16-02-2013, 04:43 PM
Video Contrast Enhancement Based on WTHE
[attachment=51239]
Abstract—
We present a fast and effective method for image
contrast enhancement based on weighted and thresholded
histogram equalization (WTHE). In our proposed method, the
probability distribution function of an image is modified by
weighting and thresholding before the histogram equalization
(HE) is performed. We show that such an approach provides a
convenient and effective mechanism to control the enhancement
process while being adaptive to various types of images. We also
discuss applications of the proposed method in video
enhancement.
Keywords—contrast enhancement; histogram; histogram
equalization; cumulative distribution function
Topic area—Multimedia processing: image post-processing
I. INTRODUCTION
Histogram equalization (HE) is widely used for image
contrast enhancement. Applications of histogram equalization
are found in many areas such as medical image processing,
speech recognition and texture synthesis. Also in recent years,
the application of the HE methods in video enhancement is
drawing much interest. It is known, however, that the
traditional HE method suffers from the following drawbacks:
(1) It lacks of a mechanism to adjust the degree of
enhancement.
(2) It often causes unpleasant visual artifacts, such as overenhancement,
level saturation and raised noise level.
(3) It could dramatically change the character of the image,
e.g., the average luminance (mean) of the image.
As a result of the above shortcomings, histogram
equalization is rarely used in its original form. Variants of the
HE method have been developed by researchers. In recent
years, many improved, HE-based enhancement techniques are
proposed, such as [1 – 8].
The recently proposed enhancement methods generally
belong to two categories: the adaptive (or local) HE methods
(AHE), such as [1,2,3], and the improved global methods
based on histogram equalization or specification, such as [4 -
8]. The AHE methods use statistical information in the
neighborhood of each pixel to perform equalization. In the
global HE-based methods, various constraints are added to the
equalization procedure to achieve better performance.
AHE methods can usually provide stronger enhancement
effects than global methods. However, due to their high
computational load, the AHE methods are usually unsuitable
for real time video enhancement. Therefore, in our research,
we focus more on the global techniques because of their
speed. In this paper, we propose a fast global HE-based
enhancement scheme. The proposed method provides
sufficiently enhanced images with significantly less artifacts,
and allows a convenient and effective control over the degree
and effect of enhancement.
Section II reviews the traditional HE method and some
recently proposed HE-based methods. In section III, we
propose a weighted thresholded HE (WTHE) enhancement
method and discuss its implementation for video
enhancement. Section IV shows experimental results together
with comparisons with some contemporary methods. Section
V draws conclusions and gives discussions.
[attachment=51239]
Abstract—
We present a fast and effective method for image
contrast enhancement based on weighted and thresholded
histogram equalization (WTHE). In our proposed method, the
probability distribution function of an image is modified by
weighting and thresholding before the histogram equalization
(HE) is performed. We show that such an approach provides a
convenient and effective mechanism to control the enhancement
process while being adaptive to various types of images. We also
discuss applications of the proposed method in video
enhancement.
Keywords—contrast enhancement; histogram; histogram
equalization; cumulative distribution function
Topic area—Multimedia processing: image post-processing
I. INTRODUCTION
Histogram equalization (HE) is widely used for image
contrast enhancement. Applications of histogram equalization
are found in many areas such as medical image processing,
speech recognition and texture synthesis. Also in recent years,
the application of the HE methods in video enhancement is
drawing much interest. It is known, however, that the
traditional HE method suffers from the following drawbacks:
(1) It lacks of a mechanism to adjust the degree of
enhancement.
(2) It often causes unpleasant visual artifacts, such as overenhancement,
level saturation and raised noise level.
(3) It could dramatically change the character of the image,
e.g., the average luminance (mean) of the image.
As a result of the above shortcomings, histogram
equalization is rarely used in its original form. Variants of the
HE method have been developed by researchers. In recent
years, many improved, HE-based enhancement techniques are
proposed, such as [1 – 8].
The recently proposed enhancement methods generally
belong to two categories: the adaptive (or local) HE methods
(AHE), such as [1,2,3], and the improved global methods
based on histogram equalization or specification, such as [4 -
8]. The AHE methods use statistical information in the
neighborhood of each pixel to perform equalization. In the
global HE-based methods, various constraints are added to the
equalization procedure to achieve better performance.
AHE methods can usually provide stronger enhancement
effects than global methods. However, due to their high
computational load, the AHE methods are usually unsuitable
for real time video enhancement. Therefore, in our research,
we focus more on the global techniques because of their
speed. In this paper, we propose a fast global HE-based
enhancement scheme. The proposed method provides
sufficiently enhanced images with significantly less artifacts,
and allows a convenient and effective control over the degree
and effect of enhancement.
Section II reviews the traditional HE method and some
recently proposed HE-based methods. In section III, we
propose a weighted thresholded HE (WTHE) enhancement
method and discuss its implementation for video
enhancement. Section IV shows experimental results together
with comparisons with some contemporary methods. Section
V draws conclusions and gives discussions.