THRESHOLDING AND REGION
BASED SEGMENTATION
2 REFERENCES
• Gonzalez and Woods, Digital Image Processing, Pearson Education,3/e,
(2008)
• The images in the presentation are taken from slides of Gonzales and
Woods and different internet websites.
26-Thresholding and Region based Segmentation 7:13 PM
3 OUTCOMES
• Thresholding Region based Segmentation
26-Thresholding and Region based Segmentation 7:13 PM
4 THRESHOLDING
When T is constant applicable to entire image,
Global thresholding
26-Thresholding and Region based Segmentation 7:13 PM
5 THRESHOLDING
The result of thresholding is a binary image,
pixels with intensity value of 1
correspond to objects,
pixels with value 0
correspond to the background.
When T varies over the image, Variable thresholding
If T depends on the properties in the neighbourhood of (x,y)
then called local or regional thresholding
26-Thresholding and Region based Segmentation 7:13 PM
6 THRESHOLDING
26-Thresholding and Region based Segmentation 7:13 PM
7 THRESHOLDING
7:13 PM
26-Thresholding and Region based Segmentation
8 THRESHOLDING
26-Thresholding and Region based Segmentation 7:13 PM
9 THRESHOLDING
26-Thresholding and Region based Segmentation 7:13 PM
10 THRESHOLDING
26-Thresholding and Region based Segmentation 7:13 PM
11 ROLE OF NOISE IN THRESHOLDING
7:13 PM
26-Thresholding and Region based Segmentation
12 REGION BASED SEGMENTATION
Find the regions directly
26-Thresholding and Region based Segmentation 7:13 PM
13 REGION BASED SEGMENTATION
Region growing
** groups pixels or sub regions
** based on predefined criteria for growth
26-Thresholding and Region based Segmentation 7:13 PM
14 REGION BASED SEGMENTATION
Region growing
** start with a set of seed points
** grow regions by appending neighbouring pixels that have predefined
properties similar to the seed
26-Thresholding and Region based Segmentation 7:13 PM
15 REGION BASED SEGMENTATION
Let R represent the entire image region
Segmentation : Partitioning R into n subgroups
26-Thresholding and Region based Segmentation 7:13 PM
16 Region Growing by Pixel Aggregation
0 0 5 6 7
1 1 5 8 7
0 1 6 7 7
2 0 7 6 6
0 1 5 6 5
Seed points (3,2) and (3,4)
Absolute diff between seed point Results in two regions
and pixel is less than threshold
Threshold=3
26-Thresholding and Region based Segmentation 7:13 PM
Region
17
Growing by Pixel Aggregation
0 0 5 6 7
1 1 5 8 7
0 1 6 7 7
2 0 7 6 6
0 1 5 6 5
Threshold=8 Results in a single region
26-Thresholding and Region based Segmentation 7:13 PM
18 Region Growing by Pixel Aggregation
Difficulties in region growing
1. Selecting the initial seeds that properly represents the ROI
2. Selection of suitable properties
3. Formulation of a stopping rule
26-Thresholding and Region based Segmentation 7:13 PM
19 Region Growing
26-Thresholding and Region based Segmentation 7:13 PM
20 Region Splitting and Merging
26-Thresholding and Region based Segmentation 7:13 PM
21 Region Splitting and Merging
Example
26-Thresholding and Region based Segmentation 7:13 PM
22 Region Splitting and Merging
26-Thresholding and Region based Segmentation 7:13 PM
23 Region Splitting and Merging
26-Thresholding and Region based Segmentation
7:13 PM
24 Region Splitting and Merging
26-Thresholding and Region based Segmentation
7:13 PM
25 Region Splitting and Merging
Example
7:13 PM
26-Thresholding and Region based Segmentation
26 Region Splitting and Merging
Example
7:13 PM
26-Thresholding and Region based Segmentation
27 CONCLUSIONS
• Thresholding
• Region based segmentation
26-Thresholding and Region based Segmentation 7:13 PM
THANK YOU
