Morphological Reconstruction in OpenCV

问题:

I'm having some problems during image processing in OpenCV. I have text on my image, and during using opening operation, I'm getting not proper output. The issue is quite similar to that provided in this article: http://www.cpe.eng.cmu.ac.th/wp-content/uploads/CPE752_06part2.pdf.

What I can see, people suggest to use reconstruction operations in that cases. My question is if there is any build-in mechanism in OpenCV or some known library/code that implement this? Thank you for help.

回答1:

This answer arrives late, but here is the basic algorithm for under-reconstruction:

1. Inputs are two images: ImReference and ImMarker, with marker <= reference
2. Intermediate image: ImRec
3. Output image: ImResult
4. Copy ImMarker into ImRec
5. copy ImRec into ImResult
6. ImDilated = Dilation(ImResult)
7. ImRec = Minimum(ImDilated, ImReference)
8. If ImRec != ImResult then return to step 5.

It's not the most optimal algorithm, but it uses only basic operations.