My data-set contains a variety of noise. One noise is in the form of different shapes(circles) of different sizes and colors at different locations. How can I remove them? I have tried colour-based segmentation and thresholding but it also detects my area of interest. I also tried using shape (circle) detection but it is also not working as these are not full circles. How can I detect half circles/ellipse?
Please give me some guidelines or some hints/code (in matlab) from where to start. Please see the attached image. I will be really thankful
Your images are corrupted by other factors, not noise. I can see the image quality is good enough, there are a little noise. In order to remove the shape that you are talking, I also recommend you use inpainting algorithms (as Nek Valous suggested above).
Here is what I would do:
Hope that helps!
Also, try some image inpainting: https://github.com/chongyangtao/Color-Image-Inpainting
What is information and what is noise in Your case? Do You mean that dark brown color circles are noise (if yes, I prefer to say artifacts)? This is the main initial step, then You or other will be able to find a proper solution.
- Badges
- Science topic
- Similar topics
- Computer Science
- Images