project: stellarPYL // file: README // author: Brunston Poon // org: UH-IFA and SPS
For licensing information see LICENCE.txt.
This is stellar spectra reduction and analysis command-line software written using the Python 3.4 version of the Anaconda Scientific Python distribution. It is work done for an internship at the University of Hawaii in conjunction with the St. Paul's School Engineering Honors program.
It aims to provide a one-click workflow for analyzing uncompressed TIFF stellar spectra images obtained from a DSLR through a diffraction grating. The goals of the project are: to automatically crop the image; to perform background subtraction; to create an intensity plot of the spectrum (accounting for non- orthogonal spectra); and to account for the use of a DSLR sensor by using either a relative response function or an absolute response function to negate the effects of sensor response on the intensity plot.
Run ui.py in the command line:
python ui.py
Be sure you are using Python3 and installed numpy and Python Imaging Library. Alternatively, you can run this using the Python3 version of Anaconda. If you do not satisfy these prerequisites, the program will not function correctly.
AS AN ALTERNATIVE TO THE BELOW, make sure you set a default threshold using 'settings_threshold', and then simply type 'auto' to have the program do the majority of the work.
You will be presented with a list of commands.
For a brand new image, run 'crop' first. Drag your file into the same directory and enter the filename including the file extension. This program will accept TIFF files, either in .tif or .tiff extension format. It will then ask you for a threshold.
The threshold is used throughout the program to determine what data is relevant and what parts of the image can be discarded without damaging the value of the data. It needs to be an integer value between 0 and 765 as the threshold is measured as the sum of the R, G, and B bin values in a pixel, therefore, each RGB value can be an integer from 0-255; total value can be from 0-765. If you do not have a value you are already using for all of your images, you can type 'pixel_d' at the command prompt to run a function that plots the distribution of binned pixel values in your image.
A typical threshold may be in the range from 100-130.
The program will run the cropping algorithm and ask for a filename to give to the new file.
The next command you should run is 'intensity_saa'. It will take an image file and a threshold and automatically perform linear regression to find the y=mx+b line on which the spectral trace lies. It will then step one pixel at a time along the spectral trace and add up all intensity values occuring along that line.
The program will graph this intensity plot, which can be saved using the tools already provided by matplotlib.
-
'autoProcess' (short 'auto') - autoProcess will take care of cropping and doing intensity plotting for you. just provide a filename. In order to use this feature you must first set a default threshold to use by using the 'settings_threshold' command.
-
'pixel_d' (short 'pd') - takes an image and shows the pixel distribution of the image over the intensity of the pixels.
-
'crop' - takes an image and crop it based on your selected threshold.
-
'image_regression' (short 'imgreg') - takes an image and finds the line which goes through the spectrum in that image.
-
'intensity_n' (short 'n') - takes an image of a spectrum and converts it into an intensity plot using the naive method of adding.
-
'intensity_saa' (short 'saa') - takes an image of a spectrum and converts it into an intensity plot using spatial anti-aliasing at a sub-sampling rate of one tenth of one pixel.
-
'show_threshold' - see exactly what could be removed (assuming no crop stop has been set) using the threshold that is currently set.
-
'show_regression' - shows regressed line overlayed on the original (cropped) image.
-
'show_walks' - shows walking lines overlayed on the original (cropped) image.
-
'about' - displays information about this program
-
'functions' - where you are now
-
'help' - brings up sample workflow
-
'settings_cropoverride' - sets manual overrides for automatic cropping on the top, bottom, and sides of an image. The default value is -1 (which is equivalent to no override) for all values.
-
'settings_default' - returns ALL settings back to default: -defaultThreshold = -1 -autoIntensity = saa -manual overrides all to -1 -step = 1 -verbose = yes -showthresh = yes
-
'settings_intensity' - sets default intensity processing method for the autoProcess feature. The default setting is saa (for spatial anti-aliasing).
-
'settings_margin' - sets margin for cropping. default is 5 pixels.
-
'settings_step' sets default step value along the spectral trace (and thus resolution of resulting intensity plot). default is 1 pixel-equivalence.
-
'settings_threshold' - sets default threshold. Set to -1 if you would like the program to always ask. The default setting is -1 (always asks).
-
'settings_verbose' - sets verboseness. 'yes' to include debug statements, 'no' is default.
-
'view_settings' - view your current settings
| 'jellyfish' | who doesn't need a jellyfish? |
The name stellarPYL is derived from three things: 'stellar' from its purpose as stellar spectra reduction software, 'PY' from its codebase written in Python, and the 'L' from 'libre' in reference to the license under which it is released, namely the GNU General Public License v3.
Pronounciation: "stellar pile"