This is a collection of tools for segmenting images of cells within MATLAB.
- Copy this entire directory to your matlab path.
- Run LoadCellImages.m
- Follow the prompts to load file and crop images.
- After loading images, the image segmenter (ProcessCellData.m) is automatically started
The image segmenter was originally designed to analyze cell footprints imaged under RICM. Image segmenting is done using 3 main parameters:
-
Ilow: Pixels below this value are included in the segmented image
-
Ihigh: Pixels above this value are included
-
StdLim: Pixels that are part of a 3x3 neighborhood with a standard deviation greater than this value are included
Other Options:
- Fill Holes: Fills in holes in the segmented image. Holes must be completely surrounded by included pixels
- LargestOnly: keeps only the largest area
- Blur: Binary segmented image is convolved with a gausian having width="Blur" number of pixels. This has the effect of smoothing the edge of the segmented image.
- Fill Holes (Gray): applies the fill operation after bluring the image.
- Levelset Value: value (from 0 to 1) above which the blured image image is thersholded.
- Minimum Size: minimum size a region of included pixels must be in order to be included in final segmented image.