Poonam et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol.

5 (3) , 2014, 4301-4303

Image Enhancement with Different

M.Tech ECE1, Assistant Professor2

Abstract- With the help of Image enhancement techniques we Image Enhancement Methods
can process an image in order to make it more appropriate for Spatial domain enhancement methods:
certain applications. The motive of Image enhancement is to The Spatial domain techniques are those techniques which
explore the detail hidden in an image or to increase the operate directly on pixels. An image processing operator in
contrast in a low contrast images. The image enhancement
process try to find to increase the interpretability or
the spatial domain may be expressed as a mathematical
perception of information in the images to provide better function T[.] applied on f(x,y) to produce a new image
input for other automated image processing steps. A g(x,y)=T[f(x,y)] as follows:
benefactor of image enhancement either may be a human
observer or a computer vision program performing some kind g(x,y)=T[f(x,y)]
of higher-level image analysis, such as target detection or Frequency domain enhancement methods:
scene understanding. Image enhancement involves mainly two The new image g(x,y) is formed by the convolution of an
techniques a simple histogram-modifying point operations or image f(x,y) and a linear position invariant operator
spatial digital filtering. Some of more complex methods h(x,y),i.e.:
involve modifying the image content in another domain, such
as the coefficient domain of a linear transformation of the
image and in this paper we are going to do a modification in g(x,y)=h(x,y)*f(x,y)
hypothesis selection filter by using the concept of global mean
and global variance with localisation function. Simple intensity transformation:
(a). Image negatives:
Keywords-Image enhancement, Image Processing , Global Negative of a digital image is obtained by the
Mean & Variance. transformation function s=T(r)=L-1-r,where L
represents number of gray levels. The result of this
1. INTRODUCTION transformation is that low intensities are made high
The main objective of image enhancement is a processing and vice versa.
on an image in order to make it more appropriate for (b). Contrast stretching:
certain applications. Image enhancement mainly sharpens Low-contrast images can occur often due to poor or
image features such as boundaries, edges or contrast and non uniform lightening conditions, or due to non
reduces the ringing artefacts. The enhancement improve the linearity, or small dynamic range in the image sensor.
quality of the images so that the information contained in Contrast stretching enhance the low contrast images.
them could be extracted in a meaningful sense. The greatest The main goal of contrast stretching is to group the
difficulty in image enhancement is quantifying the criterion gray levels
for enhancement and due to this, a large number of image (c). Compression of dynamic range
enhancement techniques are empirical and require Compression expands the values of dark pixels in an
interactive procedures to obtain satisfactory results. Image image while compressing the high level values.
enhancement in digital image processing plays important When a dynamic range of a processed image cannot be
roles in many fields, such as industrial X-ray image displayed, a way to compress the dynamic range of
processing, microscopic imaging, remote sensing, medical pixel values is to perform the following intensity
image analysis, high definition television(HDTV), etc. A transformation function: s = c log(1+|r|) where c is a
benefactor of image enhancement either may be a human scaling constant, and the logarithm function performs
observer or a computer vision program performing some the desired compression.
kind of higher-level image analysis, such as scene (d) Gray-level slicing
understanding or target detection. Image enhancement The gray-level slicing highlights a specific range of
involves mainly two techniques a histogram-modifying gray levels in an image. It involves basically two
point operations or spatial digital filtering. Some of more methods: first give a high value for all the gray levels
complex methods involve modifying the image content in in the specified range and a very low value for the
another domain i.e. the coefficient domain of a linear other gray levels and second give a high value for all
transformation of the image. the gray levels in the specified range and preserve the
background and gray level tonalities in the image.

www.ijcsit.com 4301
 Poonam et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (3) , 2014, 4301-4303

Histogram processing: longer the projection path and the larger the CNR of
(a) Histogram equalization the original data, leads to the greater improvement.
The histogram represents the frequency of occurrence Confirming the theoretical analysis, the results of an
of the various gray levels in the image. Its objective is experiment utilizing practical MRA data demonstrate
to map an input image to an output image such that its the improved visual quality of small vessels.
histogram. is uniform after the mapping. 3) In 2011 - Hongchao Song, Yuanyuan Shang, Xuefeng
The main advantage of this technique is that pictures Hou, Baoyuan Han [4] discribed the typical image
with very poor dynamic range can be enhanced. The enhancement algorithms, like median filtering,
main disadvantage of this method is that visual average smoothing, homomorphic filtering and
artifacts get introduced. histogram equalization. These algorithms are also
(b) Local enhancement verified based on Matlab. In the last part of the paper,
This is often necessary to enhance details over small the pre-and post processing image using these
areas. Influence of the number of pixels in these areas algorithms are shown. This Paper's results reveal that
may be negligible on the computation of a global median filtering has a good inhibition on salt and
transformation, so the use of global histogram pepper noise than average smoothing and the
specification does not necessarily guarantee the desired homomorphic filtering has more advantages than
local enhancement. histogram equalization in dealing with the uneven
illumination image.
RELATED RESEARCH 4) In 2012 - Snehal O.Mundhada and Prof. V. K.
1) In 2000, Image Enhancement By Nonlinear Shandilya Image enhancement is the task of applying
Extrapolation in Frequency Space Hayit Greenspan, certain alterations to an input image like as to obtain a
Charles H. Anderson, and Sofia Akber [1][2] described more visually pleasing image.[6] The alteration usually
the technique for enhancing the perceptual sharpness requires interpretation and feedback from a human
of an image . This enhancement algorithm augments evaluator of the output resulting image. The Image
the frequency content of the image using shape- enhancement is to improve the image quality so that
invariant properties of edges across scale by using a the resultant image is better than the original image for
nonlinearity that generates phase- coherent higher a specific application or set of objectives. In this paper,
harmonics. This procedure utilizes the Laplacian there is a combined approach of gray level
transform and the Laplacian pyramid image transformation algorithms, i.e. logarithmic transform
representation. The results are presented depicting the and power law transform, and with alpha rooting
power-spectra augmentation and the visual algorithm for contrast enhancement. Enhancement
enhancement of several images. The simplicity of techniques like alpha rooting operate on the transform
computations and ease of implementation allow for domain where as grey level transformations operate on
real-time applications such as high-definition individual pixel.These techniques bring about tonal
television (HDTV). changes in the images and can also generate unwanted
2) In 2001, Small Vessel Enhancement in MRA Images artifacts in some cases, and as it is not possible to
Using Local Maximum Mean Processing Yi Sun and enhance all parts of the image in balanced manner.
Dennis Parker [3] proposed using local maximum mean 5) In 2012 - Jatinder kaur, Onkar Chand In this paper, a
(LMM) processing to enhance the detectability of new and efficient algorithm for reshaping of histogram
small vessels. On each voxel in the original three- that is capable in enhancing local details as well as
dimensional (3-D) data set, the LMM over the line properly preserving the image brightness is
segments in the cube centered at the voxel is taken and presented.[5] When residual bad pixels exist in the
used to form the 3-D LMM data set. The maximum image, dynamic range of the scene will be heavily
intensity projection (MIP) is then applied to the LMM suppressed when it displayed on a regular monitor.
data to produce the two-dimensional (2-D) LMM-MIP This proposed method reduced the dynamic range
image. Through LMM processing, the variance of compression (DRC) and also improve the dynamic
background tissue is minimized, thus increasing the range and contrast. Also the proposed algorithm works
detectability of small vessels. More- over, the single on zero frequency components which exist sometimes
bright voxels are suppressed and the disconnected in the original histogram, and can enhance the contrast
small vessels can be connected. However, the LMM by redistributing the original gray scales uniformly
processing widens the larger, brighter vessels. To keep onto full Gray scale range. The dynamic range of the
the advantages provided by both the LMM-MIP and image is much improved after proposed method and
MIP images , the weight functions be used to combine the details hidden in the original image are enhanced.
them. Here, The Performance of the LMM-MIP Results of simulation show the efficient performance
algorithm is analyzed and compared with the of proposed weighting method in terms of Entropy and
performance of the MIP algorithm under three EME.
measures: The vessel voxel projection probability, the 6) In 2012 - Suprijanto, Gianto, E. Juliastuti, Azhari,
vessel receiver operating characteristic (ROC) curve Lusi Epsilawati In this, described That dental
and the vessel-tissue contrast-to-noise ratio (CNR).The panoramic radiography is one of dental imaging that
Closed forms of the three measures are obtained. The used to visualize the entirety of the maxilla and

www.ijcsit.com 4302
 Poonam et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (3) , 2014, 4301-4303

mandible jaws on the one image planes.. The quality of CONCLUSION AND FUTURE WORK :
film-based image has significant limitation due to As digital Imaging improves the quality of the image with
chemical processing and image enhancement cannot be software we say this is image enhancement. This is quite
done if required. Hence, digitized film-based image to easy, for example, to make an image darker or lighter, or to
digital image was required to allow image decrease or increase contrast. The Advanced photo
enhancements in order to improve the interpretability enhancement software also supports many filters for
quality of information in the image. Digitized film- altering images in various ways. Image enhancement is to
based image is performed using a flatbed scanner on improve the image quality so that the resultant image is
transmission and reflection mode. This paper, the better than the original image for a specific application or
contrast quality of digital image that scanned using set of objectives. In the current work we have presented the
both mode is evaluated based on statistic image image enhancement techniques. We have seen that every
characteristic. and the results showed that the quality technique has its own pros and cons.In future we will
of digitized image using transmission mode is better present a new concept for enhancing using the hybrid of
than using reflection mode. However, if direct digital HSF selection filter with global mean and variance for local
imaging is used as a standard, image enhancement on functions
digitized image is still required. All Four methods, i.e.
contrast stretching, HE, AHE, and CLAHE are used to REFERENCES
improve the quality digitized image. The evaluation of 1. Hayit Greenspan, Charles H. Anderson, and Sofia Akber, Digital
the preference image quality is performed based on Image Processing, Reading, MA: Addison-Wesley, 2000.
objective criterion. The preferable image quality for 2. Hayit Greenspan, Charles H. Anderson, and Sofia Akber, Image
Enhancement By Nonlinear Extrapolation in Frequency Space, IEEE
digitized panoramic image can be obtained by using TRANSACTIONS ON IMAGE PROCESSING, VOL. 9, NO. 6,
image enhancement based on CLAHE-Rayleigh JUNE 2000
method, which is indicated by the lowest value of 3. Yi Sun and Dennis Parker, Small Vessel Enhancement in MRA
mean, RMSE, standard deviation, and average Images Using Local Maximum Mean Processin, IEEE
difference and the higher value of NAE and SAE. TRANSACTIONS ON IMAGE PROCESSING, VOL. 10, NO. 11,
NOVEMBER 2001.
7) In 2013 – Vijay A. Kotkar, Sanjay S. Gharde Image
4. Hongchao Song, Yuanyuan Shang, Xuefeng Hou, Baoyuan Han,
enhancement is a processing on an image in order to image enhancement algorithms such as median filtering, , FL: CRC
make it more appropriate for certain applications.[7] It Press, 2011.
is used to improve the visual effects and the clarity of 5. Jatinder kaur, Onkar Chand, Contrast Enhancement with Reshaping
image or to make the original image more conducive Local Histogram using Weighting Method, ISSN: 2250-3021
for computer to process. The Contrast enhancement Volume 2, Issue 6 (June 2012).
changing the pixels intensity of the input image to 6. Snehal O.Mundhada and Prof. V. K. Shandilya, Image Enhancement
using a Combined Approach of Spatial and Transformation Domain
utilize maximum possible bins. So, We need to study Technique, International Journal of Emerging Research in
and review the different image contrast enhancement Management &Technology ISSN: 2278-9359, 2012.
techniques because contrast losses the brightness in 7. Vijay A. Kotkar, Sanjay S. Gharde, REVIEW OF VARIOUS
enhancement of image. By considering this fact, the IMAGE CONTRAST ENHANCEMENT TECHNIQUE, Vol. 2,
mixture of global and local contrast enhancement Issue 7, July 2013.
techniques may enhance the contrast of image with 8. C. Vaishnavi and Dr. P. Eswaran, Improved Colour Image
Enhancement Scheme using Mathematical Morphology, Volume 3,
preserving its brightness. We have many image Issue 4, April 2013
contrast enhancement techniques such as HE, BBHE, 9. O. Yli-Harja, J. Astola, Y. Neuvo, Analysis of the properties of
DSIHE,MHE. BPDHE, RSWHE, MMBEBHE, median and weighted median filters using threshold logic and stack
RMSHE, GHE, LHE and LGCS. In This paper author filter representation. IEEE Trans. Acoust. Speech Signal Process.,
focuses on the comparative study of contrast ASSP-39: 395-410, 1991.
enhancement techniques with special reference to local 10. P.-T. Yu, W.-H. Liao, Weighted order statistic filters - their
classification, some properties and conversion algorithm. IEEE
and global enhancement techniques. Also proposed Trans. Signal Process., 42: 2678-2691, 1994.
solution is identified to apply to this enhancement 11. Aditi Majumder, Sandy Irani, Contrast Enhancement of Images using
technique. Method will use in many fields, such as Human Contrast Sensitivity, Computer Science Department,
microscopic imaging, medical image analysis, hyper University of California, Irvine
spectral image processing, remote sensing, HDTV, 12. J. Serra, Image Analysis and Morphology: Vol. 2: The Theoretical
industrial X-ray image processing etc. Advances. London: Academic Press, 1988.

www.ijcsit.com 4303