Planet Incubator Program
About ASU's Planet Incubator Program
In 2019, Arizona State University joined forces with Planet as its first institutional data partner for higher education. Operating the largest constellation of satellites currently in low Earth orbit, Planet acquires daily global coverage of the entire landmass of our planet and its coral reefs at an astonishing three to five meter resolution. Many of these satellites are as small as a loaf of bread but, operating together, they collect over 11 terabytes of data every single day.Sign up
How You Can Benefit
All ASU students and researchers have free access to Planet’s growing catalog of imagery through the Planet Education and Research Departmental and Campus License program. Research that can benefit from this data includes: landcover change, vegetation applications, agriculture, hydrology and cryosphere research, coral reef research, as well as human geographical and social applications. The ability to observe environments frequently and at high resolution via Planet’s Dove and RapidEye satellites provides new insights into their dynamic properties.
Some of the potential educational opportunities with the data include:
- Python programming for Geoscience: Planet data is an excellent source of data with many Earth science applications.
- Geographic Information Systems (GIS): Planet data can easily be ingested into popular GIS applications, opening up a variety of project opportunities for students.
- Artificial Intelligence (AI) for Earth Science: Advanced applications such as AI and computer vision can utilize Planet imagery.
- Google Earth Engine: Planet’s API allows for users to feed their data into Google Earth Engine for advanced processing in the cloud.
Unlike most other free data, the images from Planet have a three to five meter resolution, much higher than the 30 meter data from Landsat or Sentinel. At that resolution, all kinds of opportunities open up to use object identification or artificial intelligence (AI) for advanced applications.
Not comfortable with something that sounds so techie? Not to worry, check out the tutorials below to learn how to roam through the data to identify and download the data that you need. The data exist in different flavors and formats, from visuals for general illustrations to analytic products for data science applications.
The tutorials are designed for various levels of expertise. The easiest one is to use the Planet Explorer or Basemap Viewer websites to roam through the data to filter through images by location, date range, cloud cover, and many other attributes. One you’ve identified the data that work for your project or application, you can download the data. You can also establish routine searches to always be on the look-out for the latest imagery of your area of interest (AOI). The possibilities are endless!
Wondering how you can make use of Planet imagery and don’t know where to start?
Take advantage of the new virtual office hours where a skilled remote sensing specialist can answer your questions and give you advice and guidance on using Planet imagery.
Just send an email to PlanetAdvice@asu.edu with your questions or problem description. We can even arrange a Zoom meeting to show demonstrations.
Mosaicking Planet Data Using QGIS
This tutorial will focus on the use of QGIS for mosaicking Planet Images, but similar functionality exists in ArcGIS. As an sample use case, this tutorial will create a mosaic of various tiles from a Planet basemap.
Using the Planet Explorer Plugin to QGIS
This video tutorial shows you how to install and use the Planet Explorer plugin for QGIS. This allows users to search for, order, and download data while intereacting with their other data in QGIS.
Planet Imagery with Python
This video is an excerpt from the "Planet Imagery with Python" webinar given on August 28, 2020. It describes the steps needed to install and run the Planet API client in the Google Colab environment. It also explains how to connect with Google Drive as a location from which data can be read and written in the Colab environment.
Calculating NDVI with Planet using Python
This video is an excerpt from the "Planet Imagery with Python" webinar given on August 28, 2020. This video covers how to calculate NDVI from Planet data using the Google Colab environment with the Python programming language.
Classify Python Imagery with Python
This video is an excerpt from the "Planet Imagery with Python" webinar given on August 28, 2020. This video covers how to classify a Planet image using the Python programming language in the Google Colab environment.
Planet Imagery and Point data
This video is an excerpt from the "Planet Imagery with Python" webinar given on August 28, 2020. This video covers how to use Planet imagery with vector (point) data using the Python programming language in the Google Colab environment.
Masking Clouds in Planet Data
This video is an excerpt from the "Planet Imagery with Python" webinar given on August 28, 2020. This video covers how to mask clouds using the Google Colab environment with the Python programming language.
Planet Data for Agriculture
This video tutorial (7 min., 46 sec.) shows you how to use Planet imagery to identify agricultural fields that have significantly reduced NDVI from one image to another. This can give a rough indication of what fields have been harvested during a growing season.
Machine Learning with Planet Data
This video tutorial (13 min. 24 secs.) describes and demonstrates how to extract and prepare data from a Planet satellite image to use for creating a convolutional neural network model. Subsequent videos will show how to use that data to fit a model.
The Python code file used in this video can be found here.
Machine Learning with Planet Data (Building a Model)
This video tutorial (20 min. 38 secs.) describes and demonstrates how to build a model with the data prepared and extracted as in the previous video. The model is built with the a subset of the data for training and assessed using a test set of data. The Python code file and a conda environment file can be found here.
Machine Learning with Planet Data (Tuning a Model)
This video tutorial (14 min. 57 secs.) describes and demonstrates how to tune a Machine Learning model using Keras Tuner to find the best parameters to maximized classification accuracy. The Python code file and a conda environment file can be found here.
Planet requests that you use the following reference:
Planet Team (2017). Planet Application Program Interface: In Space for Life on Earth. San Francisco, CA. https://api.planet.com.However, you may also want to include an acknowledgement that the data were provided through Planet's Education and Research Program.
Biodesign Institute C building, 5th floor
1001 South McAllister Avenue
Tempe, AZ 85287
60 Nowelo Street
Hilo HI 96720