Previous Message | Next Message Previous in Topic | Next in Topic Previous by Same Author | Next by Same Author Previous Page (May 2002) | Back to Main IMAGEJ Page
	Reply | Post a New Message | Join or Leave IMAGEJ, or Change Options | Search
	Chronologically | Most Recent First | Wrap Text (Proportional Font) | Don't Wrap Text (Non-proportional Font)

Content-Type: text/plain; charset="ISO-8859-1"
Message-ID:  <[log in to unmask]>
Date:         Wed, 29 May 2002 07:19:20 -0700
Reply-To:     ImageJ Interest Group <[log in to unmask]>
Sender:       ImageJ Interest Group <[log in to unmask]>
From:         "Cunningham, Bob" <[log in to unmask]>
Subject:      Re: Counting particles II

Actually, yes!  Years ago I had to do something similar.  For my problem,
the key was to count specular reflections rather than try to search for and
uniquely recognize partially occluded spheroids (a task I attempted and
failed to accomplish).

You may have to polish your grapes to make it work...

I tried several methods of image segmentation followed by hue/chroma
screening, but the complex diffuse polychromatic reflections in my scene
made consistent color recognition difficult.  But I did have a single
dominant direct illuminant, so I could search for specular reflections and
detect individual foreground objects.

I detected and identified specular reflections using simple (non-adaptive)
binary thresholding followed by a blobbing algorithm.

One neat property of specular reflections is that a plot of the color values
from the center to the edge of the specular reflection generates a vector in
color space that points toward the actual object color.  Conversely, if you
know the object color, then the vector points toward the illuminant.

Unfortunately, the color vectors were quite noisy, so I didn't use them in
my application.  The specular reflection alone sufficed.  And it also worked
using a monochrome camera.

I should port that algorithm to ImageJ someday...


-BobC

-----Original Message-----
From: atlan [mailto:[log in to unmask]]
Sent: Wednesday, May 29, 2002 5:10 AM
To: [log in to unmask]
Subject: Counting particles II


Has someone ever tried to count the number of fruit in a photo of a bunch
of grapes ?
thanks

Center for Information Technology National Institutes of Health Bethesda, Maryland 20892 301 594 6248 (v) 301 496 8294 (TDD) Comments and Assistance Accessibility