The paper purposes the removal of noise in digital gray scale images that often observed in scanned documents. Generally, Data i.e. picture, text can be contaminated by an additive noise during the process of scanning. This methodology prevents this type of noise known as Salt and Pepper noise (SP Noise) which causes white and black spots on the original image. We are designing a new algorithm for removal of these white and black spots after the knowledge of Median Filter, Adaptive Filter and the new proposed algorithm will definitely protect the image from noise and distortion. Firstly, Adaptive Histogram Equalization is done on the original image. Secondly apply Adaptive contrast Enhancement Technique on the resultant image. After Contrast Enhancement we apply filters Such as Homomorphic filtering. These filters are applied sequentially on distorted images for removing the image.

1. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
DOI:10.5121/ijfcst.2015.5107 67
A NEW METHODOLOGY FOR SP NOISE REMOVAL
IN DIGITAL IMAGE PROCESSING
Upma Bansal1
1
Assistant Professor, Dept. of CSE, Chandigarh University, Punjab, India
Rekha Saini2
2
Assistant Professor, Dept. of CSE, Chandigarh University, Punjab, India
Ashish Verma3
3
Assistant Professor, Dept. of CSE/IT, SSIET, PTU, India
ABSTRACT
The paper purposes the removal of noise in digital gray scale images that often observed in scanned
documents. Generally, Data i.e. picture, text can be contaminated by an additive noise during the process
of scanning. This methodology prevents this type of noise known as Salt and Pepper noise (SP Noise) which
causes white and black spots on the original image. We are designing a new algorithm for removal of these
white and black spots after the knowledge of Median Filter, Adaptive Filter and the new proposed
algorithm will definitely protect the image from noise and distortion. Firstly, Adaptive Histogram
Equalization is done on the original image. Secondly apply Adaptive contrast Enhancement Technique on
the resultant image. After Contrast Enhancement we apply filters Such as Homomorphic filtering. These
filters are applied sequentially on distorted images for removing the image.
KEYWORDS
Salt and Pepper, filter, adaptive, image, noise.
1. INTRODUCTION
An image is an artifact that represents visual observation. It may be captured by optical means
like cameras, lenses, telescopes, microscopes etc. Often the raw image is not openly suitable for
this purpose, it always needs some processing. Such a processing on image is called image
enhancement; processing by an observer to extort information is called image analysis.
Enhancement and analysis are well-known by their output and by the challenges faced and
methods exercised. Image enhancement has been done by various methods i.e. chemical, optical
and electronic means [12].
A digital image is a mathematical representation of a two-dimensional image. Digital images are
composed of pixels i.e. picture element. Each pixel represents the gray level for black and white
photos at a single point in the image, so a pixel can be represented by a tiny dot of a fussy color.
By calculating the color of an image at a large number of points, we can generate a digital
approximation of the image from which a copy of the original can be recreated. Pixels are a slight
like grain particles in a conventional photographic image, which can be arranged in a regular
pattern of rows and columns and store information differently to some extent. A digital image is a
rectangular arrangement of pixels sometimes called a bitmap.
Priyanka Kamboj and Versha Rani [1] have studied various noise model and filtering techniques.
In image processing, noise reduction and image restoration is expected to improve the qualitative
inspection of an image and the performance criteria of quantitative image analysis techniques.

2. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
68
Digital image is inclined to a variety of noise which affects the eminence of image. The purpose
of image denoising is to restore the detail of original image to the great extent. The criterion of
the noise removal problem depends on the type of noise by which is corrupting the image.
Different methods for reduction of noise and image enhancement have been considered.
Raymond H. Chan, Chung-Wa Ho, and Mila [2] put forward a two-phase scheme for removing
salt and pepper noise. In the foremost phase, an adaptive median filter is used which identify
pixels that are likely to be infected by noise. In the successive phase, the image is restored using a
specialized regularization method that applies only to those selected noise candidates. In terms of
edge perpetuation and noise suppression, their restored images represent a significant
enhancement compared to those restored by using just nonlinear filters or regularization methods
only. Strategy can remove salt-and-pepper-noise with a noise level as high as 90%.
M. S. Nair, K. Revathy, and Rao Tatavarti [3] evinced an improved decision-based algorithm for
the restoration of gray-scale and color images that are highly corrupted by Salt and Pepper noise
which efficiently removes the salt and pepper noise by preserving all details. The algorithm
utilizes formerly processed neighboring pixel values to get better image quality than the one
utilizing only the just previously applied pixel value. The projected algorithm is faster and also
produces better result than a Standard Median Filter (SMF). The advantage of the proposed
algorithm lies in removing only the noisy pixel either by the median value or by the mean of the
previously processed neighboring pixel values.
B. Singh, Ravinder Singh, Harmandeep Singh [4] proposed that removal of high density salt and
pepper noise in noisy color images using projected median filter. The presentation of improved
median filter is good at lower noise density level. The mean filter prevents little noise and gets the
worst results. The enhanced median filter is good at lower noise density levels. It suppresses most
of the noises effectively while preserving colored image details. The performance of the
algorithm is analyzed in terms of Peak signal to noise ratio (PSNR), Mean square error
(MSE),Image Enhancement Factor (IEF).
W. Luo [5] suggested that images are often corrupted by noise known as salt and pepper noise.
This noise can corrupt the images and the corrupted pixel takes either maximum or minimum
gray level. Along with these standard median filters has been established as reliable method to
remove the salt and pepper noise without harming the edge features. Though, the major problem
of standard Median Filter (MF) is that the filter is effective only at low noise densities.
2. PROPOSED ALGORITHM
The Sequence of algorithms used as follows:-
1. Firstly Adaptive Histogram Equalization Technique is performed on the input image to
improve contrast in image.
2. After this, Adaptive Contrast Enhancement is applied to improve the quality of the image.
3. Contrast Stretching is done to improve the contrast in an image by stretching the range of
intensity values it contains to span a desired range of values.
4. After Contrast Stretching, Homomorphic filtering is used for image enhancement.
Usually it normalizes the brightness across an image and improve contrast. So, homomorphic
filtering is used to remove noise.
5. Finally, we get the output noise free image.

3. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
69
The following strategy is used to get the desired results.
Fig. 1 Methodology for salt & pepper noise removal
3. PARAMETERS USED
The different parameters which are used to check the performance are as follows:
1. Mean Square Error (MSE) is the ratio of the square of difference between the input and
output image to the size of the image.
    
2
1 1
21 ,,
1
 

M
i
N
j
jifjif
MN
MSE (1)
Here f1, f2 are the input and output images respectively. M and N are the sizes of the images.
2. Peak Signal to Noise ratio (PSNR) is the logarithmic value of the ratio of size of the image
and the mean square error of the image.







MSE
LOGPSNR
2
255
10 (2)
Peak Signal to Noise Ratio should be as large as possible which means that the content of signal
in the output is large and the noise is less.
3. Signal to noise ratio (SNR) should be as large as possible which means that the content of
signal in the output is large and the noise is less.
0
BA SS
SNR

 (3)
SA is the mean intensity value of the image and the background respectively.
0 is standard deviation of the image.
Adaptive
Histogram
Equalization
Adaptive
Contrast
Enhancement
Contrast
Stretching
Homomorphic
Filtering
Output
Image
Input
Image

4. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
70
Fig. 2 Image 1 after proposed Technique Fig. 3 Image 2 after proposed technique
4. COMPARATIVE ANALYSIS BASED ON OBJECTIVE PARAMETERS
The Following table shows the comparative analysis of proposed techniques with previous
techniques of salt and pepper noise removal using PSNR, MSE and SNR.
TABLE 1: Comparison Based on Mean Square Error Parameter for Figures.
TABLE 2: Comparison Based on Peak Signal to Noise Ratio Parameter for Figures.
Technique Proposed Technique1
Fig. 3 18.433 14.8526
Fig. 3 22.713 14.322
Fig. 3 20.344 15.855
Technique Proposed Technique1
Fig. 2 0.0021 603
Fig. 2 0.0024 608
Fig. 2 0.0017 605

5. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
71
Fig. 4
TABLE 3: Comparison Based on Signal to Noise Ratio Parameter for Figures
Technique Proposed Technique1
Fig. 4 31.5275 17.4797
Fig. 4 32.6164 9.4315
Fig. 4 16.8741 10.4024
5. COMPARATIVE ANALYSIS BASED ON SUBJECTIVE PARAMETERS
The Following Table shows Comparisons Based on Subjective Parameters:-
Technique-1 is SP Noise Removal using Homomorphic Filtering.
Proposed Technique is the combination of Median Filtering and Homomorphic Filtering using
Adaptive Histogram Equalizations.
originalimage adaptivehistogrameqalization
homomorphic image
adaptivecontrastenhancement
finalimage

6. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
72
Fig. 5.1(a). Original
Image
Fig. 5.1(b). Image
after Homomorphic
Filtering
Fig. 5.1(c). Image after
Adaptive Median
Filtering
Fig. 5.1(d). Image
after Proposed
Approach
Fig. 6.1(a). Original
Image
Fig. 6.1(b). Image
after Homomorphic
Filtering
Fig. 6.1(c). Image after
Adaptive Median
Filtering
Fig. 6.1(d). Image
after Proposed
Approach

7. International Journal in Foundations of Computer Science & Technology (IJFCST), Vol.5, No.1, January 2015
73
6. CONCLUSIONS
In this paper, a recent adaptive noise reduction scheme for removing salt and pepper noise is
proposed which categorizes impulse noise and will remove the noise from the corrupted image
that is followed by image enhancement scheme to retain the details and image quality. As per the
new results, the future algorithm yields good filtering result. This is recorded by numerical
measurements like PSNR and visual observations through the experiments performed. So, we
recommend an algorithm for eliminate salt and pepper noise removal. The suggested method is
actually an adaptive median filter. The benefits of this proposed method are the fuzzy
initialization of filtering window size and the precision of median value. Comprehensive method
results expose that our filter consistently outperforms the existing filters by attaining much higher
PSNR across a wide range of noise densities, from 10% to 90%.
The purpose of such method is mainly due to highly accurate noise detection experienced by the
noise detection algorithm having high noise detection ratio and our method performs more
desirable than the median filter and other conventional edge preserving method. The PSNR, SNR
is high; MSE is low. This advised method is a fast method for removing salt and pepper noise.
REFERENCES
[1] P.Kamboj, V.Rani, Image Enhancement Using Hybrid Filtering Technique, IJSR, vol. 2(6), 2013,
214-220.
[2] R.H.Chan, C.W. Ho, M.Nikolova, Salt-and-Pepper Noise Removal by Median-Type Noise Detectors
and Detail- Preserving Regularization, IEEE Transactions on Image Processing, vol. 14, No. 10,
1479-1485, 2005.
[3] Nair, M.S., Revathy, K.; Tatavarti, Removal of Salt and Pepper Noise in Images: A new Decision
based Algorithm, IMECS’08, vol. 1, 2008.
[4] B.Singh, R.Singh, H.Singh, Removal of High Density Salt & Pepper Noise in Noisy color Images
using Proposed Median Filter, vol 2, No 2, 2013.
[5] W.Luo, Efficient Removal of Impulse Noise from Digital Images”, IEEE Transactions, 2006, 523-
527.
[6] K.S.Srinivasan, D.Ebenezer, A New Fast and Efficient Decision Based Algorithm for Removal of
High-Density Impulse Noises, IEEE signals processing letter, vol. 14, no. 3, 2007, 189-192.
[7] A.Verma, B.Sharma, Comparative Analysis in Medical Imaging, IJCA, vol. 1- No. 13, 2010.
[8] S.Pawar, P.S.Halgaonkar, J.W.Bakal, V.M.Wadhai, Implementation of PPM Image Processing and
Median Filtering, IJCA, vol. 14– no.5, January 2011.
[9] S.A.M.Saleh, H.Ibrahim, Mathematical Equations for Homomorphic Filtering in Frequency Domain,
IPCSIT, vol. 45, 2012.
[10] S.Deivalakshmi, S.Sarath, P.Palanisamy, Detection and Removal of Salt and Pepper noise in images
by Improved Median Filter, IEEE, 2011.
[11] Swapna M.Patil, R.R.Karhe, C.S.Patil, Kirti Mahajan, International Journal of Engineering Research
& Technology (IJERT) Vol. 2 Issue 1, January- 2013.
[12] B.Ho.Knag, A Review on Image and Video Processing, IJMUE, 2007, 49-63.
[13] Rekha Rani, Ashish Verma, Shiv Kumar Verma, Upma Bansal, International Journal in Foundations
of Computer Science and Technology (IJFCST) Vol. 4 Issue 5, September- 2014.