===========================================================
 Mathematical Morphology plugins for ImageJ
===========================================================
Author: Adam DeConinck
Date last modified: 20 February 2012

Summary: 
A set of plugins for the popular image processing program
ImageJ to implement morphological processing operations 
using a custom structuring element. Most of the logic
is implemented in the org.ajdecon.morphology package
included here; top-level .java files are the plugin 
user interfaces.

Simple build instructions:
1. Make sure that the ImageJ package (ij.jar) is in the
current directory, or modify the makefile.
2. To compile: make
3. To produce plugin .jar file: jar cvf Morphology_.jar ./*

Installation:
1. Install ImageJ. (Latest version found at 
http://rsbweb.nih.gov/ij/)
2. Copy Morphology_.jar into the plugins directory of 
ImageJ.

Usage:
1. Open a binary or grayscale image in ImageJ.
2. Open the Plugins>Morphology menu and select New 
Structuring Element. This will produce a simple image 
such as a square, circle, or line which will be used to 
operate on the image.
3. Open thr Plugins>Morphology menu and select another 
operation such as Erode. Select the object image and 
structuring element in the dialog which comes up. This 
will perform the given morphological operation on the 
object image.

Detail:
Mathematical morphology is a technique for processing 
geometrical structures, particularly in images. It 
generally makes use of a simple shape called a structuring
element, which is then superimposed with its center over 
each pixel of an object image (the "target" pixel). 
Depending on the values of the object pixels corresponding 
to the shape of the structuring element, the value of the
target pixel may be changed. These techniques have 
applications in various types of object recognition and
segmentation.

A simple example of a morphological operation is erosion 
in a binary image.  For each pixel, examine the values of
the pixels in the neighborhood defined by the structuring
element. If they are all foreground pixels, the target 
becomes a foreground pixel; otherwise, it becomes a 
background pixel. This has the effect of eroding the 
edges of large structures with a bias defined by the 
shape of the structuring element.

(see http://en.wikipedia.org/wiki/Mathematical_morphology
for more information)

My research over the past few years has produced large 
amounts of data in the form of microscopy images. 
Analyzing this data has required, among other things, the 
use of morphological processing to analyze the shapes and
separation of physical objects imaged. While these 
operations are all well-supported in scientific programming
languages such as Matlab or scipy, these don't make the
best tools for visual exploratory data analysis.

ImageJ is a popular image processing tool in the scientific
community, but its morphological image processing functions
are rudimentary and do not support different structuring
elements. To visually explore my data, and better learn
mathematical morphology, I wrote the attached analysis code
based on the ImageJ API and plugins for the user interface.
ImageJ may also be packaged as a convenient library for
use in other software, and I have explored using this code
for large scale morphological image processing with a 
Java framework such as Hadoop.

Contents:

makefile

plugins.config: instructions for the ImageJ user interface.

Plugins: Dilate_.java, Erode_.java, Hit_Or_Miss.java, 
Morph_Close.java, Morph_Open.java, 
Morphological_Gradient.java, New_Structuring_Element.java,
Otsu_Threshold.java, Top_Hat.java

org/ajdecon/morphology/: a package for image morphology 
using the ImageJ API, containing:

  StructElement.java: object to define structuring elements.
  Morphology.java: utility methods for image morphology.