Scribd
Upload
333 views

Skeletons in Images

Skeletons define the structure of an object within a binary image by reducing it to its core lines or "wireframe". The skeleton is computed through iterative erosions and openings of the image, where each iteration creates a subset. The final skeleton is the union of all subsets, with each subset equaling the erosion minus the opening at each level. The algorithm performs erosions until no pixels remain, extracting the medial axis or centerlines of the object.

Uploaded by

Gabriel Humpire
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
333 views

Skeletons in Images

Skeletons define the structure of an object within a binary image by reducing it to its core lines or "wireframe". The skeleton is computed through iterative erosions and openings of the image, where each iteration creates a subset. The final skeleton is the union of all subsets, with each subset equaling the erosion minus the opening at each level. The algorithm performs erosions until no pixels remain, extracting the medial axis or centerlines of the object.

Uploaded by

Gabriel Humpire
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Skeletons

Presentation By Adam Metro


CIS 467 – Image Processing
What are Skeletons?
„ Morphological Operation
„ Skeletons define the “Structure” of an
object within an image.
„ Can also be considered the “Wire Frame”
of the object.
„ Only works with Binary Images (to my
knowledge), use Threshold function to
convert.
What are Skeletons?
What are Skeletons Con’t.
„ The Skeleton S(A) of a set A can be
expressed in terms of a series of erosions
and openings.
„ Each iteration k of erosions and openings
creates a subset Sk(A)
Lets Refresh Your Memory
„ Erosion
„ Removal of pixels from edges objects within an image A.
„ Uses a structuring element B to define custom erosion.
Lets Refresh Your Memory
„ Opening
„ Opening smoothes the outer edges of an object
within an image.
„ Opening is defined as applying Erosion on an
Image A with a Structuring Element B then
applying Dilation on the resulting erosion.
Back to Skeletons
„ Formal Definition of a K
skeleton of an object S ( A) = U S k ( A)
S k ( A) = ( AΘkB) − ( AΘkB) • B
within an image A: k =0

„ Formal Definition of a
Subset k of a skeleton
of an object within an Sk ( A) = ( AΘkB) − ( AΘkB) • B
image A with
structuring element B:
In English Please…
„ K equals the level of erosions performed on the
image before opening is performed.
S ( A)
k

„ Each subset equals the result of the erosion minus


the result of the opening.
„ The final skeleton will be the union of the Kmax
subsets found.
„ Kmax equals the maximum number of S ( A)erosions
k

before no more pixels exist in the object.


Example
My Project
„ Take in binary image.
„ Perform skeletonization on the image.
„ Return and display the resulting image.
„ Will use matlab functions imerosion() and
imdialate() to perform the various erosions
and openings.
„ Will also use image subtraction function
used in a previous homework assignment.
Questions?

You might also like