Presentation is loading. Please wait.

Presentation is loading. Please wait.

Digital Image Processing

Similar presentations


Presentation on theme: "Digital Image Processing"— Presentation transcript:

1 Digital Image Processing
Image Enhancement Part II

2 Intensity Histogram

3 Intensity Histogram Example a 4x4, 4bits/pixel image 

4 From [Gonzalez & Woods]
Intensity Histogram From [Gonzalez & Woods]

5 From [Gonzalez & Woods]
Intensity Histogram From [Gonzalez & Woods]

6 Fixed Intensity Transformations

7 Basic Transformations
Negative: Log: Inverse Log: Power-law: …… From [Gonzalez & Woods]

8 Negative Transformation
From [Gonzalez & Woods]

9 From [Gonzalez & Woods]
Log Transformation From [Gonzalez & Woods]

10 Power-law (Gamma) Transformation
From [Gonzalez & Woods]

11 Power-law (Gamma) Transformation
From [Gonzalez & Woods]

12 Power-law (Gamma) Transformation
From [Gonzalez & Woods]

13 From [Gonzalez & Woods]
Thresholding m : threshold From [Gonzalez & Woods]

14 Example: Fixed Intensity Transformation
A 4x4, 4bits/pixel image passes through an intensity transformation 1  round(0.0667) = 0; 3  round(0.6) = 1; 6  round(2.4) = 2; 7  round(3.2667) = 3; 8  round(4.2667) = 4; 9  round(5.4) =5; 10  round(6.6667) = 7; 11  round(8.0667) = 8; The resulting image is:

15 Example: Histogram Change

16 Contrast Stretch

17 General Idea: Make Best Use of the Dynamic Range
From [Gonzalez & Woods]

18 Contrast Stretch General form:
Special case  Full-scale contrast stretch: Typically used:

19 Example: Full-Scale Contrast Stretch
Full-scale contrast stretch of a 4x4, 4bits/pixel image Find 4  round(0) = 0; 6  round(4.29) = 4; 7  round(6.43) = 6; 8  round(8.57) = 9; 9  round(10.71) = 11; 10  round(12.86) = 13; 11  round(15) = 15; The resulting image is:

20 Example: Histogram Change

21 Histogram Equalization

22 Histogram Equalization
From [Gonzalez & Woods]

23 Histogram Equalization
From [Gonzalez & Woods]

24 Example A 4x4, 4bits/pixel image
First try: full-scale contrast stretch 2  round(0) = 0; 3  round(1.67) = 2; 8  round(10.00) = 10; 9  round(11.67) = 12; 10  round(13.33) = 13; 11  round(15) = 15; The resulting image is:

25 Example: Histogram Change
full-scale contrast stretch original big gap

26 Cumulative Histogram H(k) Q(k)

27 Intermediate Image intermediate image original

28 Full-Scale Contrast Stretch of Intermediate Image
2  round(0) = 0; 6  round(4.29) = 4; 9  round(7.50) = 8; 12  round(10.71) = 11; 14  round(12.86) = 13; 16  round(15) = 15; final result: histogram equalized image

29 direct full-scale contrast stretch
Histogram Comparison original direct full-scale contrast stretch histogram-equalized more equalized

30 Summary of the Histogram Equalization Algorithm
original image histogram H(k) cumulative histogram Q(k) intermediate image full-scale contrast stretch histogram-equalized image


Download ppt "Digital Image Processing"

Similar presentations


Ads by Google