Input: Colour/ grayscale image
1 Read input image
2 Construct M×N = n image cells
3 Perform statistical analysis of the image cells
Determine whether ith image cell required
segmentation
if > threshold value
proceed to local normalised cuts segmentation
else
4
compute background node in ith image cell
5 First stage segmentation
Normalised cuts local segmentation into k1 segments
on ith image cell until nth image cell.
Nodes computation
Each of the segments from every image cell is
6
represented by a node.
7 Second stage segmentation
Perform normalised cuts segmentation based on the
computed nodes to k2 segments.
Output:
Produce segmented segments and display result