I am doing research on processing medical images which contain some cells which adhere to other cells. Considering the fact that I should extract the cell’s number, I need to find a method in order to make these cells separated.
It would be kind of you to give me your valued suggestions.
Thank you.
Marjan
Dear M. Awad,
For example, in this image there are about 17 cells, but after running my code 12 is produced as the number of cells. I used GA for finding the proper threshold, but I want to find a more convenient way for this purpose.
Thank you so much.
There are some ideas that can be used to find objects from an image which are included in the following code. The code is used to demonstrate different morphology function in my image processing class.
Additionally, you may also try to use the functions "bwconncomp" and "regionprops" in Matlab to find the bounding box of each object. Please see the help of these two functions in Matlab. There are examples on how to use them.
Check out these different methods which can be combined with morphology to get the desired results.
Take a look to the document below. It is based on ImageJ and uses the function 'Watershed'. This function can often (but not always) accurately cut them apart by adding a 1 pixel thick line where it feels the division should be.
http://www.unige.ch/medecine/bioimaging/tricks/imagejtutorials/CellCounting.pdf
Following two threads are about the same issue: separating adjacent cells. You could use watershed or calculate angular deficits to pinpoint concavities in contours.
https://www.researchgate.net/post/The_best_way_of_separating_overlapping_objects?_tpcectx=home_feed
https://www.researchgate.net/post/Which_techniques_are_useful_for_determination_of_cell-cell_contacts_in_live_cell_microscopy_images?_tpcectx=home_feed
See also:
C. Zhang, C. Sun, R. Su, and T. D. Pham, Segmentation of Clustered Nuclei Based on Curvature Weighting
and
http://blogs.mathworks.com/steve/2006/06/02/cell-segmentation/
I have done such a project as my B.Sc. thesis before. you can use structuring elements so as to to do this. the following link provides you a good introduction to structuring elements.
https://www.cs.auckland.ac.nz/courses/compsci773s1c/lectures/ImageProcessing-html/topic4.htm
please also see structuring elements in Matlab help
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Signal Processing
- Image Processing