|
That the image is multi-channel is shown in the first screen, yes. You can use the Channel transform to extract one of the channels into a separate image and run watersheds on that. That is: select the Channel transform shown on your first screen, leave the Channel parameter at 0, press Add Component at the bottom of the Transform pane, switch to the Project pane, open the created image, use the Style pane to set up formatting if you have trouble seeing the data (switch to the Style pane, the image should be using the single channel for all of R, G, B, select these R, G and B in the list of channels by Ctrl-clicking, then right-click the selected channels and select Full Range in the context menu), you should now be able to use the watershed transforms. One caveat: extracting channel 0 is just a guess, your original image can be just a picture showing heights without actual height data, you should consult what the image really contains.
|