Skip to content
Converting SMART to Python3
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
testing
tracker
LICENSE
README.md
config.ini
detect.py
detect_core.py
input_data.py
magnetic_properties.py
plot_detections.py
position_properties.py
process_magnetogram.py
psl_properties.py
requirements.txt
test.py
wrapper.py

README.md

smart_python

A line-by-line translation of SMART IDL. A work in progress...

Take an SDO magnetogram and run SMART to detect magnetically interesting regions. In default mode, a .json property file and .png image of detections will be output. Currently to run use the wrapper, e.g.,

import wrapper
wrapper.main(fits_file='hmi.M_720s.20170906_120000_TAI.fits')

Optional keyword inputs:

  • fits_file: run SMART on specifically defined fits file (default will take latest available fits from JSOC)

Optional outputs:

  • Save fits files of magnetogram and detections (to be added)

Note that the SMART code can be tailored to whatever user requirements are necessary via the config.ini file. Here, a data directory can be defined for where the data files are kept (default is /data/smart/), as well as various thresholds changed for detection purposes.

JSON output

The code will output a JSON file in the format yyyymmdd_HHMM_properties.json, which contains property values for all detected SMART regions (blank file if none), including position, magnetic, and polarity separation line information*

Section Subsection Description
meta dateobs time of observation (yyyymmdd_HHMM)
dimension pixel resolution
instrument observation source
posprop arid SMART detection no.
x(y)bnd x(y) pixel coordinates of box boundary of detection
x(y)cenflx(area) centre of detection in pixel coordinates based on flux(area)
hcx(y)bnd heliocentric coordinate boundaries in arcseconds
hcx(y)flx(area) heliocentric coordinates of centre of detection based on flux(area) in arcseconds
hglat(lon)bnd solar heliographic latitude(longitude) coordinate boundaries
hglat(lon)flx(area) solar heliographic latitude(longitude) coordinates of centre of detection based on flux(area)
carlonbnd carrington longitude of boundary
carlonflx(area) carrington longitude of centre of detection based on flux(area)
magprop arid SMART detection no.
areabnd total area of boundary box region (millionths of a solar hemisphere)
posareabnd area of positive magnetic field part of boundary box region (millionths of a solar hemisphere)
negareabnd area of negative magnetic field part of boundary box region (millionths of a solar hemisphere)
totarea total area of detection (millionths of a solar hemisphere)
posarea area of positive magnetic field part of detection (millionths of a solar hemisphere)
negarea area of negative magnetic field part of detection (millionths of a solar hemisphere)
totflx total magnetic flux of detection (maxwell)
posflx magnetic flux of positive magnetic field part of detection (maxwell)
negflx magnetic flux of negative magnetic field part of detection (maxwell)
imbflx magnetic flux imbalance (maxwell)
frflx flux fraction of detection, i.e. (posflx - negflx) / totflx
bmin total negative magnetic field strength of detection (gauss)
bmax total positive magnetic field strength of detection (gauss)
bmean mean magnetic field strength of detection (gauss)
pslprop arid SMART detection no.
psllength polarity inversion line length of detection (megameters)
pslsglength strong gradient PIL length (megameters)
pslcurvature curvature of PIL
rvalue r value (maxwell)
wlsg gradient-weighted integral length of PIL (gauss)
bipolesep_mm bipole separation of detection (megameters)
bipolesep_px bipole separation of detection (pixels)

*note PIL properties might have zeros since deprojection not implemented (see issue #3).

Tracking

SMART can be run in 'tracking' mode, where the evolution of SMART properties will be plotted over time since the last detection (default six hours previous). The code, developed by Sean Blake, can also be run stand alone, as long as default json and optional processed fits files are available (see optional outputs above - to be added). More description to be added once integration is complete!

External dependencies

astropy
matplotlib
numpy
opencv
pandas
scikit_image
scipy
sunpy

See requirements.txt for latest development versions created with pipreqs.

License

The content of this project is licensed under the Creative Commons Attribution 4.0 license, and the underlying source code used to format and display that content is licensed under the MIT license. Please reference the original Higgins et al publication in all instances and relevant GitHub repositries if using the code.

You can’t perform that action at this time.