Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Image Segmentation
Group similar components (such as, pixels in an
image, image frames in a video)
Applications: Finding tumors, veins, etc. in medical
images, finding targets in satellite/aerial images,
finding people in surveillance images, summarizing
video, etc.
Similarity
Thresholding, region growing, and region splitting/merging.
1 if f ( x, y ) T
g ( x, y )
0 if f ( x, y ) T
Global Thresholding
Local/Adaptive Thresholding
1/5/2020 Image Segmentation 5
Global Thresholding
Single threshold value for entire image
Fixed ?
Automatic
Intensity histogram
Multilevel thresholding
T mean
T median
max min
T
2
P( Ri ) true
P( Ri R j ) True
Region Splitting/Merging
Stop when no further split or merge is possible
24
Classification
25
Classification
26
Classification
27
Clustering
There are two types of fruit in the basket,
separate them into two ‘groups’
28
Clustering
Groups - Clusters
29
Classification vs. Clustering
Category “A”
Category “B”
Clustering
Classification
30
Pattern Class
31
Pattern Class
Intra-class variability
Inter-class similarity
Identical twins
33
K-Means Clustering
D. Comaniciu and P.
Meer, Robust Analysis
of Feature Spaces:
Color Image
Segmentation, 1997.