International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)

Underwater Image Segmentation using CLAHE

Enhancement and Thresholding
Rajesh kumar Rai1, Puran Gour2, Balvant Singh3
1, 2, 3

Department of Electronics & Communication

NRI Institute of Information Science & Technology Bhopal, INDIA
1

raj.rai1008@gmail.com,
purrangour@rediffmail.com,
3
balvant.er@gmail.com

The researchers have reviewed several techniques related

to images enhancement viz contrast stretching histogram
equalization [6] contrast limited adaptive histogram
equalization (CLAHE) [5]. Here we are going to compare
three enhancement techniques on the basis of SNR and
mean square error as parameter.
After enhancing the image quality segmentation is
performed to extract the desired objects. Segmentation is
typically associated with pattern recognition problems. It is
considered the first phase of a pattern recognition process
and is sometimes also referred to as object isolation as
shown in Fig. 1. In this paper histogram thresholding is
used for segmenting the background and objects in
underwater environment. Then performance of global, local
and multilevel thresholding is compared for different
underwater environments. It found that with the
combination of CLAHE method segmentation performs
well for most of the images.

Abstract - The objects in the underwater images are not

clearly visible due to low contrast and scattering of light and
the large noise present in the environment. Hence it is difficult
to segmentation in such environment without losing the details
of the objects. In this paper a method of segmentation is
presented for underwater images, in which the image quality
is first enhanced using contrast limited adaptive histogram
equalization method, and then histogram thresholding is used
to segment the objects. Paper discusses the comparative
analysis of various Histogram thresholding techniques. For
comparing the performance, mean square error and SNR
used as parameters. The method is tested on various type of
underwater image environment.
Keywords - Image enhancement, Contrast limited Adaptive
Histogram Equalization, image segmentation, Global
Thresholding, Multi level Thresholding

I.

INTRODUCTION

No segmentation technique is universally applicable,

which works equally well for all kinds of images.
Therefore, various segmentation approaches [1, 12, 15, and
16] have been developed which perform differently for
different applications as well for different types of images.
It is difficult to segmentation in underwater environment
without losing the details of the objects. Although the low
light in camera develops fast in recent years, and imaging
velocity and image resolution ratio are improved greatly.
But underwater images still have low contrast; unbalance
gray scales and fuzzy edge of objects under the influence of
the imaging condition and some character of water media.
In underwater environment image get blurred due to poor
visibility conditions and effects like absorption of light,
reflection of light, bending of light, denser medium
(800 times denser than air), and scattering of light etc.
hence it is required to enhance the quality of the underwater
images before processing. There is lot of research done for
the improvement of image quality as [4, 12, 13, and 14].

Fig.1 Basic of Segmentation

The remaining of the paper is organized as follows:

Next Section describe the various type of image
enhancement method and algorithm, section II describe the
contrast stretching histogram equalization techniques,
section III describe CLAHE method and its flow chart.
Section IV discusses the Segmentation methods and, In
Section V the results are compared. Section VI gives
conclusion and future work.

118

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)
II.

IMAGE ENHANCEMENT

Image enhancement is a process of improving the quality

of image by improving its features. Here we have given
three method of image enhancement for underwater images
(i) Contrast stretching (ii) Histogram equalization and
Contrast limited adaptive histogram equalization. A number
of contrast measures were proposed for complex images as
underwater images [1, 2, 3, and4].

Fig 2 Transformation for the histogram equalization

Using histogram equalization can be a good approach

when automatic enhancement is desired, although there are
still situations where basing image enhancement on a
uniform histogram may not be the best approach. In these
situations, histogram equalization effects may be too
severe. So other histogram techniques may need to be used,
such as adaptive histogram equalization.
In this paper, a novel Contrast limited adaptive
histogram equalization (CLAHE) enhancement method is
proposed which can yield the optimal equalization and also
limit the contrast of the image. The suggested method is
very useful for the video image broadcasting, where the
brightness requirement is high such as in geographical
channels.

In local Contrast Stretching method, the Contrast is

stretched between the limit of lower threshold and upper
threshold. It is an intensity based contrast enhancement
method as Io(x,y) = f(I(x,y)), where the original image is
I(x,y), the output image is Io(x,y) after contrast
enhancement, and f is the transformation function. Another
contrast based method is histogram equalization. Histogram
is defined as the statistic probability distribution of each
gray level in a digital image. Histogram equalization (HE)
is one of the well-known methods for enhancing the
contrast of given images, making the result image have a
uniform distribution of the gray levels. It flattens and
stretches the dynamic range of the images histogram and
results in overall contrast improvement. HE has been
widely applied when the image needs enhancement
however, it may significantly change the brightness of an
input image and cause problem in some applications where
brightness preservation is necessary.

III. CONTRAST L IMITED ADAPTIVE H ISTOGRAM

EQUALIZATION (C LAHE)
CLAHE was originally developed for medical imaging
and has proven to be successful for enhancement of lowcontrast images such as portal films. The CLAHE
algorithm partitions the images into contextual regions and
applies the histogram equalization to each one. This evens
out the distribution of used grey values and thus makes
hidden features of the image more visible. The full grey
spectrum is used to express the image.
Contrast Limited Adaptive Histogram Equalization,
(CLAHE) is an improved version of AHE, or Adaptive
Histogram Equalization. Both overcome the limitations of
standard histogram equalization. A variety of adaptive
contrast limited histogram equalization techniques
(CLAHE) are provided. Sharp field edges can be
maintained by selective enhancement within the field
boundaries. Selective enhancement is accomplished by first
detecting the field edge in a portal image and then only
processing those regions of the image that lie inside the
field edge. Noise can be reduced while maintaining the high
spatial frequency content of the image by applying a
combination of CLAHE, median filtration and edge
sharpening. A variation of the contrast limited technique
called adaptive histogram clip (AHC) can also be applied.

It is an intensity based contrast enhancement method as

Io(x,y) = f(I(x,y )) , where the original image is I (x,y),the
output image is Io (x,y) after contrast enhancement, and f is
the transformation function. The contrast stretching is a
method to make brighter portion more brighter and darker
portion darker.
There are many applications, wherein we need a flat
histogram. This cannot be achieved by histogram
stretching. so we use Histogram Equalization. Histogram
equalization maps grey levels r of an image into grey levels
s of an image in such a way that grey levels s are uniform.
This expands the range of grey levels (contrast) that are
near the histogram maxima, and compresses the range of
grey levels that are near the histogram minima.
To equalize the histogram first its PDF is calculated and
then cumulative density function (CDF) is calculated. The
CDF of the probability function is flat as shown in Fig. 2.
Hence transformed gray levels are equalized.

119

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)
AHC automatically adjusts clipping level and moderates
over enhancement of background regions of portal
images.The expression of modified gray levels for standard
CLAHE method with Uniform Distribution can be given as

Where

( )

(1)

Maximum pixel value

Minimum pixel value

a) Original Image

b) Contrast Stretching

c) HE Image

d) CLAHE Image

g is the computed pixel value

( )=CPD (Cumulative probability distribution)

For exponential distribution gray level can be

adapted as

Where

( )
[
is the clip parameter?

( )]

(2)

Fig. 3 Comparison of the Enhancement methods

CLAHE method operates on small regions in the image,

called tiles, rather than the entire image. Each tile's
contrast is enhanced, so that the histogram of the output
region approximately matches the histogram.
IV.

SEGMENTATION

Image segmentation is application dependent and success

of the segmentation is measured by the needs of the
application. Paper discusses the various histogram based
segmentation methods. Viz. Global, local and multilevel
segmentation.
A.

Fig. 4 Histogram of an Underwater Image

Global Thresholding

Global segmentation is concerned with segmenting a

whole image. Local segmentation deals with segmenting
sub-images which are small windows on a whole image.
Although a sub image is still a valid image, it is also a
fragment of a larger scene being processed in isolation. The
image consists of two global segments: a light cross on a
dark background. Each of the segments is homogeneous
and fully connected.
The global thresholding is given by
( (

))

It simply means that in global thresholding we partition

the histogram of image by using single thresholding
function T. In this case ( ) is an entire image. Because
of single threshold T it is called as global thresholding. This
kind of thresholding can be accepted to succeed only if the
histogram is well partitioned. That means there is a marked
difference in the gray level of the objects and the
background. As shown in the Fig 4.
The number of pixels available to local segmentation is
much lower than what most global segmentation algorithms
would expect. Global segmentation algorithms, in an
attempt to reduce over-segmentation, often disallow
segments containing lesser than a specified number of
pixels. A local segmentation algorithm should expect to
diagnose many small segments, including those consisting
of a single pixel. The one pixel segment may be considered
noise by global segmentation, but local segmentation must
be more lenient and allow for the fact that a lone pixel may
be part of a larger global segment.

(3)

120

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)
Global segmentation algorithms which take spatial
information into consideration usually outperform their
clustering-based counterparts. Spatial information is useful
because most segments corresponding to real world objects
consist of pixels which are spatially connected. Evaluating
the quality of segmentation algorithms is an inherently
difficult problem; this section will discuss the main
approaches to spatial segmentation and assess their
suitability to global segmentation.
B.

Local Thresholding

Global thresholding methods use the same threshold for

every pixel in an image. Difficulties arise when the
illumination of a scene varies across the image. In this case
dynamic threshold could be used based on the histogram of
an appropriately sized sub-image encompassing each pixel.
This is known as adaptive, variable or local thresholding.
Here we divide entire image into the sub images of size
of A, and select separate threshold fr the each sub image A.
[
(
) (
)
(4)
C.

Fig. 5 Multilevel thresholding [16]

V. RESULTS
In this section, some experimental results for
performance comparison of various contrast based
enhancement method are presented for the purpose of
segmentation. Source images are taken from different
underwater environment as shown in Fig. 6. Images are
firstly segmented without application of CLAHE by using
global and multilevel thresholding.

Multi Level Thresholding

Thresholding works well when a gray level histogram of

the image groups separates the pixels of the object and the
background into two dominant modes. Then a threshold T
can be easily chosen between the modes. The same
approach can be used with more than one threshold value as
shown in Fig. 5, for threshold T1 and T2, any point which
satisfies the relation
(
)
would be labeled
as an object point and all others would be labeled
background points. This is called as multilevel
thresholding.

a) See fish image

b) Fish plant image

c)

d) See object image

See plant image

Fig.6 Original test images used for the paper

121

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)

a) Original See fish image b) global thresholding

a) Original See image b) global thresholding

a) Original See Plant image b) global thresholding

c) Multilevel Segmentation with n=2 d) global threshold with CLAHE

c)Multilevel Segmentation with n=2 d) global threshold with CLAHE

c) Multilevel Segmentation with n=2 d) global threshold with CLAHE

Fig. 7 Comparison of the Segmentation method

After that CLAHE is applied on the same images with

global thresholding and segmentation is done it is clear
that the CLAHE method not only gives

Better equalization but also improves the contrast of

image. This can be clearly seen from the histograms of
the results.
S.N.
1
2
3

Images

Contrast
Histogram
Stretching
Equalization
See Fish1
32.4317
64.72
See Plant
17.295
43.462
See Fish2
16.4221
47.68
Table 1 Comparison of MSE

CLAHE
34.771
26.967
13.378

a) Original Histogram b) Histogram of Contrast stretching

SN
1
2
3
c) Histogram of Equalization d) Histogram of CLAHE
method

Images

Contrast
Histogram
Stretching
Equalization
See Fish
5.95
2.996
See Plant
6.895
2.7643
Fish plant
7.842
2.722
Table 2 Comparison of SNR

CLAHE
5.558
4.4257
9.627

VI. CONCLUSION

Fig. 8 Comparison of Histograms for See plant image

The comparison of enhancement method is presented in

the paper. It is found that CLAHE method not only
improve the contrast but also equalizes the image
histogram efficiently.

122

International Journal of Emerging Technology and Advanced Engineering

Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 1, January 2012)
It is observed that SNR is improved with CLAHE
method in comparison with HE method. The
enhancement method effectively improves the visibility
of underwater images. The comparison of segmentation
shows that with CLAHE objects are more clearly visible.

[8] Etta D, Pisano, S. Zong, R. E Jhonston Contrast limited

adaptive histogram equalization image processing to
improve the detection of simulated speculation in Dense
Monograms, Journal of Digital Imaging, vol. 11, No. 4,
pp 193-200, Nov. 1998

REFERENCES

[9]. R. Gonzalez, R. Woods, Digital Image Processing, 2nd

Edition, Prentice Hall, 2002.

[1]

A. Daskalakis, D. Cavouras, P. Bougioukos, S

Kostopoulos, and George Nikiforidis, An efficient
CLAHE based, spot adaptive, image segmentation

[10] Gonzalez, RC, Woods, RE & Eddins, SL 2004, Digital

Image processing using MATLAB, Pearson Prentice Hall.

[2] Sos S. Agaian,, , Blair Silver, and Karen A. Panetta,,

Transform
Coefficient
Histogram-Based
Image
Enhancement Algorithms Using Contrast Entropy , IEEE
Transaction on image processing, VOL. 16, NO. 3,
MARCH 2007.

[11] Kashif Iqbal, Rosalina Abdul Salam, Azam Osman and

Abdullah Zawawi Talib Underwater Image Enhancement
Using an Integrated Colour" Model ,IAENG
International Journal of Computer Science, 34:2,
IJCS_34_2_12, 2009.
[12] Shilong Wang, Yuru Xu and Yongjie Pang. A Fast
Underwater Optical Image Segmentation Algorithm
Based on a Histogram Weighted Fuzzy C-means
Improved by PSO, journal of Machine science, No.10 pp
70-75, April 2011.
[13] Schechner, Y & Karpel, N, 'Clear Underwater Vision',
Computer Vision & Pattern Recognition, vol. 1, pp. 53643, 2008
[14] Petit. F, Capelle Laize, Carre, P., Underwater image
enhancement by attenuation inversionwith quaternions,
IEEE International Conference on Acoustics, Speech and
Signal Processing , ICASSP 2009, 26.may 2009.
[15] Milton Wider, Yin M. Myint, Eko Supriyanto Comparison
Histogram Thresholding Methods for Ultrasound

[3] R. Dale-Jones and T. Tjahjadi, A study and modification

of the local histogram equalization algorithm, Pattern
Recognition, vol. 26, no. 9, pp. 13731381, 2007.
[4] Tristan J. Lambert,Digital Enhancement Techniques for
Underwater Video Image Sequences, submitted at School
of Computing In Nov.2005,
[5]. K. Zuiderveld, "Contrast Limited Adaptive Histogram
Equalization", Graphics Gems IV, pp. 474-485
[6] B. Silver, S. S. Agaian, and K. A. Panetta, Logarithmic
transform coefficient histogram matching with spatial
equalization, presented at the SPIE Defense and Security
Symp., Mar. 2005.
[7]

B. Silver, S. S. Agaian, and K. A. Panetta, Contrast

entropy based
image enhancement and logarithmic
transform coefficient histrogram shifting, presented at the
IEEE ICASSP, Mar. 2005.

Appendix Image Extraction. International journal of

computers Issue 4, Volume 5, 2011

[16] Salem Saleh Al-amri, N.V. Kalyankar and Khamitkar S.

, Image Segmentation by Using ThershodTechniques .
Journal of computing volume 2, issue 5, may 2010 issn
2151-9617

123