Planetary body limb and plume labels for NASA images

This data set was compiled to aid in evaluating methods for automated analysis of images to detect planetary bodies and limbs.  It contains manually generated labels for 308 NASA images of planets and moons.  The labels annotate the location of the limb (edge) of the body and plumes emitted by the body, if any.  "Plume" in this context refers to any bright material emitted from the body, such as icy plumes from Enceladus or volcanic plumes from Io. 112 of the labeled images contain plumes.

Contents:

This data set covers images collected by the following instruments:

1. Cassini Imaging Science Subsystem (ISS)
2. Galileo Solid-State Imaging (SSI)
3. MESSENGER Mercury Dual Imaging System (MDIS)
4. New Horizons Long Rang Reconnaissance Imager (LORRI)                                                                     

The target bodies include the planet Mercury; Jupiter's moons Callisto, Europa, Ganymede, and Io; and Saturn's moon Enceladus.    

There is a directory for each instrument_target combination:

• cassini_iss_enceladus/: Cassini ISS narrow-angle camera observations of Saturn's moon Enceladus
• galileo_ssi_callisto/: Galileo SSI observations of Jupiter's moon Callisto
• galileo_ssi_europa/: Galileo SSI observations of Jupiter's moon Europa
• galileo_ssi_ganymede/: Galileo SSI observations of Jupiter's moon Ganymede
• galileo_ssi_io/: Galileo SSI observations of Jupiter's moon Io
• messenger_mdis_mercury/: MESSENGER MDIS narrow-angle and wide-angle observations of Mercury 
• new_horizons_lorri_io/: New Horizons LORRI observations of Jupiter's moon Io

Source images:  The images that are associated with each label file can be obtained from the Planetary Data System (PDS) at https://pds-imaging.jpl.nasa.gov/search .  For Cassini ISS, MESSENGER MDIS, and New Horizons LORRI images, search on the product id from the label filename.  For example, the product id for  

lor_0035092814_0x630_sci_label.yml

is

lor_0035092814_0x630_sci

For Galileo SSI images, the filename does not include the product id. A list of the source product ids is included in the file named 

galileo_image_ids.txt

Label format:

Labels are stored in YAML format.  The limb is annotated as a series of points marked along the limb such that a least-squares circle fit of those points provides a model of the body's limb ("points" field).  Plumes, when present, are indicated as one or more angular ranges (in radians) around the limb within which plume activity is present ("plumes"->"intervals" field).  Angles are specified starting with 0 radians (up) and proceeding clockwise.  The user who generated the labels is recorded in the "user" field.   

Example (galileo_ssi_io/0085r_label.yml):

Six points define the limb of the body, and there are two areas of plume activity. 

comment: Points are in (x, y), i.e. (col, row), order.                          
plumes:                                                                         
  comment: Intervals in radians                                                 
  intervals:                                                                    
  - [2.8540078295092326, 3.020573420947303]                                     
  - [3.219430581132992, 3.352047930386058]                                      
  user: mcameron                                                                
points:                                                                         
- [123.09103311855094, 322.15933747194225]                                      
- [143.78220150957688, 79.06162214909591]                                       
- [89.23475042871942, 219.0261628709045]                                        
- [256.9650789538681, 407.0275241755675]                                        
- [176.84241267100913, 375.9514805780413]                                       
- [113.07652569773596, 121.72097703075002]                                      
user: mcameron

Attribution:

If you use this data set in your own work, please cite this DOI: 10.5281/zenodo.2556063 .