Design of Novel Filter for the Removal of Gaussian Noise in Plasma Images L. LAKSHMI PRIYA PG Scholar, Department of ETCE, Sathyabama University, Chennai llakshmipriyabe@gmail.com Dr.M.S.GODWIN PREMI Professor, Dept. of ETCE Sathyabama University, Chennai msgodwinpremi@gmail.com T. BOOBALAN ME Assistant Professor Department of ECE ACET, Pollachi Tamilnadu mails2boobalan@gmail.com L. MALATHI ME Assistant Professor Department of ECE SRIT, Coimbatore Tamilnadu lmalathigraj@gmail.com ABSTRACT Noise can degrade the image at the time of capturing or transmission of the image. The noise removal algorithms are depends on the type of noise present in the image. Different noise models including additive and multiplicative types are used. They include Gaussian noise, salt and pepper noise and speckle noise. To ensure the quality of image in image processing, noise estimation and removal are very important step before analysis or using image. So the original image will get distorted during the transmission. Noisy image consist of unwanted data which may reduce the shape and size of objects in the image and blurring of edges or dilution of fine details in the image. So eliminating such noise is an important pre-processing task. This process of removing noise is called De-noising. The important property of a good image denoising model is that should completely remove noise as far as possible also the PSNR value should be high. The main focus of this project is To estimate the Gaussian Noise and designing of a Novel Filter to remove the Gaussian noise in plasma images. Keywords: Gaussian Noise, Plasma Images, PSNR. I. INTRODUCTION Image de-noising is a vital image processing task i.e. as a process itself as well as a component in other processes. There are many ways to de-noise an image or a set of data and methods exists. The important property of a good image denoising model is that it should completely remove noise as far as possible as well as preserve edges. Traditionally, there are two types of models i.e. linear model and non-liner model. Generally, linear models are used. The benefits of linear noise removing models is the speed and the limitations of the linear models is, the models are not able to preserve edges of the Page 64
images in a efficient manner i.e the edges, which are recognized as discontinuities in the image, are smeared out. Additive noise is the signal that gets added to the original image to generate the resultant noisy image. The image s(x,y) is blurred by a linear operation and noise n(x,y) is added to form the degraded image w(x,y). This is convolved with the restoration procedure g(x,y) to produce the restored image z(x,y). In Multiplicative model the noisy image is generated by multiplication of the original image and the noise signal [6]. Image noise can be classified as Impulse noise (Salt-and-pepper noise), Amplifier noise (Gaussian noise), Shot noise, Quantization noise (uniform noise), Film grain, on-isotropic noise, Multiplicative noise (Speckle noise) and Periodic noise. In this paper the experiment results shows the PSNR values for the novel filter to remove the Gaussian Noise. Figure 1: Denoising concept The Linear operation shown in Figure 1.1 is the addition or multiplication of the noise n(x,y) to the signal s(x,y). Once the corrupted image w(x,y) is obtained, it is subjected to the denoising technique to get the denoised image z(x,y). Different algorithms are used depending on the noise model. Most of the natural images are assumed to have additive random noise [1] is observed in ultrasound images where Rician noise [2] affects MRI images. The quality of image can be measured by the peak signal-to-noise ratio (PSNR). The scope of this paper is to design a novel filter to remove the Gaussian noise in Plasma images. II. TYPES OF NOISE All types of noises categorized into two models: can be III. GAUSSIAN NOISE Gaussian noise is statistical noise having a probability density function (PDF) equal to that of the normal distribution, which is also known as the Gaussian distribution Generally Gaussian noise mathematical model represents the correct approximation of real world scenarios. In this noise model, the mean value is zero, variance is 0.1 and 256 gray levels in terms of its PDF [2], which is shown in Fig. 2. 1. Additive Noise Model 2. Multiplicative Noise Model Page 65
FILTER 1 Noise Estimation Figure 3: Adding white Gaussian noise Figure 2: Denoising concept Due to this equal randomness the normalized Gaussian noise curve look like in bell shaped. The PDF of this noise model shows that 70% to 90% noisy pixel values of degraded image in between μ s and μ + s. The shape of normalized histogram is almost same in spectral domain. IV. EXPERIMENT RESULT For the testing purpose we have added white Gaussian noise to the girl.tif image and the noise is removed using two filters Adaptive Normalised Weightage Filter and Bi-square Adaptive Normalised Weightage Filter. Finally the PSNR values were compared between two filters. Figure 4: Denoised image FILTER 2 Noise Estimation Figure 5: Adding white Gaussian noise Page 66
Figure 9: Denoised image Figure 6: Denoised image FILTER 2 V. PROPOSED APPROACH In proposed system the noise have been added in the original Plasma image and the noise estimation & de-noising were done. Noise Estimation and removal of Gaussian Noise in Plasma Image Original Plasma Image Figure 10: Adding white Gaussian noise Figure 7: Plasma Image (original Image) FILTER 1 Removal of Gaussian Noise in Plasma Image Figure 11: Adding white Gaussian noise Figure 8: Adding white Gaussian noise Page 67
3. UVLSI design and no of flip flops per sq meter is minimal. Filter Selection: Median filter Remove the outlier without reducing the sharpness of the image. So performs better result in Plasma image for noise removal [5]. Adaptive filter Requires less computation time. Currently we are woking to derive the function for noise estimation using curve fitting toolbox in Matlab R2010a and Novel filter design to remove Gaussian Noise using curve fitting toolbox in Plasma Image. VII. CONCLUSION Removing Noise by Median Filter: In signal processing its desirable to perform noise reduction on an image. Median filter is a non-linear filtering technique used to remove noise. After denoising the edges has to be preserved of the digital image. Median filter remove the noise from the image as well as preserve the edges of image, so median filter is widely used in image processing. The main idea of the median filter is to run through the signal entry by entry, replacing each entry with the median of neighboring entries. The pattern of neighbors is called the "window", which slides, entry by entry, over the entire signal. The median is much less sensitive than the mean to external values. Therefore it is better to remove the external values without reducing the image sharpness. It is one kind of smoothing technique. Its performance is not much better than Gaussian blur for high levels of noise, whereas, for speckle noise and impulsive noise it is effective. Because of this, median filtering is very widely used in digital image processing [3]. VI. FUTURE WORK Advantages of Filter 1 and Filter 2 are: 1. Hardware design implementation is easy. 2. Power dissipation is minimal. We used the girl.jpg image and added the white Gaussian noise and compared the PSNR value. The PSNR value is increased in filter 2 than filter 1. Median filter is better able to remove noise without reducing the sharpness of the image and the performance of the Median filter after denoising for Gaussian noise is better than other filter (Mean and Wiener filter). VIII. REFERENCES [1] H. Guo, J. E. Odegard, M. Lang, R. A. Gopinath, I. W. Selesnick, and C. S. Burrus, "Wavelet based speckle reduction with application to SAR based ATD/R," First Int'l Conf. on Image Processing, vol. 1, pp. 75-79. [2] Signal & Image Processing : An International Journal (SIPIJ) Vol.6, No.2, April 2015 DOI : 10.5121/sipij.2015.6206 63A REVIEW PAPER: NOISE MODELS IN DIGITAL IMAGE PROCESSING Ajay Kumar Boyat1 and Brijendra Kumar Joshi2 1Research Scholar, Department of Electronics Telecomm and Computer Engineering, Military College of Tele Communication Engineering, Military Head Quartar of War (MHOW), Ministry of Defence, Govt. of India, India 2Professor, Department of Electronics Telecomm and Computer Engineering, Military College of Tele Communication Engineering, Military Page 68
Head Quartar of War (MHOW), Ministry of Defence, Govt. of India, India. [3] International Research Journal of Engineering and Technology (IRJET) eissn: 2395-0056 Volume: 02 Issue: 08 Nov-2015 www.irjet.net p-issn: 2395-0072 2015, IRJET ISO 9001:2008 Certified Journal Page 127 REVIEW PAPER: TO STUDY THE IMAGE DENOISING TECHNIQUES Kalpana1, Harjinder Singh2 1 Student of M.tech,Electronics and Communication,Punjabi University Patiala,Punjab,India 2 Assistant Professor, Electronics and Communication,Punjabi University Patiala,Punjab,India. [4] IEEE TRANSACTIONS ON IMAGE PROCESSING, JUNE 2016 Estimation of Gaussian, Poissonian-Gaussian, and Processed Visual Noise and its Level Function Meisam Rakhshanfar, Student Member, IEEE, and Maria A. Amer, Senior Member, IEEE [5] Comparison of Various Filters for Noise Removal in MRI Brain Image 1 Anisha, S.R PG Scholar,2Dr J Venugopala Krishnan 1ME VLSI DESIGN,PG scholar,2professor/hod International Conference on Futuristic Trends in Computing and Communication (ICFTCC2015). Gurjeet Kaur [6] A REVIEW ON IMAGE DENOISING METHODS. International Journal of Advanced Engineering and Global Technology I Vol-03, Issue-11, December 2015. Dept.of Computer MGC, Fatehgarh sahib, India Dr Balkrishan Dept.of Computer YCOE, Talwani sabo, India Harmandeep singh Dept.of Computer COEM, Rampura Phul, India Gurjeet Kaur Dept.of Computer MGC, Fatehgarh sahib, India Page 69