Well  Prof Giuliani , the eternal local vs global optimization for # of segments is what I can think of.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4294284/

Adapting some combinatorial techniques

I would suggest the following survey for deformable models segmentation:

Pablo Mesejo, Oscar Ib´a˜nez, Oscar Cord´on, Stefano Cagnoni. A Survey on Image Segmentation using Metaheuristic-based Deformable Models: State of the Art and Critical Analysis. Applied Soft Computing, Elsevier, 2016, 44, pp.1-29. 

https://hal.archives-ouvertes.fr/hal-01282678/document

The following slide deck is an overview of the use of metaheuristics in image processing

Daniela Zaharie, Bio-inspired Metaheuristics and Image Processing, SSIP - Summer School on Image Processing, Szeged, Hungary, 2016.

http://www.inf.u-szeged.hu/~ssip/2016/wp-content/uploads/2016/07/Zaharie_ssip2016.pdf

Thanks a lot. I obtained promising results with FA algorithm for segmentation of grayscale images, I show you an example in the included file

I would like to extend the method for colour images

Regards Donatella

Dear Prof Giuliani,

The below reference is very similar to your work aim.

http://www.sciencedirect.com/science/article/pii/S1877050915001283.

Although,I work in brain image segmentation and trying to present new hybrid approach using FCM and metaheuristics algorithms; I work in grayscale images.

Best regards

thanks so much!

I would like to add one thing is depending on the application and the properties of input images, the selection of color space can be crucial for separating unique colors under varied light conditions. While working with brainbow color images and underwater images for long time, it had been my impression for getting better segmentation. Here is a paper that can help us selecting the appropriate color space for representing varied types of color images 

Khan, Rahat, et al. "Discriminative color descriptors." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2013.

http://ieeexplore.ieee.org/abstract/document/6619213/?reload=true

Dear Prof Giuliani,

Few of the best suitable papers for image segmentation using metaheuristic algorithms are listed as:

1. http://waset.org/publications/9999738/nature-inspired-metaheuristic-algorithms-for-multilevel-thresholding-image-segmentation-a-survey

2. http://elcvia.cvc.uab.es/article/view/567

3. http://www.sciencedirect.com/science/article/pii/S2352146517301989

Thanks a lot for your suggestion and helpful considerations

Regards

Donatella