scan through a map of {0,1}s, work out how many regions there are, then decide which regions belong together to form an object, finally seperate binary map into seperate images, each contain regions belong to one image.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
scan through a map of {0,1}s, work out how many regions there are, then decide which regions belong together to form an object, finally seperate binary map into seperate images, each contain regions belong to one image.
Currently(v29):
imlabel.m - label image into regions, also produce a region count.
imlink.m - produce table showing which regions are linked.
The link table is alright, but I don't like the look of it :(.. Don't like doing branches and tree type of structures.
I hope it's avoidable. Or, maybe I just have to get use to it.
Have to think up of some smarter algorithms.
So many small things have to take into consideration...