Image connected components
Objectives
- Recognize the difference between different types of connected components (or “blobs” as they are often called)
- Explain what a labeled image is
- List some of the properties that can be used to describe a connected component
- Recognize that blobs can be filtered according to their properties
- Recall that the watershed algorithm can be used to separate touching components
Accessing files
See GitHub for instructions on how to download the course material to your local computer.
The source files for this lecture are at <repo>/lectures/image_conncted_components
Other useful material
Core materials
- Lecture videos
- PowerPoint file
Image processing connected components
A video explaining different types of connected components.
A demonstration of creating a labeled image.
A third video which shows how to extract blob properties (in this case their size and centroid).
A demonstration of filtering blobs by size, clearning the image boundary, and identifying the largest blob in an image.
A final video showing how to apply a watershed algorithm to separate a blob into two separate regions.