Click here to Skip to main content
15,891,372 members
Please Sign up or sign in to vote.
4.50/5 (2 votes)
See more:
Hi,
I am working on a C# project which requires recognition of objects inside an image file. The image will be gray-scale. The objects will be usually some small shapes like circle, rectangle, arrows, semi-circle, and so on. In my problem domain, those objects will be of very small size like only 0.1% to 1% size of the whole image. The difficult part is those objects can be situated on top of some bigger objects. Which means there can be some lines crossing these objects. I must still be able to recognize them without those blocking or touching lines.

So in general I need to select an object using some kind of magic wand like algorithm and add or remove parts from it to refine the object selection. Any help or guide on how to proceed or where to research on or where to get similar algorithms will be helpful. I am totally new to image processing. I worked on many C# projects but this is most challenging to me.
Posted

Try using Matlab for reference.You can find a lot of functions and examples related to image/video analysis. You can then use those principles and implement them in your own app. I used to do stuff like that during university, but back then I had more pressing matters.
 
Share this answer
 
Hi
If an object can be visually distinguish from its surrounding then its edge and the surrounding might be different. What you need to do is trial with different edge detecting algorithms and choose the best works for you. I have seen such algorithms separating blood cell types from a smear micro photograph using edge detecting and shape recognization algorithms
 
Share this answer
 
Hope this[^]article might give you an idea.
 
Share this answer
 

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)



CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900