kevin218/least_asymmetry

This code is to find the center of a distribution of light in an image using the least asymmetry method. The code also contains center of light and fitting a Gaussian routines as well. All functions are designed to take optional weights

Least asymmetry

This is a python module for calculating the center of distributions using the method of determining the point of least asymmetry. A description of the algorithm can be found in the docs directory. The description is a stub of a paper demonstrating its use for astrophysical measurements found here.

This module may either be used directly from a clone by running the ./build.sh script to build the c++ extension, or by using the setup.py file to install the module into the site packages directory. The actr routine itself is accessible from the module itself, or from the asym.py file. As a side effect of how the routine works, this module also provides centering by fitting a gaussian (fitgaussian) or finding a weighted average (col) and can thus be used as a general centering package

Module level Example:

# create a demo image
import numpy as np
indy, indx = np.indices((30, 30))
gaus = a.gaussian(15, 14.5, 14.5, 3, 3, 0)
image = gaus(indy, indx)

noise = np.random.normal(0, 2, len(image.flatten()))
noise = noise.reshape(image.shape)
nimage = image+noise

# Use the routine
import least_asymmetry as a
center = a.actr(nimage, [14,14])[0]