"MATLAB" image thinning algorithm

Refinement algorithm

Image thinning, which generally refers to a manipulation operation of the skeleton (image skeletonization) of a two-value image.

The so-called refinement is after a layer of stripping, from the original picture to remove some points, but still have to maintain the original shape, until the skeleton of the image is obtained. Skeleton, which can be understood as the middle axis of the image.

Evaluation index:

• Convergence of
• Ensure the connectivity of thin lines after refinement
• Keep the basic shape of the original
• Reduces distortion at the intersection of strokes
• The refinement result is the center line of the original image
• Reduced speed and number of iterations

which

• Non-iterative algorithm:

The skeleton is generated once, such as the method based on distance transformation, length coding refinement, etc.

• Iterative algorithm:

  Repeat delete the image edge to meet certain conditions of the pixel, and finally get the single-pixel broadband skeleton

• Serial algorithm

  Whether to delete pixels is a fixed order in the execution of each iteration, it depends not only on the results of the previous iteration, but also on the distribution of pixel points that have been processed in this iteration

• Parallel algorithms

  Whether pixel deletion or not is independent of the order in the pixel value image, depends only on the result of the previous iteration

Effect

"MATLAB" image thinning algorithm