How to perform morphological reconstruction by dilation?

I am performing retinal blood vessel segmentation and i need to perform morphological reconstruction by dilation.

I am implementing a paper titled " AUTOMATIC DETECTION OF MICROANEURYSMS AND HEMORRHAGES IN COLOR EYE FUNDUS IMAGES" by Sérgio Bortolin Júnior1 and Daniel Welfer2 .

I quote from paper," the detection of the blood vessels was performed from a morphological reconstruction by dilation, denoted by R. In this operation, we used a diamond-shaped structuring element with radius of 1 pixel."

In the below image, first image is f9 and the second image is f10 (the reconstructed output)

I am attaching an example of f9 as f9.jpg

Can somebody help me do this task?

Thank you.

Answer 1

[Constantino Carlos Reyes-Aldasoro]

Dear Krishna

I have seen several of your questions, I think that you would greatly benefit if you would study some Matlab and Image Processing before you try to tackle your problems, which require you to know some basics like labelling, dilation, erosion, etc. There are many websites and books that can teach you these and a week of study would reap huge benefits later on. I have written one book on the topic, but there are many others, the classic textbooks of Gonzalez & Woods, or Sonka are very good.

This is the book I mentioned:

https://onlinelibrary.wiley.com/doi/book/10.1002/9781118657546

[Krishna Chaitanya]

Thank you for the book! I would definitely study it this weekend,but I need to complete this part urgently as there are many other tasks depending on it.

It would be a great help if you kindly help me with this task.

Just a pseudocode with the necessary methods would be a great help.

Thank you.

Answer 2

[Constantino Carlos Reyes-Aldasoro]

Dilation is rather simple, but perhaps there are more steps to what you want:

f9_dil = imdilate(f9,strel('diamond',3));