The key is the optical system. As for detection algorithm,you can do it using another software,such as HALCON,OpenCV,there are many files to check.
. I have developed some relevant products.What you did is machine vision system.So,the system setup is very important .
------------------ Original ------------------
From: "Toufic El Arnaout"<
[hidden email]>
Date: Tue, Nov 11, 2014 10:08 PM
To: "IMAGEJ"<
[hidden email]>;
Subject: detection of crystals
Hello everyone,I would like to know from experts in object detection. I currently have some images of crystals in solution. The problems with these are a- the internal reflection and shadows in each or many crystals, b- inside the crystals the color is same as as the background, c- many crystals overlap (especially at high crystal concentration),d- reflections and shadows inside the same crystal (some sides of it for example) or sometimes from a neighboring crystal
I tried many algorithms such as Canny edge detection, FeatureJ, EdgeFitter, Biovoxxel, etc etc... As a total beginner, I also played around with all tools preinstalled in ImageJ/Fiji, specially thresholding, fill object, image pretreatment, ...The problem is that even if the image is very well optimized for processing, I need objects to be well separated/extracted, and their edges useful for other applications such as binary conversion, etc and particle analysis.Any hint, update, recent developments and useful tools for these types of images will be appreciated.I understand that the magic of brain is fantastic, and that we always have to ask the question if what/how we see really present there for the program to detect... But help me with tips specially if useful in a macro.Kind regardstoufic
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
