i am having an issue with my morphological segmentation, i have tried researching months on how to do a smooth segmentation and quantify my objects but to no avail, can you please make some recommendations, especially pertaining to the contours that i am having problems with in my image.
To start @reubensey - it’s best to never use the JPEG file format for saving/processing images for quantitative analysis of any kind. You can read more about why JPEGs are not a good idea - ever (follow the link). So be sure to save/process your images as TIFFs or the proprietary file format provided by your system… and use Bio-Formats to import your images.
But as a quick test… I just ran a quick Gaussian blur (2 pixel radius), Minimum filter (2 pixel radius), and then Variance (2 pixels again)… then running the Morphological Segmentation plugin with MorphoLibJ I got decent segmentation:
Obviously - this can be refined… but gives you a start. This was also under the assumption that you want to segment those lighter-colored blobs.
thanks for the education and the quick response. I will avoid the Jpeg next time. However, you will realize that, the contours are very challenging as displayed in the image, considering the wonderful results displayed in this piece, “http://www1.metalurji.org.tr/immc2016/459.pdf” execution i can say was 100%. that’s what i am trying to achieve best. Nonetheless, thanks. any other information will be very much appreciated. Attached are two images from the link i mentioned.