IJCSMS International Journal of Computer Science and Management Studies, Vol.

11, Issue 02, Aug 2011 114

ISSN (Online): 2231-5268
www.ijcsms.com

An Enhanced Approach to improve the contrast of

Images having bad light by Detecting and
Extracting their Background
Gurpreet Bhatia1, Deepak Goel2
1
Scholar, M.tech,CSE Department, VCE, Rohtak
[email protected]
2
Assistant Professor, CSE Department, VCE, Rohtak
[email protected]

ABSTRACT the difference between the highest and lowest intensity

values in an image gives a measure of its contrast. The
The application of mathematical morphology to image
processing and analysis has initiated a new approach for first work dealing with contrast theory was carried out by
solving a number of problems in the image enhancement field. Meyer and Serra.
So, we have used morphological operators to detect and extract
the background of images. Various operators used are erosion, II. BASIC CONCEPTS
dilation, opening and closing. After the background is
extracted, contrast of bad lightning image is improved by the
application of operators based on Weber’s law notion. The Morphological operators
operators used are block analysis and opening by
reconstruction. In this paper, we will present the result of The identification of objects within an image can be a
images on applying these operators. very difficult task. One way to simplify the problem is to
Keywords: Morphological operators, block analysis, opening change the grayscale image into a binary image, in which
by reconstruction, contrast. each pixel is restricted to a value of either 0 or 1. The
techniques used on these binary images go by such
names as: blob analysis, connectivity analysis,
I. INTRODUCTION and morphological image processing (from the Greek
word morph, meaning shape or form). The foundation of
In this paper, firstly we give introduction about various morphological processing is in the mathematically
morphological operators and then we apply them on a rigorous field of set theory; however, this level of
bad light image and extract the background of that image sophistication is seldom needed. Morphological
and then improve contrast of that image. Image operators often take a binary image and a structuring
enhancement is a useful technique in image processing element as input and combine them using a set operator
that permits the improvement of the visual appearance of (intersection, union, inclusion, complement). They
the image or provides a transformed image that enables process objects in the input image based on
other image processing tasks (image segmentation, for characteristics of its shape, which are encoded in the
example).Methods in image enhancement are generally structuring element. Usually, the structuring element is
classified into spatial methods and frequency domain sized 3×3 and has its origin at the center pixel. It is
ones. The present work is focused on the spatial shifted over the image and at each pixel of the image its
methods, and in particular, to the use of morphological elements are compared with the set of the underlying
image transformations. Mathematical morphology pixels. If the two sets of elements match the condition
approach is based on set theoretic concepts of shape. In defined by the set operator (e.g. if the set of pixels in the
morphology objects present in an image are treated as structuring element is a subset of the underlying image
sets. pixels), the pixel underneath the origin of the structuring
Quite often a recorded image suffers from a common element is set to a pre-defined value (0 or 1 for binary
degradation like poor contrast. The range of intensity i.e. images).

IJCSMS
www.ijcsms.com
IJCSMS International Journal of Computer Science and Management Studies, Vol. 11, Issue 02, Aug 2011 115
ISSN (Online): 2231-5268
www.ijcsms.com

A morphological operator is therefore defined by its The two methodologies we used to enhance the contrast
structuring element and the applied set operator. For the of images is based on Weber’s law notion and are:-
basic morphological operators the structuring Block analysis method and opening by reconstruction
element contains only foreground pixels (i.e. ones) and method.
`don't care's'. These operators, which are all a
combination of erosion and dilation, are often used to In block analysis method, each block is the sub image of
select or suppress features of a certain the original image. The maximum and minimum
shape, e.g. removing noise from images or selecting intensity values are denoted as Mi and mi. For each
objects with a particular direction. analyzed block, maximum (Mi) and minimum (mi)
values are used to determine the background measures.
In erosion, every object pixel that is touching a Ti is used to select the background parameters.
background pixel is changed into a background pixel. In Background parameters line between clear (f >Ti) and
dilation, every background pixel that is touching an dark (f <=Ti) intensity levels. and dark intensity levels.
object pixel is changed into an object pixel. Erosion Once Ti is calculated, this value is used to select the
makes the objects smaller, and can break a single object background parameter associated with the analyzed
into multiple objects. block.

In MM, there is other class of transformations that allows

Dilation makes the objects larger, and can merge
multiple objects into one. As shown in (d), opening is the filtering of the image without generating new
defined as an erosion followed by a dilation. Figure (e) components; these transformations are called
transformations by reconstruction. The normal
shows the opposite operation of closing, defined as a
morphological opening is an erosion followed by a
dilation followed by an erosion.
dilation. The erosion "shrinks" an image according to the
shape of the structuring element, removing objects that
As illustrated by these examples, opening removes small are smaller than the shape. Then the dilation step
islands and thin filaments of object "regrows" the remaining objects by the same shape. The
pixels. Likewise, closing removes islands and thin dilation step in the opening operation restored the
filaments of background pixels. vertical strokes, but the other strokes of the characters
are missing. How can we get the entire characters
containing vertical strokes. The answer is to use
morphological reconstruction. For binary images,
reconstruction starts from a set of starting pixels (or
"seed" pixels) and then grows in flood-fill fashion to
include complete connected components. To get ready to
use reconstruction, first define a "marker" image. This is
the image containing the starting or seed locations.

III. CONTRAST ENHANCEMENT

PROCESS
The contrast enhancement process consists of different
steps:-

• Image Acquisition
Fig 1 Morphological operations four basic morphological operations • Applying Morphological operators
are used in the processing of binary image: erosion, dilation, opening • Detecting and extracting the background
and closing.Fig (a) shows an example binary image. Fig (b) to (e) • Applying contrast enhancement operators:-
shows the result of applying these operations to the image in (a).
block analysis and opening by reconstruction
• Applying image enhancement techniques like
image sharpening etc.
Contrast enhancement operators
• Final image

IJCSMS
www.ijcsms.com
IJCSMS International Journal of Computer Science and Management Studies, Vol. 11, Issue 02, Aug 2011 116
ISSN (Online): 2231-5268
www.ijcsms.com

Fig 2 Proposed methodology to compute the problem

Result of Step 3

Then in this step we separate the background from the

IV. IMPLEMENTATION RESULTS OF image.
VARIOUS STEPS

Result of Step1

In this step, we acquire image which has poor lightning

or dull, from a specified place.

Result of Step 4

Then in the next step, we segment the image into sub-

images. As the source image is difficult to deal with in a
general view. Thus we decompose it into simpler ways in
this processing stage. The decomposed sub-images are
Result of Step 2 processed by a morphological filter to emphasize the
character region and suppress the small islands of noises.
In this step, firstly we convert image into grayscale
image and then apply various morphological operations
on the image such as erosion, dilation, opening and
closing to see the exact location of foreground image.

IJCSMS
www.ijcsms.com
IJCSMS International Journal of Computer Science and Management Studies, Vol. 11, Issue 02, Aug 2011 117
ISSN (Online): 2231-5268
www.ijcsms.com

Result of Step 5

Finally the sub-images are united to obtain the resulting

image. Then, we enhance the image by various functions
like image sharpening etc. By doing this, finally our
image will be of good contrast and free from bad
lightning.

Fig 4: Histogram of improved image

VI. CONCLUSION
The proposed methodology in this paper is used to detect
the image background and to enhance the contrast in
grey level images with bad lighting. In this paper, firstly
V. COMPARISON OF IMAGES we give introduction about basic concepts used in this
methodology like Morphological operators, contrast
Then, we will compare the contrast of both images by enhancement operators. Then we provide various steps to
Histogram Equalization process. During the histogram perform the methodology. Then, we provide
equalization process, grey level intensities are reordered implementation results of steps involved in proposed
within the image to obtain an uniform distributed methodology. And finally, we provide the comparison
histogram. between two images i.e. original image and contrast
improved image by the help of Histogram equalization
process.

There is a disadvantage of contrast enhancement

transformations i.e. they can only be used satisfactorily
in images with poor lighting; in a future work this
problem will be considered.

VII. REFERENCES
[1] I. R. Terol-Villalobos, “Morphological image
enhancement and segmentation,” in Advances in Imaging
and Electron Physics,P.W.Hawkes, Ed. New York:
Academic, 2001, pp. 207–273.
Fig 3 : Histogram of original image [2] Angélica R. Jiménez-Sánchez, Jorge D. Mendiola-
Santibañez, “Morphological Background Detection and
Enhancement of Images With Poor Lighting”, IEEE
transactions on image processing, Vol. 18, No. 3, March
2009.
[3] I. R. Terol-Villalobos, “Morphological connected
contrast mappings based on top-hat criteria: A multiscale
contrast approach,” Opt. Eng., vol. 43, no. 7, pp. 1577–
1595, 2004.
[4] A. Toet, “Multiscale contrast enhancement with
applications to image fusion,” Opt. Eng., vol. 31, no. 5,
1992.

IJCSMS
www.ijcsms.com
IJCSMS International Journal of Computer Science and Management Studies, Vol. 11, Issue 02, Aug 2011 118
ISSN (Online): 2231-5268
www.ijcsms.com

[5] S. Mukhopadhyay and B. Chanda, “A multiscale variable pose and illumination,” in Proc. IEEE Int. Conf.
morphological approach to local contrast enhancement,” Automatic Face and Gesture Recognition, 2000, pp. 277–
Signal Process., vol. 80, no. 4, pp. 685–696, 2000. 284.
[6] J. Kasperek, “Real time morphological image contrast [25] Z. Liu, C. Zhang, and Z. Zhang, “Learning-based
enhancement in virtex FPGA,” in Lecture Notes in perceptual image quality improvement for video
Computer Science. New York: Springer, 2004. conferencing,” presented at the IEEE Int. Conf.
[7] P. Salembier and J. Serra, “Flat zones filtering, connected Multimedia and Expo (ICME), Beijing, China, Jul. 2007.
operators and filters by reconstruction,” IEEE Trans.
Image Process., vol. 3, no.8, pp. 1153–1160, Aug. 1995.
[8] E. Peli, “Contrast in complex images,” J. Opt. Soc.
Amer., vol. 7, no.10, pp. 2032–2040, 1990.
[9] J. Short, J. Kittler, and K. Messer, “A comparison of
photometric normalization algorithms for face
verification,” presented at the IEEE Int.Conf. Automatic
Face and Gesture Recognition, 2004.
[10] Lixu Gu, Toyohisa Kaneko, “Morphological
segmentation applied to character extraction from color
cover images”
[11] G. M. Matheron. Random sets and integral in Geometry.
Wiley,NewYork, 1975.
[12] J. Serra. Image analysis using mathematical morphology.
Academic Press, London, 1982, Conf. Automatic Face
and Gesture Recognition, 2004.
[13] C. R. González and E.Woods, Digital Image Processing.
Englewood Cliffs, NJ: Prentice Hall, 1992.
[14] R. H. Sherrier and G. A. Johnson, “Regionally adaptive
histogram equalization of the chest,” IEEE Trans. Med.
Imag., vol. MI-6, pp.1–7, 1987.
[15] L. Vincent, Gray scale area openings and closings, their
efficient implementation and applications, in Proc. of the
workshop Mathematical Morphology and its Applications
to Signal Processing, Barcelona, Spain, pp. 22-27, May
1993.
[16] J. Serra, Mathematical Morphology Vol. I. London, U.K.:
Academic, 1982.
[17] P. Soille, Morphological Image Analysis: Principle and
Applications. New York: Springer-Verlag, 2003.
[18] H. Heijmans, Morphological Image Operators. New
York: Academic, 1994
[19] L. Vincent and E. R. Dougherty, “Morphological
segmentation for textures and particles,” in Digital Image
Processing Methods, E. R.Dougherty, Ed. New York:
Marcel Dekker, 1994, pp. 43–102.
[20] L.Vincent, “Morphological grayscale reconstruction in
image analysis: Applications and efficient algorithms,”
IEEE Trans. Image Process., vol. 2, no. 2, pp. 176–201,
Feb. 1993.
[21] E. Peli, “Contrast in complex images,” J. Opt. Soc.
Amer., vol. 7, no. 10, pp. 2032–2040, 1990.
[22] A. Edgar, R. Araiza, J. D. Mendiola Santibañez, G.
Herrera Ruiz , C. A. G. Gutiérrez, M. T. Perea, and G. J.
R. Moreno, “Contrast Enhancement and Illumination
Changes Compensation,” Computación y Sistemas, vol.
10, no. 4, pp. 357–371, 2007.
[23] P. Maragos and R. Schafer, “Morphological filters—Part
I: Their settheoretical analysis and relations to linear shift
invariant filters,” IEEE Trans. Acoust. Speech Signal
Process., vol. 35, pp. 1153–1169, 1987.
[24] A. S. Georghiades, P. N. Belhumeur, and D. J.
Kriegman, “Generative models for recognition under
IJCSMS
www.ijcsms.com