9 morphological image processing.ppt [相容模式]

34
9 MORPHOLOGICAL IMAGE PROCESSING Chih-Wei Tang Department of Communication Engineering National Central University JhongLi, Taiwan

Upload: others

Post on 10-Feb-2022

15 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: 9 Morphological Image Processing.ppt [相容模式]

9 MORPHOLOGICAL IMAGE PROCESSING

Chih-Wei TangDepartment of Communication Engineering

National Central UniversityJhongLi, Taiwan

Page 2: 9 Morphological Image Processing.ppt [相容模式]

Outline• Preliminaries• Dilation and erosion• Opening and closing• Some basic morphological algorithms• Extensions to gray-scale images

C.E., NCU, Taiwan 2

Page 3: 9 Morphological Image Processing.ppt [相容模式]

Basic Concepts from Set Theory

C.E., NCU, Taiwan 3

• Sets: objects in an image• Set theory: the language of mathematical

morphology

Page 4: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 4

Page 5: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 5

Page 6: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 6

Page 7: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 7

Page 8: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 8

Page 9: 9 Morphological Image Processing.ppt [相容模式]

Translation and Reflection

C.E., NCU, Taiwan 9

• They are extended from morphology but not found in basic texts on set theory.

Page 10: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 10

Outline

Preliminaries Dilation and erosion Opening and closing Some basic morphological algorithms Extensions to gray-scale images

Page 11: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 11Reflection

Translation

Page 12: 9 Morphological Image Processing.ppt [相容模式]

Use of Morphological Dilation for Bridging Gaps

C.E., NCU, Taiwan 12

Page 13: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 13

Page 14: 9 Morphological Image Processing.ppt [相容模式]

Use of Morphological Erosion for Removing Image Components

C.E., NCU, Taiwan 14

Page 15: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 15

Outline

Preliminaries Dilation and erosion Opening and closing Some basic morphological algorithms Extensions to gray-scale images

Page 16: 9 Morphological Image Processing.ppt [相容模式]

Opening and Closing• Opening

• smooth the contour of an object• breaks narrow isthmuses• eliminates thin protrusions

• Closing• smooth sections of contours• generally fuses narrow breaks and long thin gulfs• eliminates small holes• fills gaps in the contour

C.E., NCU, Taiwan 16

Page 17: 9 Morphological Image Processing.ppt [相容模式]

Opening

C.E., NCU, Taiwan 17

Page 18: 9 Morphological Image Processing.ppt [相容模式]

Closing

C.E., NCU, Taiwan 18

Page 19: 9 Morphological Image Processing.ppt [相容模式]

Opening

Closing

C.E., NCU, Taiwan 19

Page 20: 9 Morphological Image Processing.ppt [相容模式]

Use of Opening and Closing for Morphological Filtering

C.E., NCU, Taiwan 20

Page 21: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 21

Outline

Preliminaries Dilation and erosion Opening and closing Some basic morphological algorithms Extensions to gray-scale images

Page 22: 9 Morphological Image Processing.ppt [相容模式]

Boundary Extraction

C.E., NCU, Taiwan 22

StructureElement

Erosion

Page 23: 9 Morphological Image Processing.ppt [相容模式]

Example – Boundary Extraction

C.E., NCU, Taiwan 23

Page 24: 9 Morphological Image Processing.ppt [相容模式]

Region Filling

C.E., NCU, Taiwan 24

Dilation

Complement

Page 25: 9 Morphological Image Processing.ppt [相容模式]

Example – Region Filling

C.E., NCU, Taiwan 25

Page 26: 9 Morphological Image Processing.ppt [相容模式]

Extraction of Connected Components

C.E., NCU, Taiwan 26

Converge!!!1 If kk XX

kXY Let

Page 27: 9 Morphological Image Processing.ppt [相容模式]

C.E., NCU, Taiwan 27

Outline

Preliminaries Dilation and erosion Opening and closing Some basic morphological algorithms Extensions to gray-scale images

Page 28: 9 Morphological Image Processing.ppt [相容模式]

2-D Dilation

C.E., NCU, Taiwan 28

• General effects• If all the values of the structuring element

are positive, the output image tends to be brighter than the input.

• Dark details either are reduced or eliminated, depending on how their values and shapes relate to the structuring element used for dilation.

Page 29: 9 Morphological Image Processing.ppt [相容模式]

1-D Dilation

C.E., NCU, Taiwan 29

Correct?

Page 30: 9 Morphological Image Processing.ppt [相容模式]

2-D Erosion

• General effects• If all the elements of the structuring element

are positive, the output image tends to be darker than the input image.

• The effect of bright details in the input image that are smaller in area than the structuring element is reduced.

C.E., NCU, Taiwan 30

Page 31: 9 Morphological Image Processing.ppt [相容模式]

1-D Erosion

C.E., NCU, Taiwan 31

Correct?

Page 32: 9 Morphological Image Processing.ppt [相容模式]

Dilation and Erosion on A Gray-Scale Image

C.E., NCU, Taiwan 32

Page 33: 9 Morphological Image Processing.ppt [相容模式]

Gray-Scale Opening and Closing

C.E., NCU, Taiwan 33

Page 34: 9 Morphological Image Processing.ppt [相容模式]

Gray-Scale Opening and Closing –Example

C.E., NCU, Taiwan 34