# Context Satellite imagery provides unique insights into various markets, including agriculture, defense and intelligence, energy, and finance. New commercial imagery providers, such as [Planet](https://www.planet.com/) and [BlackSky](https://www.blacksky.com/), are using constellations of small satellites to exponentially increase the amount of images of the earth captured every day. This flood of new imagery is outgrowing the ability for organizations to manually look at each image that gets captured, and there is a need for machine learning and computer vision algorithms to help automate the analysis process. The aim of this dataset is to help address the difficult task of detecting the location of airplanes in satellite images. Automating this process can be applied to many issues including monitoring airports for activity and traffic patterns, and defense intelligence. Continusouly updates will be made to this dataset as new Planet imagery released. Current images were collected as late as July 2017. # Content Provided is a zipped directory `planesnet.zip` that contains the entire dataset as .png image chips. Each individual image filename follows a specific format: {label} __ {scene id} __ {longitude} _ {latitude}.png - **label:** Valued 1 or 0, representing the "plane" class and "no-plane" class, respectively. - **scene id:** The unique identifier of the PlanetScope visual scene the image chip was extracted from. The scene id can be used with the [Planet API](https://www.planet.com/docs/reference/data-api/) to discover and download the entire scene. - **longitude_latitude:** The longitude and latitude coordinates of the image center point, with values separated by a single underscore. The dataset is also distributed as a JSON formatted text file `planesnet.json`. The loaded object contains **data**, **label**, **scene_ids**, and **location** lists. The pixel value data for each 20x20 RGB image is stored as a list of 1200 integers within the **data** list. The first 400 entries contain the red channel values, the next 400 the green, and the final 400 the blue. The image is stored in row-major order, so that the first 20 entries of the array are the red channel values of the first row of the image. The list values at index *i* in **labels**, **scene_ids**, and **locations** each correspond to the *i*-th image in the **data** list. ## Class Labels The "plane" class includes 8000 images. Images in this class are near-centered on the body of a single airplane, with the majority of the plane's wings, tail, and nose also visible. Examples of different aircraft sizes, orientations, and atmospheric collection conditions are included. Example images from this class are shown below.  The "no-plane" class includes 24000 images. A third of these are a random sampling of different landcover features - water, vegetion, bare earth, buildings, etc. - that do not include any portion of an airplane. The next third are "partial planes" that contain a portion of an airplane, but not enough to meet the full definition of the "plane" class. The last third are "confusers" - chips with bright objects or strong linear features that resemble a plane - that have previously been mislabeled by machine learning models. Example images from this class are shown below.    # Acknowledgements Satellite imagery used to build PlanesNet is made available through Planet's [Open California](https://www.planet.com/products/open-california/) dataset, which is [openly licensed](https://creativecommons.org/licenses/by-sa/4.0/). As such, this dataset is also available under the same CC-BY-SA license. Users can sign up for a free Planet account to search, view, and download thier imagery and gain access to their API.