Mixture Modeling Thresholding

QR for this page

From Fiji

Jump to: navigation, search
Mixture Modeling (ImageJ)
Author Christopher Mei, Maxime Dauphin
Maintainer
File Mixture_Modeling.jar (18,454 Bytes)
Source in .jar file
Initial release 15 December 2003
Development status unknown
Category Segmentation
Website [1]


Purpose

This plugin automatically threshold an image using the Mixture Modeling algorithm. It is an histogram-based technique that assumes that the histogram distribution is represented by two Gaussian curves.

Documentation

This algorithm separates the histogram of an image into two classes using a Gaussian model. It then calculates the image threshold as the intersection of these two Gaussians. This thresholding technique has the advantage of finding a threshold that is in certain cases closer to real world data. The Gaussian parameters can also be used to describe the two regions obtained.

The plugin returns a histogram with the two Gaussians, the parameters obtained (average, standard deviation, threshold) and the thresholded image.

It has the same drawbacks as the Otsu Thresholding plugin (by the same author), which is that it behaves annoyingly with stacks.