Prune

Prune barbs off a skeletonized image and turn a skeletonized image into separate pieces.

plantcv.morphology.prune(skel_img, size=0, mask=None)

returns Pruned skeleton image, segmented image, segment objects

  • Parameters:
    • skel_img - Skeleton image (output from plantcv.morphology.skeletonize)
    • size - Pieces of skeleton smaller than size should get removed.(Optional) Default size=0.
    • mask - Binary mask for debugging (optional). If provided, debug images will be overlaid on the mask.
  • Context:
    • This "prunes" spurious branches/barbs off a skeleton. The function prunes barbs that are size pixels or smaller from a skeleton image. If the default size=0 is used, the pruned skeleton will be identical to the input skeleton image.

from plantcv import plantcv as pcv

# Set global debug behavior to None (default), "print" (to file), 
# or "plot" (Jupyter Notebooks or X11)
pcv.params.debug = "plot"

pruned_skeleton, segmented_img, segment_objects = pcv.morphology.prune(skel_img=skeleton, size=70)

Skeleton before pruning

Screenshot

Pruned Skeleton (image getting returned)

Screenshot

Debugging Image

Screenshot

Segmented Skeleton (image getting returned)

Screenshot


from plantcv import plantcv as pcv

# Set global debug behavior to None (default), "print" (to file), 
# or "plot" (Jupyter Notebooks or X11)
pcv.params.debug = "plot"

# The image created for debugging purposes allows for line thickness 
# adjustments with the global line thickness parameter. Try setting 
# pcv.params.line_thickness = 3 for thinner lines (default 5)
pcv.params.line_thickness = 3 

pruned_skeleton, segmented_img, segment_objects = pcv.morphology.prune(skel_img=skeleton, 
                                                                       size=70, 
                                                                       mask=plant_mask)

Pruned Skeleton (image getting returned)

Screenshot

Debugging Image with Mask

Screenshot

Segmented Skeleton with Mask (image getting returned)

Screenshot

Source Code: Here