Texture Featuring Based on Watershed Karsten Rodenacker, Uta Jütting GSF-Institut für Biomathematik und Biometrie Neuherberg 11.05.98
Download ReportTranscript Texture Featuring Based on Watershed Karsten Rodenacker, Uta Jütting GSF-Institut für Biomathematik und Biometrie Neuherberg 11.05.98
Slide 1
Texture Featuring Based on
Watershed
Karsten Rodenacker, Uta Jütting
GSF-Institut für Biomathematik und Biometrie
Neuherberg
11.05.98
Slide 2
Content
Introduction
Some remarks about watershed
Textural featuring
Examples
Summary
Slide 3
Introduction
Textures
Slide 4
Some remarks about watershed
Definition
Given
an image and a local maximum at point
m. Consider all monotonous decreasing paths
w(m,l) starting from m and ending in l.
The set of all points l is a connected region.
The set of all such regions is a partition of the
input image, the order corresponds with the
number of local maximums.
The borders of the regions are the watersheds
Slide 5
Some remarks about watershed
Definition (continued)
The
inversion of the above definition
concerning maximum by minimum and
decreasing by increasing delivers a second type
of watershed.
Slide 6
Some remarks about watershed
Explanation
1-dimensional
Slide 7
Some remarks about watershed
Explanation
2-dimensional
Slide 8
Some remarks about watershed
Explanation
2-dimensional
Slide 9
Some remarks about watershed
Result of transformation
Input
image (original, smoothed 3x3)
Slide 10
Some remarks about watershed
Result of transformation
Watersheds
(labelled regions)
Slide 11
Some remarks about watershed
Result of transformation
Watersheds
Slide 12
Some remarks about watershed
Result of transformation
Lower,
upper ricefield
Slide 13
Some remarks about watershed
Result of transformation
Half
height segmentation HU (black)/HL (white)
Slide 14
Textural featuring
global
Statistical
estimators (moments) from:
– half height segmentations and original
– topological gradient
Derived
parameters
– stereological (volume densities)
– densitometric
– morphometric
Slide 15
Textural featuring
global, e.g.
V VU
P AU
HUA
volume density
A
HUA
mean particle area
HUNO
E HUNO HLNO
RDD
H U M 1 H LM 1
M1
Euler number
Rel. density difference
Slide 16
Textural featuring
heuristical
neighborhood
related of connected regions
(graph theoretical approach)
– connectivity
– distances
neighborhood AND
– pattern matching
...
intensity related
Slide 17
Examples
global texture features:
VVU
Volume density
RDD Relative density difference
Slide 18
Examples
Projects
Differentiation
of hormone status of breast
carcinoma patients
VVU, RDD
Differentiation of osteoblasts under different
growth conditions (Osteoporosis)
RHUA
Differentiation of neuroendocrine lung tumours
HUCV
Slide 19
Slide 20
Examples
global texture features: RDD
Slide 21
Examples
Slide 22
Examples
Projects
Slide 23
Summary
+ Topology by watershed
Taxonomy
of transformation results
+ Segmentation
+ Parameter free featuring
- Noise sensitivity of watershed
- Difficulties of interpretation for natural
images (reflected light, shadows, ...)
Texture Featuring Based on
Watershed
Karsten Rodenacker, Uta Jütting
GSF-Institut für Biomathematik und Biometrie
Neuherberg
11.05.98
Slide 2
Content
Introduction
Some remarks about watershed
Textural featuring
Examples
Summary
Slide 3
Introduction
Textures
Slide 4
Some remarks about watershed
Definition
Given
an image and a local maximum at point
m. Consider all monotonous decreasing paths
w(m,l) starting from m and ending in l.
The set of all points l is a connected region.
The set of all such regions is a partition of the
input image, the order corresponds with the
number of local maximums.
The borders of the regions are the watersheds
Slide 5
Some remarks about watershed
Definition (continued)
The
inversion of the above definition
concerning maximum by minimum and
decreasing by increasing delivers a second type
of watershed.
Slide 6
Some remarks about watershed
Explanation
1-dimensional
Slide 7
Some remarks about watershed
Explanation
2-dimensional
Slide 8
Some remarks about watershed
Explanation
2-dimensional
Slide 9
Some remarks about watershed
Result of transformation
Input
image (original, smoothed 3x3)
Slide 10
Some remarks about watershed
Result of transformation
Watersheds
(labelled regions)
Slide 11
Some remarks about watershed
Result of transformation
Watersheds
Slide 12
Some remarks about watershed
Result of transformation
Lower,
upper ricefield
Slide 13
Some remarks about watershed
Result of transformation
Half
height segmentation HU (black)/HL (white)
Slide 14
Textural featuring
global
Statistical
estimators (moments) from:
– half height segmentations and original
– topological gradient
Derived
parameters
– stereological (volume densities)
– densitometric
– morphometric
Slide 15
Textural featuring
global, e.g.
V VU
P AU
HUA
volume density
A
HUA
mean particle area
HUNO
E HUNO HLNO
RDD
H U M 1 H LM 1
M1
Euler number
Rel. density difference
Slide 16
Textural featuring
heuristical
neighborhood
related of connected regions
(graph theoretical approach)
– connectivity
– distances
neighborhood AND
– pattern matching
...
intensity related
Slide 17
Examples
global texture features:
VVU
Volume density
RDD Relative density difference
Slide 18
Examples
Projects
Differentiation
of hormone status of breast
carcinoma patients
VVU, RDD
Differentiation of osteoblasts under different
growth conditions (Osteoporosis)
RHUA
Differentiation of neuroendocrine lung tumours
HUCV
Slide 19
Slide 20
Examples
global texture features: RDD
Slide 21
Examples
Slide 22
Examples
Projects
Slide 23
Summary
+ Topology by watershed
Taxonomy
of transformation results
+ Segmentation
+ Parameter free featuring
- Noise sensitivity of watershed
- Difficulties of interpretation for natural
images (reflected light, shadows, ...)