K-mean & watershed segmentation
3 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Hello all.I am new to Matlab software. I am trying to implement a project with the following steps: K-means & watershed step1: k-means :- primary segmentation step2: Euclidean distance :- calculate for each data point to the mean of each cluster. step3: Gradient magnitude :- using the primary segmentation apply sobel filter. step4: Automatic thresholding:- automatic thresholding perform on gradient magnitude(i.e based on the histogram of the normalized gradient magnitude. step5: Apply rainfall simulation:- Is applied on the edge map. step6: Steepest descent:- steepest descent of rainfall is implemented using 3x3 window centred on each pixel of the gradient map. step7: Compute steepest gradient:- The gradient direction from 8-connected neighbouring pixels.(the neighbouring pixel along steepest direction is marked and the window shifted along the direction). Note: The process of marking pixels and shifting window is repeated until the path reaches a minimum. the pixel constituting the part adopt the label of that minimum.Repeat the rainfall simulation by tracing a path of steepest descent for all pixels that are unlabelled. the paths reaching a common minimum adopt the same label as that minimum, and constitute a catchment basin, which refers to a partition in the image . this partition constitute the initial segmentation. step8: implement post-segmentation merging on the improved watershed algorithm. (the post segmentation merging is based on spatial criteria, this is to reduce the number of partitions without affecting the accuracy of the segmentation map.
Thank you.
2 commentaires
the cyclist
le 10 Sep 2013
Josh,
If you look at what you posted, you might notice two things:
- The formatting makes it clear that you lazily cut & paste your assignment in a way that makes a dense block of text that is painful to read.
- You have not actually asked a question.
If you expect some help from this volunteer corps of people, please put some effort it on your side. Make a more readable question. Ask a specific question. Post some code that you tried.
We won't just do your homework for you from start to finish.
Réponses (0)
Voir également
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!