Image processing segmentation
Objectives
- Recognize that segmentation can be used to identify features in an image
- Distinguish between global and regional thresholding
- Explain the algorithm underlying Otsu’s method for thresholding
- Use different color-spaces to simplify thresholding
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_processing_segmentation
Other useful material
- Steve on Image Processing
- Digital Image Processing, Gonzalez and Woods
- A blog-post on k-means segmentation
Core materials
- Lecture videos
- PowerPoint file
Image processing segmentation
A video explaining the basic concept of image segmentation.
A demonstration of finding pixel values with intensities above a certain threshold.
A third video which explains Otsu’s method for automatic thresholding and demonstrates adaptive background correction to allow for changes in overall brightness.
A video showing how to use MATLAB’s Color Segmenter App to explore whether different features are easier to segment in non RGB color spaces.