrument filter via the Labs panel.

You can then apply search filters in the search panel for PS2, PS2.SD or Both. PS2 refers to previous PlanetScope imagery. PS2.SD refers to Next-Generation PlanetScope.

Going forward, Next-Generation PlanetScope imagery will be published. Some archive imagery, dating back to January 2019, may also get published.

The imagery products, item, and asset types remain the same, however there are some differences, all of which are explained in more detail in the Combined Imagery Product Specification. The changes most likely to impact customer image processing pipelines are:

The instrument field will have a value of PS2.SD

Scene height is nearly double that of previous PlanetScope scenes

The spectral response is sharper so analysis results may be different

Yes. While Next-Generation PlanetScope offers improved spectral resolution, for most applications, it can be used in tandem with PlanetScope imagery for analysis.

If you are doing quantitative analysis and need to harmonize the two data sources further, an xml file will accompany all new PlanetScope image products, with coefficients to transform the new PlanetScope sensor radiance values to match those of the previous PlanetScope satellites.

No; there will be no changes to our commercial offerings. These will all remain the same and going forward, will now include this new imagery.

We have launched tech demonstration satellites with advanced capabilities that we're testing and seeing great initial results so far. We aren't diving into specifics yet, but stay tuned.

That depends! Wedon'tknow your pipelines and workflows like you do so wecan'ttell you exactly what to change. However we are happy to help support you and answer specific questions. Options include (and are not limited to):

No action. Accept and use the new imagery alongside original Planetscope imagery. This is applicable for example for users doing only visual analysis

Filter out NextGeneration Imagery. If youaren'tquite ready to incorporate this new data, you can explicitly filter it out using the instrument filter.

Harmonization. Use the Harmonization coefficients to align this NextGeneration data with the classic PlanetScope data.