plugins - image type and flood filling

Posted by Ben Tupper on Sep 18, 2008; 2:45pm
URL: http://imagej.273.s1.nabble.com/plugins-image-type-and-flood-filling-tp3695038.html

Hi All,

I have a two part question for which I have one example plugin.

The first question is about image type.  In the plugin I create  
duplicate of a 32-bit image. I subsequently threshold that duplicate  
and then convert it to a mask which makes it 8-bit.  I then select a  
pixel in the mask that I know has a value of 255 *in the display*. I  
then programmatically get the value of the pixel a number of ways  
using ImageProcessor methods and using one ImagePlus method.  Here's  
the output of from running the plugin (pasted in below).

Mask is a Gray32? true
Mask bit depth = 8
[13, 294]: Orig= 224.0
Mask with ImageProcessor.getPixel() = 1130364928
Mask with ImageProcessor.getPixelValue() = 224.0
Mask with ImageProcessor.getf() = 224.0
Mask with ImagePlus.getPixel() = 255

The first 2 lines of output are confusing because the Mask image  
processor thinks it is 32-bit even though it also thinks it is 8-bit.

The next line shows the pixel coordinates I chose and the original  
image value.

The next 4 lines show the value of the pixel using different methods  
- only the last gives me what I *think* should be the correct value.

Can someone shed light on (a) why the Mask thinks it is both 32-bit  
and 8-bit and (b) what is the "right" way to retrieve  pixel values  
using ImageProcessor directly?


The second question is about FloodFiller. I believe that I can use a  
seed pixel derived from one image to flood fill a second image. What  
I can seem to fathom is what is the effect on the second image?  
Using the Macro.Functions.floodFill(x,y) as an example, I have tried  
to flood fill the mask (above) with pixels derived for the Edge  
image.  In the plugin some 6700 pixels qualify as seed pixels, but I  
can't see what effect the filling has on the mask.  When I use the  
macro tool (the bucket icon on the toolbar) it has an obvious effect.  
How do I manage the FloodFiller so that I can distinguish unqualified  
and unfilled blobs from the qualified and filled blobs?

Thanks!
Ben

import ij.*;
import ij.process.*;
import ij.gui.*;
import java.awt.*;
import ij.plugin.*;

public class Test_GetPixel implements PlugIn {

   ImagePlus origImp, maskImp, edgeImp;
   ImageProcessor origIp, maskIp, edgeIp;

        public void run(String arg) {
         //load the image and make it "just so"
                IJ.run("Cell Colony (31K)");
                IJ.run("Invert");
                IJ.run("32-bit");
               
                ImagePlus origImp = IJ.getImage();
                ImageProcessor origIp = origImp.getProcessor();

   //create a mask of the image as a 0-1 binary float
     ImagePlus maskImp = new ImagePlus("Mask", origIp.duplicate());
     ImageProcessor maskIp = maskImp.getProcessor();
     maskIp.resetMinAndMax();
     maskIp.setThreshold(72, 5000000, FloatProcessor.RED_LUT);
     maskImp.show();
     IJ.run("Convert to Mask");
     maskImp.updateAndDraw();

     int i = 13;
     int j = 294;
     int w = maskIp.getWidth();
     int h = maskIp.getHeight();

     IJ.log("Mask is a Gray32? " + (maskImp.GRAY32 == 2));
     IJ.log("Mask bit depth = " + maskImp.getBitDepth());
     IJ.log("["+i+", "+j+"]: Orig= " + origIp.getPixelValue(i,j));
     IJ.log("Mask with ImageProcessor.getPixel() = " +  
maskIp.getPixel(i,j));
     IJ.log("Mask with ImageProcessor.getPixelValue() = " +  
maskIp.getPixelValue(i,j));
     IJ.log("Mask with ImageProcessor.getf() = " + maskIp.getf(i,j));
     IJ.log("Mask with ImagePlus.getPixel() = " + (maskImp.getPixel
(i,j))[0]);


   //create an edge image
                ImagePlus edgeImp = new ImagePlus("Edge", origIp.duplicate());
                ImageProcessor edgeIp = edgeImp.getProcessor();
     edgeIp.resetMinAndMax();
                int[] gauss = {3, 5, 3, 5, 8, 5, 3, 5, 3};
     int[] laplace = {0, -1, 0, -1, 4, -1, 0, -1, 0};
                edgeIp.convolve3x3(laplace);
     edgeIp.convolve3x3(gauss);
                edgeImp.show();

     FloodFiller ff = new FloodFiller(maskIp);
     int c = 0;
     for ( i = 0; i < w; i++){
                  for ( j = 0; j < h; j++){
                    if ( (edgeIp.getPixelValue(i,j) > 18.0) && (maskImp.getPixel
(i,j))[0] == 255) {
                      //IJ.log("Fill= " + i + ", " + j);
                      ff.fill(i,j);
                      maskImp.updateAndDraw();
                      c++;
                    }
        IJ.showProgress(i, w);
       } //j loop
     }//i loop
     IJ.log("");
     IJ.log("Found " + c + " qualifying seed pixels for flood filling");
   } //run

}