and in the input image C ONTRAST E NHANCEMENT BASED ON L AYERED D IFFERENCE R EPRESENTATION Chulwoo Lee Chul Lee and ChangSu Kim wiserain kayne changsukim ID: 795590
Download The PPT/PDF document "2D histogram The number of pairs of adja..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
2D histogramThe number of pairs of adjacent pixels with value and in the input image.
C
ONTRAST
E
NHANCEMENT BASED ON
L
AYERED
D
IFFERENCE
R
EPRESENTATION
Chulwoo Lee,
Chul
Lee, and Chang-Su Kim
{
wiserain
,
kayne
,
changsukim
} @ korea.ac.kr
School of Electrical Engineering, Korea University , Seoul, Korea
Overview
Basic Concept
Intra-Layer Optimization
Elementary Properties
[1]
Q. Wang
and R. K. Ward, “Fast Image/Video Contrast Enhancement Based on Weighted
Thresholded
Histogram Equalization,”
IEEE Trans.
Consum
. Electron.
,
vol. 53, no. 2, pp. 757–764, 2007.
[2]
T.
Arici
, S.
Dikbas
, and Y.
Altunbasak
, “A Histogram Modification Framework and Its Application for Image Contrast Enhancement,”
IEEE Trans. Image Process.
,
vol. 18, no. 9, pp. 1921–1935, 2009.
Experimental Results
(a) Original
(b) WTHE
[1]
(c) WAHE
[2]
(d) Proposed
Input
WTHE
[1]
WAHE
[2]
Proposed
AMBE
-
6.482
7.647
5.208
DE
2.154
2.144
2.135
2.113
EME
9.878
9.485
9.265
12.811
PixDist
26.315
33.164
32.694
34.125
Proportional relationship
: Normalizing constant
Estimation of
By summing up
for all possible
’s,
At layer 1
At layer 2
At layer
By
, we have a linear equation,
Global contrast enhancement
Gray-level difference
Proposed approach
Conventional approach
Gray-level intensity
Relationship between
and
for
Relationship between
and
Inter-Layer Aggregation
First, we need to smooth the obtained by the Gaussian kernel
: user-controllable parameter, e.g.
For example,
A large
Frequent gray-level pairs
Input image
Increase
difference variable
Image contrast is
enhanced
Output image
Constrained optimization problem
Solution to the problem
Quadratic cost function
The active-set method
Then, we aggregate by the weighted averaging process
Objective assessment
Subjective assessment