You are on page 1of 99



A Project Report Submitted in partial fulfillment of the Requirement for the award of the Degree of Master of Technology IN Communication Engineering & Signal Processing By SIDDABATHUNI RAMATULASI (Y11MTEC813) Under the guidance of

P.SIVA PRASAD, M.Tech Assistant professor

Department of Electronics & Communication Engineering R.V.R. & J.C.COLLEGE OF ENGINEERING

AUTONOMOUS (Approved by A.I.C.T.E)

(Affiliated to Acharya Nagarjuna University)

Chandramoulipuram GUNTUR 522 019, Andhra Pradesh, INDIA [2011-2013]

Department of Electronics & Communication Engineering


TRIMMED MEDIAN FILTER FOR VIDEO RESTORATION that is being submitted by Siddabathuni Ramatulasi (Y11MTEC813) in partial fulfillment for the award of the Degree of Master of Technology in Communication Engineering & Signal Processing to the Acharya Nagarjuna University is a record of bonafide work carried out by her under my guidance and supervision. The results embodied in this project report have not been submitted to any other University or Institute for the award of any degree or diploma.


Signature of Guide P.SIVA PRASAD, M.Tech Assistant Professor

Signature of HOD Dr.V.V.K.D.V.PRASAD, Ph.D Professor & Head

I would like to express my sincere gratitude to my project guide, P.SIVA PRASAD, for his advice, encouragement and his help to solve many practical problems and his patience to answer my many questions. His deep understanding and immense knowledge helped me solve many difficult problems. He guided me throughout this project, even during his extremely busy schedule. I am also grateful to my Head of Department Dr.V.V.K.D.V.PRASAD, for his support and supervision. And also I sincerely thank to Dr.A.SUDHAKAR, Principal for providing all facilities to done the project successfully. This master thesis project has been a very valuable learning experience. It has given me the chance to learn better ways of achieving goals from more experienced personnel. Above all, the most important asset I have taken from this experience is the willingness to learn. The working atmosphere and especially the nice persons of this department have encouraged me in my work. I would like to thank them all for their hospitality. I would like to thank all the lecturers and technical staff in Electronics and Communications Department for their warm hearted support during difficult times. Last but not least to my parents, I extend my deepest love. They have always motivated me to continue my higher studies.

TABLE OF CONTENTS Chapter No Description Abstract List of Figures List of Symbols List of Abbreviations Chapter 1 INTRODUCTION 1.1 Introduction 1.2 Literature Survey Chapter 2 DIGITAL IMAGE PROCESSING 2.1 Introduction 2.2 Digital image 2.2.1 Types of digital images Black and white images Color images 2.3 Image file sizes 2.4 Image file formats 2.5 Digital image processing 2.6 Advantages of Digital Image Processing 2.7 Disadvantages of Digital Image Processing 2.8 Fundamental steps in digital image processing 2.8.1 Image acquisition 2.8.2 Image enhancement 2.8.3 Image restoration Page No i ii v vi 1 2 3 7 8 8 10 10 10 11 11 12 13 13 14 14 15 16

2.9 Color image processing 2.9.1 Segmentation 2.9.2 Image compression 2.10 Project description Chapter 3 DIGITAL VIDEO 3.1 Introduction 3.2 History 3.3 Advantages of Digital video 3.4 Frame separation 3.4.1 Number of Frames per Second 3.4.2 Interlaced Vs Progressive 3.5 Video quality Chapter 4 NOISE 4.1 Introduction 4.2 Noise 4.3 Types of Noise 4.3.1 Short Noise 4.3.2 Gaussian Noise 4.3.3 White Noise 4.3.4 Additive White Gaussian Noise 4.3.5 Poison Noise 4.3.6 Speckle Noise 4.3.7 Salt and Pepper Noise 4.4 Noise Generation 4.4.1 Signal to Noise Ratio 35

17 17 18 19 21 22 25 27 29 29 30 31 32 33 33 34

35 36 36 37 37 38 38 39

4.4.2 Peak Signal to Noise Ratio 4.5 Detector Noise 4.6 Crimmins Speckle Removal Chapter 5 IMAGE RESTORATION 5.1 Introduction 5.2 Restoration 5.3 Requirements for Restoration 5.4 Degradation 5.4.1 Image Degradation 5.4.2 Image Degradation for Blur Parameters 5.5 Thresholding Chapter 6 REMOVING OF HIGH DENSITY SALT AND PEPPER NOISE USING DIFFERENT FILTERS 6.1 Introduction 6.2 Salt and Pepper Noise 6.3 Disadvantages of Salt and Pepper Noise 6.4. Mean Filter 6.5 Median Filter 6.5.1 Advantages of Median Filter 6.5.2 Disadvantage of the Median Filter 6.5.3 Comparison between the Median Filter and the Average filter 6.6 Removing of Salt and Pepper Noise using Different Filters 6.6.1 Standard Median Filter 45

40 41 42 44

45 46 46 47 47 49

50 51 51 52 52 53 55 56


59 59

6.6.2 Adaptive Median Filter 6.6.3 Tolerance Based Switched Median Filter 6.6.4 Decision Based Algorithm 6.6.5 Unsymmetric Trimmed Median Filter 6.6.6 Decision Based Unsymmetric Trimmed Median Filter 6.6.7 Modified Decision based Unsymmetric Trimmed Median Filter Chapter 7 APPLICATIONS

60 61 62 62


64 66 67 68 69 70 81 83

7.1 Photoshop 7.2 Satellites 7.3 Medical line process Chapter 8 Chapter 9 SIMULATION RESULTS CONCLUSIONS


It is important to remove or minimize the degradations, noises in valuable ancient blurred color images. The traditional available filtering methodologies are applicable for fixed widow dimensions only these are not applicable for varying scale images. In this project we propose a new technique for digital image restoration, in this the noise free and noisy pixels are classified based on empirical multiple threshold values. Then the median filtering technique is applied. So that noise free pixels are getting preserved and only noisy pixels get restored. In this project, a novel decision-based filter, called the multiple thresholds switching (MTS) filter, is proposed to restore images corrupted by salt-pepper impulse noise. The filter is based on a detection-estimation strategy. The impulse detection algorithm is used before the filtering process, and therefore only the noisecorrupted pixels are replaced with the estimated central noise-free ordered mean value in the current filter window. The new impulse detector, which uses multiple thresholds with multiple neighborhood information of the signal in the filter window, is very precise, while avoiding an undue increase in computational complexity. For impulse noise suppression without smearing fine details and edges in the image, extensive experimental results demonstrate that our scheme performs significantly better than many existing, well-accepted decision-based methods. The performance of our proposed algorithm will be analyzed based PSNR and MSE values.


Figure No Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 2.7 Figure 2.8 Figure 2.9 Figure 2.10 Figure 2.11 Figure 2.12 Figure 3.1 Figure 4.1

Figure Title Normal and Pesedo-photo graph Image Pixel value of a Image Color values form 32-bit table Metafile format Images Digital Image Processing block diagram Digital Camera Scanners Enhancement Image Image Restoration Black and white to color image Image Segmentation Block diagram of proposed scheme Example of frame separation 1D Gaussian distributions with mean 0 and Standard deviation 1

Page No 8 9 10 12 14 15 15 16 16 17 17 20 29

41 48 49 53 54 55 56

Figure 5.1 Figure 5.2 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5

Blur Length Blur Angle 33 averaging kernel often used in mean filtering Illustrates an example of median filtering Calculating the median value of a pixel neighborhood The original image and the same image of median filter Comparison of the nonlinear median filter and the linear

mean filter. Figure 6.6 Figure 6.7 Figure 6.8 Figure 6.9 Figure 6.10 Standard median filter Adaptive median filter Tolerance based switched median filter Decision based algorithm Modified decisions based un-symmetric trimmed median Filter Figure 8.1 Figure 8.2 Figure 8.3 Figure 8.4 Figure 8.5 Figure 8.6 Figure 8.7 Figure 8.8 Figure 8.9 Figure 8.10 Figure 8.11 Figure 8.12 Figure 8.13 Figure 8.14 Figure 8.15 Figure 8.16 Figure 8.17 Project execution first step Select the video sample Browsing the video sample Frame separation Frame selection Selecting the frame Generating the noise in first level Generating the noise in second level Generating the noise in third level Generating the noise in fourth level Generating the noise in fifth level The output of the standard median filter The output of the adaptive median filter The output of the tolerance based switched median filter The output of the decision based algorithm The output values of the existing methods The output of the decision based unsymmetric trimmed Median filter

57 59 61 62 62

65 71 72 72 73 73 74 74 75 75 76 76 77 77 78 78 79


Figure 8.18

The output of the decision based unsymmetric trimmed Median filter for video Restoration 80 80

Figure 8.19

The output value of the modified method

LIST OF SYMBOLS Noise Variance


Noisy Approximation




Center Weighted Median Filter Progressive Switching Median Filter Portable Network Graphics Joint Photographic Experts Group Graphics Interchange Format Television Computed Tomography Magnetic Resonance Imaging Peak Signal-to-Noise Ratio Image Enhancement Factor Mean Square Error Frames per Second Compact Disc Non-linear Editing Workstation High Definition Television Moving Picture Experts Group Digital Video Disc Central Processing Unit Time Base Correctors Nippon Electric Corporation


Digital Video Effects Ampex Digital Optics Video Tape Recorders Electronic Field Production High Definition Video Phase Alternating Line Sequential Couleur Avec Memoire National Television Standards Committee Cathode Ray Tube Liquid Crystal Display Additive White Gaussian Noise Signal to Noise Ratio Peak Signal to Noise Ratio Point Spread Function Decision Based Algorithm Alpha Trimmed Midpoint Standard Median Filter Un-symmetric Trimmed Median Filter Adaptive Median Filter Modified Decision Based Algorithm


Modified Decision Based Unsymmetric Trimmed Median Filter


Fuzzy Noise Reduction Method Photoshop Document Photoshop Big Random Access Memory Read only Memory Adaptive Two-pass Median Filter Support Vector Machines


1.1 Introduction: In image processing it is usually necessary to perform high degree of noise reduction in an image before performing higher-level processing steps, such as edge detection. The median filter is a non-linear digital filtering technique, often used to remove noise from images or other signals. The idea is to examine a sample of the input and decide if it is representative of the signal. This is performed using a window consisting of an odd number of samples. The values in the window are sorted into numerical order; the median value, the sample in the center of the window, is selected as the output. The oldest sample is discarded, a new sample acquired, and the calculation repeats. Median filtering is a common step in image processing. It is particularly useful to reduce speckle noise and salt and pepper noise. Its edge-preserving nature makes it useful in cases where edge blurring is undesirable Image synthesis is the process of creating new images from some form of image description. The kinds of images that are typically synthesized include Test Patterns - scenes with simple two dimensional geometric shapes. Image Noise - images containing random pixel values usually generated from specific parameterized distributions. Computer Graphics - scenes or images based on geometric shape descriptions. Often the models are three dimensional, but may also be two dimensional. Synthetic images are often used to verify the correctness of operators by applying them to known images. They are also often used for teaching purposes, as the operator output on such images is generally `clean', whereas noise and

uncontrollable pixel distributions in real images make it harder to demonstrate unambiguous results. The images could be binary, grey level or color. 1.2 Literature Survey: Median filters based on fuzzy rules and its application to image restoration: A novel median-type filter controlled by fuzzy rules is proposed in order to remove impulsive noises on signals such as images. Median filter is well known for removing impulsive noises but this filter distorts the fine structure of signals as well. The filter proposed here is obtained as a weighted sum of the input signal and the output of the median filter, and the weight is set based on fuzzy rules concerning the states of the input signal sequence. Moreover, this weight is obtained optimally by a learning method, so that the mean square error of the filter output for some training signal data can be the minimum. Some results of image processing show the high performance of this filter. Moreover, the influences of the training signal on the filter performance. Selective Removal of Impulse Noise Based on Homogeneity Level Information: We propose a decision-based, signal adaptive median filtering algorithm for removal of impulse noise. Our algorithm achieves accurate noise detection and high SNR measures without smearing the fine details and edges in the image. The notion of homogeneity level is defined for pixel values based on their global and local statistical properties. The co-occurrence matrix technique is used to represent the correlations between a pixel and its neighbors, and to derive the upper and lower bound of the homogeneity level. Noise detection is performed at two stages: noise candidates are first selected using the homogeneity level, and then a refining process follows to eliminate false detections. The noise detection scheme does not use a quantitative decision measure, but uses qualitative structural information, and it is not

subject to burdensome computations for optimization of the threshold values. Empirical results indicate that our scheme performs significantly better than other median filters, in terms of noise suppression and detail preservation. A new efficient approach for the removal of impulse noise from highly corrupted images: In this paper, a novel adaptive filter, called the adaptive two-pass median (ATM) filter based on support vector machines (SVMs), is proposed to preserve more image details while effectively suppressing impulse noise for image restoration. The proposed filter is composed of a noise decision-maker and two pass median filters. Our new approach basically uses an SVM impulse detector to judge whether the input pixel is noise or not. If a pixel is detected as a corrupted pixel, the noise-free reduction median filter will be triggered to replace it. Otherwise, it keeps unchanged. Then, to improve the quality of the restored image, a decision impulse filter is put to work in the second pass filtering procedure. As for the noise suppressing on both fixed-valued and random-valued impulses without degrading the quality of the fine details, the results of our extensive experiments demonstrate that the proposed filter outperforms earlier median-based filters in the literature. In addition, our new filter also provides excellent robustness at various percentages of impulse noise.

Application of partition-based median type filters for suppressing noise in images:

An adaptive median based filter is proposed for removing noise from images. Specifically, the observed sample vector at each pixel location is classified into one of M mutually exclusive partitions, each of which has a particular filtering operation. The observation signal space is partitioned based an the differences defined between the current pixel value and the outputs of CWM (center weighted median) filters with variable center weights. The estimate at each location is formed as a linear combination of the outputs of those CWM filters and the current pixel value. To control the dynamic range of filter outputs, a location-invariance constraint is imposed upon each weighting vector. The weights are optimized using the constrained LMS (least mean square) algorithm. Recursive implementation of the new filter is then addressed. The new technique consistently outperforms other median based filters in suppressing both random-valued and fixed-valued impulses, and it also works satisfactorily in reducing Gaussian noise as well as mixed Gaussian and impulse noise.

A noise-filtering method using a local information measure: A nonlinear-noise filtering method for image processing, based on the entropy concept is developed and compared to the well-known median filter and to the center weighted median filter (CWM). The performance of the proposed method is evaluated through subjective and objective criteria. It is shown that this method performs better than the classical median for different types of noise and can perform better than the CWM filter in some cases.

Progressive Switching Median Filter for the Removal of Impulse Noise from Highly Corrupted Images: A new median-based lter, progressive switching median (PSM) lter, is proposed to restore images corrupted by saltpepper impulse noise. The algorithm is developed by the following two main points: Switching scheme an impulse detection algorithm is used before ltering, thus only a proportion of all the pixels will be ltered. Progressive methods both the impulse detection and the noise ltering procedures are progressively applied through several iterations. Simulation results demonstrate that the proposed algorithm is better than traditional median-based lters and is particularly effective for the cases where the images are very highly corrupted.


2.1 Introduction: When using digital equipment to capture, store, modify and view photographic images, they must first be converted to a set of numbers in a process called digitization or scanning. Computers are very good at storing and manipulating numbers, so once your image has been digitized you can use your computer to archive, examine, alter, display, transmit, or print your photographs in an incredible variety of ways. 2.2 Digital Image: An image is a two-dimensional picture, which has a similar appearance to some subject usually a physical object or a person. Image is a two-dimensional, such as a photograph, screen display, and as well as a three-dimensional, such as a statue. They may be captured by optical devices such as cameras, mirrors, lenses, telescopes, microscopes, etc. and natural objects and phenomena, such as the human eye or water surfaces.

Fig 2.1 Normal and Pesedo-photo graph Image

The word image is also used in the broader sense of any two-dimensional figure such as a map, a graph, a pie chart, or an abstract painting. In this wider sense,

images can also be rendered manually, such as by drawing, painting, carving, rendered automatically by printing or computer graphics technology, or developed by a combination of methods, especially in a pseudo-photograph. An image is a rectangular grid of pixels. It has a definite height and a definite width counted in pixels. Each pixel is square and has a fixed size on a given display. However different computer monitors may use different sized pixels. The pixels that constitute an image are ordered as a grid (columns and rows); each pixel consists of numbers representing magnitudes of brightness and color.

Fig 2.2 Pixel value of a Image Each pixel has a color. The color is a 32-bit integer. The first eight bits determine the redness of the pixel, the next eight bits the greenness, the next eight bits the blueness, and the remaining eight bits the transparency of the pixel.

Fig 2.3 Colour values form 32-bit table 2.2.1 Types of Digital Images: For photographic purposes, there are two important types of digital images color and black and white. Color images are made up of colored pixels while black and white images are made of pixels in different shades of gray. Black and White: A black and white image is made up of pixels each of which holds a single number corresponding to the gray level of the image at a particular location. These gray levels span the full range from black to white in a series of very fine steps, normally 256 different grays. Since the eye can barely distinguish about 200 different gray levels. Assuming 256 gray levels, each black and white pixel can be stored in a single byte (8 bits) of memory. Color Images: A color image is made up of pixels each of which holds three numbers corresponding to the red, green, and blue levels of the image at a particular location. Red, green, and blue (sometimes referred to as RGB) are the primary colors for mixing lights these so called additive primary colors are different from the subtractive primary colors used for mixing paints (cyan, magenta, and yellow). Any color can be created by mixing the correct amounts of red, green, and blue light. Assuming 256 levels for each primary, each color pixel can be stored in three bytes (24 bits) of memory. This corresponds to roughly 16.7 million different possible colors.

2.3 Image File Sizes: Image file size is expressed as the number of bytes that increases with the number of pixels composing an image, and the color depth of the pixels. The greater the number of rows and columns, the greater the image resolution, and the larger the file. Also, each pixel of an image increases in size when its color depth increases, an 8-bit pixel (1 byte) stores 256 colors, a 24-bit pixel (3 bytes) stores 16 million colors, the latter known as true color. Image compression uses algorithms to decrease the size of a file. High resolution cameras produce large image files, ranging from hundreds of kilobytes to megabytes, per the camera's resolution and the image-storage format capacity. High resolution digital cameras record 12 megapixel (1MP = 1,000,000 pixels / 1 million) images, or more, in true color. For example, an image recorded by a 12 MP camera; since each pixel uses 3 bytes to record true color, the uncompressed image would occupy 36,000,000 bytes of memory, a great amount of digital storage for one image, given that cameras must record and store many images to be practical. Faced with large file sizes, both within the camera and a storage disc, image file formats were developed to store such large images. 2.4 Image File Formats: Image file formats are standardized means of organizing and storing images. This entry is about digital image formats used to store photographic and other images. Image files are composed of either pixel or vector (geometric) data that are rasterized to pixels when displayed (with few exceptions) in a vector graphic display. Including proprietary types, there are hundreds of image file types. The PNG, JPEG, and GIF formats are most often used to display images on the Internet.

Fig 2.4 Metafile format Images In addition to straight image formats, Metafile formats are portable formats which can include both raster and vector information. The metafile format is an intermediate format. Most Windows applications open metafiles and then save them in their own native format. 2.5 Digital Image Processing: The field of digital image processing refers to processing digital images by means of a digital computer. Note that a digital image is composed of a finite number of elements, each of which has a particular location and value. These elements are referred to as picture elements, image elements, pels, and pixels. Pixel is the term most widely used to denote the elements of a digital image. Vision is the most advanced of our senses, so it is not surprising that images play the single most important role in human perception. Digital image processing is a subcategory or field of digital signal processing, digital image processing has many advantages over analog image processing. It allows a much wider range of algorithms to be applied to the input data and can avoid problems such as the build-up of noise and signal distortion during processing. Since

images are defined over two dimensions (perhaps more) digital image processing may be modeled in the form of multidimensional systems. 2.6 Advantages of Digital Image Processing: The processing of images is faster and more cost-effective. One needs less time for processing, as well as less film and other photographing equipment. When shooting a digital image, one can immediately see if the image is good or not. By changing the image format and resolution, the image can be used in a number of media. Digital image processing made digital image can be noise free. Digital imaging is the ability of the operator to post process the image. It means manipulate the pixel shades to correct the image density and contrast. Digital imaging allows the electronic transmission of images to third party providers. The expensive reproduction is faster and cheaper. 2.7 Disadvantages of Digital Image Processing: The initial cost can be high depending on the system used. Misuse of copyright is now easier than it earlier was. For instance, images can be copied from the Internet just by clicking the mouse a couple of times. Work has become more technical, which may not be a disadvantage for everyone. If computer is crashes then pics that have not been printed and filed into book albums that are lost.

2.8. Fundamental Steps in Digital Image Processing:

Fig 2.5 Digital Image Processing block diagram 2.8.1 Image Acquisition: Image Acquisition is to acquire a digital image. To do so requires an image sensor and the capability to digitize the signal produced by the sensor. The sensor could be monochrome or color TV camera that produces an entire image of the problem domain every 1/30 sec. the image sensor could also be line scan camera that produces a single image line at a time. In this case, the objects motion past the line.

Fig 2.6 Digital camera

Scanner produces a two-dimensional image. If the output of the camera or other imaging sensor is not in digital form, an analog to digital converter digitizes it. The nature of the sensor and the image it produces are determined by the application.

Fig 2.7 Scanners

2.8.2 Image Enhancement: Image enhancement is among the simplest and most appealing areas of digital image processing. Basically, the idea behind enhancement techniques is to bring out detail that is obscured, or simply to highlight certain features of interesting an image. A familiar example of enhancement is when we increase the contrast of an image because it looks better. It is important to keep in mind that enhancement is a very subjective area of image processing.

Fig 2.8 Enhancement Image

2.8.3 Image restoration: Image restoration is an area that also deals with improving the appearance of an image. However, unlike enhancement, which is subjective, image restoration is objective, in the sense that restoration techniques tend to be based on mathematical or probabilistic models of image degradation.

Fig 2.9 Image Restoration

Enhancement, on the other hand, is based on human subjective preferences regarding what constitutes a good enhancement result. For example, contrast stretching is considered an enhancement technique because it is based primarily on the pleasing aspects it might present to the viewer, where as removal of image blur by applying a deblurring function is considered a restoration technique. 2.9 Colour Image Processing: The use of colour in image processing is motivated by two principal factors. First, colour is a powerful descriptor that often simplifies object identification and extraction from a scene. Second, humans can discern thousands of color shades and

intensities, compared to about only two dozen shades of gray. This second factor is particularly important in manual image analysis.

Fig 2.10 Black and white to colour image

2.9.1 Segmentation: Segmentation procedures partition an image into its constituent parts or objects. In general, autonomous segmentation is one of the most difficult tasks in digital image processing. A rugged segmentation procedure brings the process a long way toward successful solution of imaging problems that require objects to be identified individually.

Fig 2.11 Image Segmentation

On the other hand, weak or erratic segmentation algorithms almost always guarantee eventual failure. In general, the more accurate the segmentation, the more likely recognition is to succeed. Digital image is defined as a two dimensional function f(x, y), where x and y are spatial (plane) coordinates, and the amplitude of f at any pair of coordinates (x, y) is called intensity or grey level of the image at that point. The field of digital image processing refers to processing digital images by means of a digital computer. The digital image is composed of a finite number of elements, each of which has a particular location and value. The elements are referred to as picture elements, image elements, pels, and pixels. Pixel is the term most widely used. 2.9.2 Image Compression: Digital Image compression addresses the problem of reducing the amount of data required to represent a digital image. The underlying basis of the reduction process is removal of redundant data. From the mathematical viewpoint, this amounts to transforming a 2D pixel array into a statically uncorrelated data set. The data redundancy is not an abstract concept but a mathematically quantifiable entity. If n1 and n2 denote the number of information-carrying units in two data sets that represent the same information, the relative data redundancy RD [2] of the first data set (the one characterized by n1) can be defined as,

RD 1

1 CR

Where C R called as compression ratio [2]. It is defined as

CR =

n1 n2

In image compression, three basic data redundancies can be identified and exploited: Coding redundancy, inter pixel redundancy, and phychovisal redundancy.

Image compression is achieved when one or more of these redundancies are reduced or eliminated. The image compression is mainly used for image transmission and storage. Image transmission applications are in broadcast television; remote sensing via satellite, air-craft, radar, or sonar; teleconferencing; computer communications; and facsimile transmission. Image storage is required most commonly for educational and business documents, medical images that arise in computer tomography (CT), magnetic resonance imaging (MRI) and digital radiology, motion pictures, satellite images, weather maps, geological surveys, and so on. 2.10 Project Description: According to recent literatures introduced, a modified decision based unsymmetrical trimmed median filter for the video restoration of gray scale, and color images [10]. This filter removes the high density salt and pepper noise and restores the gray scale, and color frames that are highly corrupted by salt and pepper noise. This filter gives better Peak Signal-to-Noise Ratio (PSNR) and Image Enhancement Factor (IEF). The Filter can remove the high density noise, its computational speed is also higher than existing filters, it will take the feedback from noisy or corrupted pixels, and tested against Different color video frames and it gives better Peak Signal-toNoise Ratio (PSNR) and Mean Square Error (MSE).

Block Diagram:

Input Video

Frame Separation

Input Frame

Addition of Noise

Application of Diff Filters


Fig 2.12 Block diagram of proposed scheme


3.1 Introduction of Digital Video: Digital video refers to the capturing, manipulation, and storage of moving images that can be displaced on computer screens. This requires that the moving images be digitally handled by the computer. The word digital refers to a system based on discontinuous events, as opposed to analog, a continuous event. Computers are digital systems; they do not process images the way the human eye does. Before the Digital Era, to display analog video images on a computer monitor, the video signal had to first be converted from analog to digital form. A special video digitalizing overlay board or hardware on the motherboard had to be installed in your computer to take the video signal and convert it to digital information. To do this, however, required a very powerful computer to be able to read and digitalize every frame repetitively. So the next step in digital video evolution was to eliminate the analog videotape. Thus, the entire procedure, including the capturing of video, is in digital form. First, a camera and a microphone capture the picture and sound of a video session and send analog signals to a video-capture adapter board. The board only captures half of the number of frames per second that movies use in order to reduce the amount of data to be processed. Second, there is an analog-to-digital converter chip on the video-capture adapter card, and it converts the analog signals to digital patterns (0s and 1s). Third, a compression/decompression chip or software reduces the data to a minimum necessary for recreating the video signals. In this procedure, no analog was involved, making the process more efficient. Digital video comprises a series of orthogonal bitmap digital images displayed in rapid succession at a constant rate. In the context of video these images are called frames. We measure the rate at which

frames are displayed in frames per second (FPS).Since every frame is an orthogonal bitmap digital image it comprises a raster of pixels. If it has a width of W pixels and a height of H pixels we say that the frame size is WxH. Pixels have only one property, their colour. The colour of a pixel is represented by a fixed number of bits. The more bits the more subtle variations of colours can be reproduced. This is called the colour depth (CD) of the video. Digital video cameras come in two different image capture formats: interlaced and deinterlaced progressive scan. Interlaced cameras record the image in alternating sets of lines: the odd-numbered lines are scanned, and then the even-numbered lines are scanned, then the odd-numbered lines are scanned again, and so on. One set of odd or even lines is referred to as a "field", and a consecutive pairing of two fields of opposite parity is called a frame. Deinterlaced cameras records each frame as distinct, with all scan lines being captured at the same moment in time. Thus, interlaced video captures samples the scene motion twice as often as progressive video does, for the same number of frames per second. Progressive-scan camcorders generally produce a slightly sharper image. However, motion may not be as smooth as interlaced video which uses 50 or 59.94 fields per second, particularly if they employ the 24 frames per second standard of film. Digital video can be copied with no degradation in quality. No matter how many generations of a digital source is copied, it will still be as clear as the original first generation of digital footage. However a change in parameters like frame size as well as a change of the digital format can decrease the quality of the video due to new calculations that have to be made. Digital video can be manipulated and edited to follow an order or sequence on an NLE or non-linear editing workstation, a computer based device intended to edit video and audio. More and more, videos are edited on

readily available, increasingly affordable consumer-grade computer hardware and software. However, such editing systems require ample disk space for video footage. The many video formats and parameters to be set make it quite impossible to come up with a specific number for how many minutes need how much time. Digital video has a significantly lower cost than 35 mm film. The tape stock itself is very inexpensive. Digital video also allows footage to be viewed on location without the expensive chemical processing required by film. Also physical deliveries of tapes and broadcasts do not apply anymore. Digital television (including higher quality HDTV) started to spread in most developed countries in early 2000s. Digital video is also used in modern mobile phones and video conferencing systems. Digital video is also used for Internet distribution of media, including streaming video and peer-to-peer movie distribution. However even within Europe are lots of TV-Stations not broadcasting in HD, due to restricted budgets for new equipment for processing HD.

Many types of video compression exist for serving digital video over the internet and on optical disks. The file sizes of digital video used for professional editing are generally not practical for these purposes, and the video requires further compression with codecs such as Sorenson, H.264 and more recently Apple ProRes especially for HD. Probably the most widely used formats for delivering video over the internet are MPEG4, Quicktime, Flash and Windows Media, while MPEG2 is used almost exclusively for DVDs, providing an exceptional image in minimal size but resulting in a high level of CPU consumption to decompress.

As of 2011, the highest resolution demonstrated for digital video generation is 35 megapixels (8192 x 4320). The highest speed is attained in industrial and scientific

high speed cameras that are capable of filming 1024x1024 video at up to 1 million frames per second for brief periods of recording.

3.2 History:

Starting in the late 1970s to the early 1980s, several types of video production equipment were introduced, such as time base correctors (TBC) and digital video effects (DVE) units (one of the former being the Thomson-CSF 9100 Digital Video Processor, an internally all-digital full-frame TBC introduced in 1980, and two of the latter being the Ampex ADO, and the Nippon Electric Corporation (NEC) (DVE). They operated by taking a standard analog composite video input and digitizing it internally. This made it easier to either correct or enhance the video signal, as in the case of a TBC, or to manipulate and add effects to the video, in the case of a DVE unit. The digitized and processed video information from these units would then be converted back to standard analog video.

Later on in the 1970s, manufacturers of professional video broadcast equipment, such as Bosch (through their Fernseh division), RCA, and Ampex developed prototype digital videotape recorders (VTR) in their research and development labs. Bosch's machine used a modified 1" Type B transport, and recorded an early form of CCIR 601 digital video. Ampex's prototype digital video recorder used a modified 2" Quadruplex VTR (an Ampex AVR-3), but fitted with custom digital video electronics, and a special "octaplex" 8-head headwheel (regular analog 2" Quad machines only used 4 heads). The audio on Ampex's prototype digital machine, nicknamed by its developers as "Annie", still recorded the audio in analog as linear tracks on the tape, like 2" Quad. None of these machines from these manufacturers were ever marketed commercially, however.

Digital video was first introduced commercially in 1986 with the Sony D-1 format, which recorded an uncompressed standard definition component video signal in digital form instead of the high-band analog forms that had been commonplace until then. Due to its expense, D-1 was used primarily by large television networks. It would eventually be replaced by cheaper systems using video compression, most notably Sony's Digital Betacam (still heavily used as an electronic field production (EFP) recording format by professional television producers) that were introduced into the network's television studios.

One of the first digital video products to run on personal computers was PACo the PICS Animation Compiler from The Company of Science & Art in Providence, RI, which was developed starting in 1990 and first shipped in May 1991. PACo could stream unlimited-length video with synchronized sound from a single file on CD ROM. Creation required a Mac; playback was possible on Macs, PCs, and Sun Sparcstations. In 1992, Bernard Luskin, Philips Interactive Media, and Eric Doctorow, Paramount Worldwide Video, successfully put the first fifty videos in digital MPEG 1 on CD, developed the packaging and launched movies on CD, leading to advancing versions of MPEG, and to DVD.

QuickTime, Apple Computer's architecture for time-based and streaming data formats appeared in June, 1991. Initial consumer-level content creation tools were crude, requiring an analog video source to be digitized to a computer-readable format. While low-quality at first, consumer digital video increased rapidly in quality, first with the introduction of playback standards such as MPEG-1 and MPEG-2 (adopted for use in television transmission and DVD media), and then the introduction of the DV tape format allowing recording direct to digital data and simplifying the editing

process, allowing non-linear editing systems (NLE) to be deployed cheaply and widely on desktop computers with no external playback/recording equipment needed.

The widespread adoption of digital video has also drastically reduced the bandwidth needed for a high-definition video signal (with HDV and AVCHD, as well as several commercial variants such as DVCPRO-HD, all using less bandwidth than a standard definition analog signal) and tapeless camcorders based on flash memory and often a variant of MPEG-4.

3.3 Advantages of Digital Video: What is it about digital video that makes it so attractive? Isnt videotape good enough? Here are three of many reasons that explain why digital videos are becoming more popular than ever. Ease of Manipulation is the difference between analog and digital is like comparing a typewriter with a word processor. Just like the cut and paste function is much easier and Faster with a word processor, editing is easier and faster with a digital video. Also, many effects that were exclusive for specialized post production houses are now easily achieved by bringing in files from Photoshop, Flash, and Sound Edit as components in a video mix. In addition, the ability to separate sound from image enables editing one without affecting the other. Preservation of Data is not true that DV is better simply because it is digital. Big screen films are not digital and are still highly esteemed as quality images. However, it is easier to maintain the quality of a digital video. Traditional tapes are subject to wear and tear more so than DVD or hard drive disks. Also, once done, a digital video can be copied over and over without losing its

original information. Analog signals can be easily distorted and will lose much of the original data after a few transfers. Internet is a digital video can be sent via the Internet to countless end users without having to make a copy for every viewer. It is easy to store, retrieve, and publish. Compression of a digital video files can be very large. For example, one single frame from a television image with a resolution of 720 x 576 pixels and a color depth of 16 bits has a size of 1.35 MB (Fisher & Schroeder, 1999). Multiply that by 25 frame per second and then by the duration of a movie! It is not practicalsometimes impossibleto have videos of this size. Thus compression, the process of reducing file size by eliminating unnecessary data for reconstruction purposes, is a must. There are two types of compression, lossless and lossy. The lossless compression retains the original data so that the individual image sequences remain the same. It saves space by removing image areas that use the same color. The compression rate is usually no better than 3:1 (Fisher & Schroeder.). The low rate makes most lossless compression less desirable. The lossy compression methods remove image and sound information that is unlikely to be noticed by the viewer. Some information is lost, but since it is not differentiated by the human perception, the quality perceived is still the same, while the volume is dramatically decreased. There are many compression formats. Here are the few most widely used. 3.4 Frame Separation: Frame processing is the main step in the modified decision based unsymmetric trimmed median filter. Frame rate (also known as frame frequency) is the frequency (rate) at which an imaging device produces unique consecutive images

called frames. The term applies equally well to film and video cameras, computer graphics, and motion capture systems. Frame rate is most often expressed in frames per second (FPS) and is also expressed in progressive scan monitors as hertz(Hz). After frame separation we can get number frames in our data base. We have to select one frame from that number of frames. For that particular we have to generate the noise to measure the performance.

Fig 3.1 Example of frame separation 3.4.1 Number Of Frames Per Second: Frame rate, the number of still pictures per unit of time of video, ranges from six or eight frames per second (frames) for old mechanical cameras to 120 or more frames per second for new professional cameras. PAL (Europe, Asia, Australia, etc.) and SECAM (France, Russia, parts of Africa etc.) standards specify 25 frame/s, while NTSC (USA, Canada, Japan, etc.) specifies 29.97 frame/s. Film is shot at the slower frame rate of 24 photograms, which complicates slightly the process of transferring a cinematic motion picture to video. The minimum frame rate to achieve the illusion of a moving image is about twelve to fifteen frames per second.

3.4.2 Interlaced Vs Progressive Video can be interlaced or progressive. Interlacing was invented as a way to reduce flicker in early mechanical and CRT video displays without increasing the number of complete frames per second, which would have required sacrificing image detail in order to remain within the limitations of a narrow bandwidth. The horizontal scan lines of each complete frame are treated as if numbered consecutively and captured as two fields an odd field (upper field) consisting of the odd-numbered lines and an even field (lower field) consisting of the even-numbered lines. Analog display devices reproduce each frame in the same way, effectively doubling the frame rate as far as perceptible overall flicker is concerned. When the image capture device acquires the fields one at a time, rather than dividing up a complete frame after it is captured, the frame rate for motion is effectively doubled as well, resulting in smoother, more lifelike reproduction (although with halved detail) of rapidly moving parts of the image when viewed on an interlaced CRT display, but the display of such a signal on a progressive scan device is problematic. In progressive scan systems, each refresh period updates all of the scan lines of each frame in sequence. When displaying a natively progressive broadcast or recorded signal, the result is optimum spatial resolution of both the stationary and moving parts of the image. When displaying a natively interlaced signal, however, overall spatial resolution will be degraded by simple line doubling and artifacts such as flickering or "comb" effects in moving parts of the image will be seen unless special signal processing is applied to eliminate them. A procedure known as deinterlacing can be used to optimize the display of an interlaced video signal from an analog, DVD or satellite source on a progressive scan device such as an LCD

Television, digital video projector or plasma panel. Deinterlacing cannot, however, produce video quality that is equivalent to true progressive scan source material. 3.5 Video Quality Video quality is a characteristic of a video passed through a video transmission/processing system, a formal or informal measure of perceived video degradation (typically, compared to the original video). Video processing systems may introduce some amounts of distortion or artefacts in the video signal, so video quality evaluation is an important problem. Since the time when the world's first video sequence was recorded, many video processing systems have been designed. In the ages of analog video systems, it was possible to evaluate quality of a video processing system by calculating the system's frequency response using some traditional test signal (for example, a collection of color bars and circles). Nowadays, digital video systems are replacing analog ones, and evaluation methods have changed. Performance of a digital video processing system can vary significantly and depends on dynamic characteristics of input video signal (e.g. amount of motion or spatial details).


4.1 Introduction In communication systems, the noise is an error or undesired random disturbance of a useful information signal, introduced before or after the detector and decoder. The noise is a summation of unwanted or disturbing energy from natural and sometimes man-made sources. Noise is, however, typically distinguished other unwanted

from interference,

(e.g. cross-talk,

deliberate jamming or

electromagnetic interference from specific transmitters), for example in the signal-tonoise ratio (SNR), signal-to-interference ratio (SIR) and signal-to-noise plus

interference ratio (SNIR) measures. Noise is also typically distinguished from distortion, which is an unwanted alteration of the signal waveform, for example in the signal-to-noise and distortion ratio (SINAD). In a carrier-modulated passband analog communication system, a certain carrier-to-noise ratio (CNR) at the radio receiver input would result in a certain signal-to-noise ratio in the detected message signal. While noise is generally unwanted, it can serve a useful purpose in some applications, such as random number generation or dithering. 4.2 Noise In common use the word noise means unwanted sound or noise pollution. In electronics noise can refer to the electronic signal corresponding to acoustic noise (in an audio system) or the electronic signal corresponding to the (visual) noise commonly seen as 'snow' on a degraded television or video image. In signal processing or computing it can be considered data without meaning; that is, data that is not being used to transmit a signal, but is simply produced as an unwanted byproduct of other activities. In Information Theory, however, noise is still considered to be information. In a broader sense, film grain or even advertisements in web pages

can be considered noise. Noise can block, distort, or change the meaning of a message in both human and electronic communication. In many of these areas, the special case of thermal noise arises, which sets a fundamental lower limit to what can be measured or signaled and is related to basic physical processes at the molecular level described by well known simple formulae. 4.3 Types of noises Noise is random, undesirable electrical energy that enters the communications system via the communicating medium and interferes with the transmitted message. However, some noise is produced in the receiver. Noise can be classified two categories 1. External noises: Noise whose sources are external. External noise may be classified into the following three types Atmospheric noises. Extraterrestrial noises. Man-made noises or industrial noises. 2. Internal noise: In communication, i.e. noises which get, generated within the receiver or communication system. Internal noise may be put into the following four categories. Thermal noise or white noise or Johnson noise. Shot noise. Transit time noise. Miscellaneous internal noise. External noise cannot be reduced except by changing the location of the receiver or the entire system. Internal noise on the other hand can be easily evaluated

mathematically and can be reduced to a great extent by proper design. As already said, because of the fact that internal noise can be reduced to a great extent, study of noise characteristics is a very important part of the communication engineering. 4.3.1 Shot Noise: Short noise is a type of electronic noise which originates from the discrete nature of electric charge. The term also applies to photon counting in optical devices, where shot noise is associated with the particle nature of light. Shot noise exists because phenomena such as light and electric current consist of the movement of discrete (also called "quantized") 'packets'. Consider light a stream of discrete photons coming out of a laser pointer and hitting a wall to create a visible spot. The fundamental physical processes that govern light emission are such that these photons are emitted from the laser at random times; but the many billions of photons needed to create a spot are so many that the brightness, the number of photons per unit time, varies only infinitesimally with time. However, if the laser brightness is reduced until only a handful of photons hit the wall every second, the relative fluctuations in number of photons, i.e., brightness, will be significant, just as when tossing a coin a few times. These fluctuations are shot noise. 4.3.2 Gaussian Noise: Gaussian noise is statistical noise that has a probability density function of the normal distribution (also known as Gaussian distribution). In other words, the values that the noise can take on are Gaussian-distributed. It is most commonly used as additive white noise to yield additive white Gaussian noise (AWGN).

4.3.3 White noise: In signal processing, white

noise is

random signal with


(constant) power spectral density. In other words, a signal that contains equal power within any frequency band with a fixed width. white noise refers to a statistical model for signals and signal sources, rather than to any specific signal. The term is also used for a discrete signal whose samples are regarded as a sequence of serially uncorrelated random variables with zero mean and

finite variance. Depending on the context, one may also require that the samples be independent and have the same probability distribution. The samples of a white noise signal may be sequential in time, or arranged along one or more spatial dimensions. In digital image processing, the samples (pixels) of a white noise image are typically arranged in a rectangular grid, and are assumed to be independent random variables with uniform probability

distribution over some interval. A random signal is considered "white noise" if it is observed to have a flat spectral power density over the visible band.

4.3.4 Additive White Gaussian Noise (AWGN): AWGN is a channel model in which the only impairment to communication is a linear addition of wideband or white noise with a constant spectral

density(expressed as watts per hertz of bandwidth) and a Gaussian distribution of amplitude. The model does not account for fading, frequency selectivity, interference, nonlinearity or dispersion. However, it produces simple and tractable mathematical models which are useful for gaining insight into the underlying behaviour of a system before these other phenomena are considered. Wideband Gaussian noise comes from many natural sources, such as the thermal vibrations of atoms in conductors (referred to as thermal noise or Johnson-

Nyquist noise), shot noise, black body radiation from the earth and other warm objects, and from celestial sources such as the Sun. The AWGN channel is a good model for many satellite and deep space communication links. It is not a good model for most terrestrial links because of multipath, terrain blocking, interference, etc. However, for terrestrial path modelling, AWGN is commonly used to simulate background noise of the channel under study, in addition to multipath, terrain blocking, interference, ground clutter and self interference that modern radio systems encounter in terrestrial operation. 4.3.5 Poisson Noise: Poisson noise has a probability density function of a Poisson distribution. That expresses the probability of a given number of events occurring in a fixed interval of time and/or space if these events occur with a known average rate

and independently of the time since the last event. The Poisson distribution can also be used for the number of events in other specified intervals such as distance, area or volume. 4.3.6 Speckle noise: Speckle noise is a granular noise that inherently exists in and degrades the quality of images. Speckle noise is a multiplicative noise, i.e. it is in direct proportion to the local grey level in any area. The signal and the noise are statistically independent of each other.

4.3.7 Salt & Pepper Noise: It represents itself as randomly occurring white and black pixels. An effective noise reduction method for this type of noise involves the usage of a median filter. Salt and pepper noise creeps into images in situations where quick transients, such as faulty switching, take place. The image after distortion from salt and pepper noise looks like the image attached. 4.4 Noise Generation: Noises are random background events which have to be dealt with in every system processing real signals. They are not part of the ideal signal and may be caused by a wide range of sources, e.g. variations in the detector sensitivity, environmental variations, the discrete nature of radiation, transmission or quantization errors, etc. The characteristics of noise depend on their source, as does the operator which best reduces their effects. Many image processing packages contain operators to artificially add noise to an image. Deliberately corrupting an image with noise allows us to test the resistance of an image processing operator to noise and assess the performance of various noise filters. Noise can generally be grouped in two classes

independent noise, and Noise which is dependent on the image data.

Image independent noise can often be described by an additive noise model, where the recorded image f(i,j) is the sum of the true image s(i,j) and the noise n(i,j).

The recorded image is give by the equation 4.1.

f(i,j) = s(i,j) + n(i,j) The noise n(i,j) is often zero-mean and described by its variance

(4.1) . The impact of the

noise on the image is often described by the signal to noise ratio (SNR), which is given by the equation 4.2






are the variances of the true image and the recorded image,

respectively. In many cases, additive noise is evenly distributed over the frequency domain (i.e. white noise), whereas an image contains mostly low frequency information. Hence, the noise is dominant for high frequencies and its effects can be reduced using some kind of lowpass filter. This can be done either with a frequency filter or with a spatial filter. In the second case of data dependent noise, (e.g. arising when monochromatic radiation is scattered from a surface whose roughness is of the order of a wavelength, causing wave interference which results in image speckle), it can be possible to model noise with a multiplicative, or non-linear, model. These models are mathematically more complicated, hence, if possible, the noise is assumed to be data independent. 4.4.1 Signal to Noise Ratio: Signal-to-noise ratio is a measure used in science and engineering that compares the level of a desired signal to the level of background noise. It is defined as the ratio of signal power to the noise power. A ratio higher than 1:1 indicates more signal than noise.

4.4.2 Peak Signal to Noise Ratio: Peak signal-to-noise ratio, often abbreviated PSNR, is an engineering term for the ratio between the maximum possible power of a signal and the power of corrupting noise that affects the fidelity of its representation. Because many signals have a very wide dynamic range, PSNR is usually expressed in terms of the logarithmic decibel scale. PSNR is most commonly used to measure the quality of reconstruction of lossy compression codecs (e.g., for image compression). The signal in this case is the original data, and the noise is the error introduced by compression. When comparing compression codecs, PSNR is an approximation to human perception of

reconstruction quality. Although a higher PSNR generally indicates that the reconstruction is of higher quality, in some cases it may not. One has to be extremely careful with the range of validity of this metric; it is only conclusively valid when it is used to compare results from the same codec (or codec type) and same content. PSNR is most easily defined via the mean squared error (MSE). Given a noise free mn monochrome image I and its noisy approximation K, MSE is defined as equation 4.3.



The PSNR is defined as equation 4.4.

PSNR in dB = 10


4.5 Detector Noise: One kind of noise which occurs in all recorded images to a certain extent is detector noise. This kind of noise is due to the discrete nature of radiation, i.e. the fact that each imaging system is recording an image by counting photons. Allowing some assumptions (which are valid for many applications) this noise can be modeled with an independent, additive model - where the noise n(i,j) has a zero-mean Gaussian distribution described by its standard deviation ( ), or variance. (The 1-D Gaussian distribution has the form shown in Figure 1.) This means that each pixel in the noisy image is the sum of the true pixel value and a random, Gaussian distributed noise value.

Fig 4.1 1D Gaussian distribution with mean 0 and standard deviation 1

4.6 Crimmins Speckle Removal: Crimmins Speckle Removal reduces speckle from an image using the Crimmins complementary hulling algorithm. The algorithm has been specifically designed to reduce the intensity of salt and pepper noise in an image. Increased iterations of the algorithm yield increased levels of noise removal, but also introduce a significant amount of blurring of high frequency details. Crimmins Speckle Removal works by passing an image through a speckle removing filter which uses the complementary hulling technique to reduce the speckle index of that image. The algorithm uses a non-linear noise reduction technique which compares the intensity of each pixel in an image with those of its 8 nearest neighbours and, based upon the relative values, increments or decrements the value of the pixel in question such that it becomes more representative of its surroundings. The noisy pixel alteration (and detection) procedure used by Crimmins is more complicated than the ranking procedure used by the non-linear median filter. It involves a series of pairwise operations in which the value of the `middle' pixel within each neighbourhood window is compared, in turn, with each set of neighbours (N-S, E-W, NW-SE, NE-SW) in a search for intensity spikes. For each iteration and for each pair of pixel neighbours, the entire image is sent to a Pepper Filter and Salt Filter. In the example case, the Pepper Filter is first called to determine whether the each image pixel is darker than i.e. by more than 2 intensity levels its northern neighbours. Comparisons where this condition proves true cause the intensity value of the pixel under examination to be incremented twice lightened, otherwise no change is affected. Once these changes have been recorded, the entire image is passed through the Pepper Filter again and the same series of comparisons are made between the current pixel and its southern neighbour. This

sequence is repeated by the Salt Filter, where the conditions lighter than and darken are, again, instantiated using 2 intensity levels. Over several iterations, the effects of smoothing in this way propagate out from the intensity spike to infect neighboring pixels. In other words, the algorithm smoothes by reducing the magnitude of a locally inconsistent pixel, as well as increasing the magnitude of pixels in the neighborhood surrounding the spike. It is important to notice that a spike is defined here as a pixel whose value is more than 2 intensity levels different from its surroundings. This means that after 2 iterations of the algorithm, the immediate neighbors of such a spike may themselves become spikes with respect to pixels lying in a wider neighborhood.


5.1 Introduction: Image restoration is the operation of taking a corrupted/noisy image and estimating the clean original image. Corruption may come in many forms such as motion blur, noise, and camera misfocus. Image restoration is different from image enhancement in that the latter is designed to emphasize features of the image that make the image more pleasing to the observer, but not necessarily to produce realistic data from a scientific point of view. Image enhancement techniques (like contrast stretching or de-blurring by a nearest neighbour procedure) provided by "Imaging packages" use no a priori model of the process that created the image [9]. Image enhancement noise can effectively be removed by sacrificing some resolution, but this is not acceptable in many applications. In a Fluorescence Microscope resolution in the z-direction is bad as it is. More advanced image processing techniques must be applied to recover the object. Deconvolution is an example of image restoration method. It is capable of Increasing resolution, especially in the axial direction removing noise increasing contrast [9]. 5.2 Restoration: There are many definitions of restoring a photo. You may want to fix the color, the tone, the contrast, or you may need to repair some damage on an old family memory. All of these can make a photo look new, or at lease restore the appearance that you want. Although some color and tonal corrections are relatively ease to achieve, repairing damage can take long hours of slow, methodical work. Most of the restorations that you might want to do to a photo will be successful if you take the necessary time [9].

5.3 Requirements for Restoration: The successful restoration of blurred image requires accurate estimation of PSF parameters. In our project, we deal with images, which are blurred by the relative motion between the imaging system and the original scene. Thus, given a motion blurred and noisy image, the task is to identify the point spread function parameters and apply the restoration filter to get an approximation to the original scene. Parameter estimation is based on the observation that image characteristics along the direction of motion are different than the characteristics in other directions. The PSF of motion blur is characterized by two parameters namely, blur direction and blur length [9]. 5.4 Degradation: It is used to improve the appearance of an image by application of a restoration process that uses a mathematical model for image degradation. Types of degradation Blurring caused by motion or atmospheric disturbance Geometric distortion caused by imperfect lenses Superimposed interference patterns caused by mechanical systems Noise from electronic sources. 5.4.1 Image Degradation: Image restoration suppressing image degradation using knowledge about its nature. Most image restoration methods are based on convolution applied globally to the whole images. Causes of image degradation defects of optical lenses,

nonlinearity of the electro-optical sensor, graininess of the film material, relative motion between an object and camera wrong focus, atmospheric turbulence in remote sensing or astronomy, The objective of image restoration is to reconstruct the original image from its degraded version. 5.4.2 Degradation of Blur Parameters: The problem of restoration of images blurred by relative motion between the camera and the object scene is important in a large number of applications. The solution proposed here identifies important parameters with which to characterize the point spread function (PSF) of the blur, given only the blurred image itself. This identification method is based on the concept that image characteristics along the direction of motion are different from the characteristics in other directions. Depending on the PSF shape, the homogeneity and the smoothness of the blurred image in the motion direction are greater than in other directions. Furthermore, in this direction correlation exists between the pixels forming the blur of the original unblurred objects. By filtering the blurred image we emphasize the PSF characteristics at the expense of the image characteristics. The method proposed here identifies the direction and the extent of the PSF of the blur and evaluates its shape which depends on the type of motion during the exposure. Correct identification of the PSF parameters permits fast high resolution restoration of the blurred image.

Blur parameters:

When the intensity of the observed point image is spread over several pixels, this is known as the Point Spread Function (PSF).

Length: Blur Length is the number of pixels by which the image is degraded. It is the number of pixel positions by which a pixel is shifted from its original position.




Fig 5.1 Blur Length

Angle: Blur Angle is the angle at which the image is degraded.


Original Fig 5.2 Blur Angle


5.5 Thresholding: The simplest property that pixels in a region can share is intensity. So, a natural way to segment such regions is through thresholding, the separation of light and dark regions. Thresholding creates binary images from grey-level ones by turning all pixels below some threshold to zero and all pixels about that threshold to one [8]. The major problem with thresholding is that We consider only the intensity, not any relationships between the pixels. There is no guarantee that the pixels identied by the thresholding process are contiguous. We can easily include extraneous pixels that arent part of the desired region, and we can just as easily miss isolated pixels within the region. These effects get worse as the noise gets worse, simply because its more likely that pixels intensity doesnt represent the normal intensity in the region. We typically have to play with it, sometimes losing too much of the region and sometimes getting too many extraneous background pixels.


6.1 Introduction: Images and videos are often corrupted by impulse noises during acquisition and transmission. This impulse noise present in an image due to bit errors in transmission or introduced during the signal acquisition stage. There are two types of impulse noise, they are salt and pepper noise and random valued noise. This is based on the noise values. The noise which is easier-to-restore is called salt- and-pepper noise and the noise more difficult random valued is called impulse noise [1]. Salt-and-pepper noise (also called impulse noise, shot noise or spike noise) typically caused by malfunctioning pixel element in camera sensors, faulty memory locations, or timing errors in digitization process. Images and videos are often corrupted by impulse noises during acquisition and transmission. This impulse noise present in an image due to bit errors in transmission or introduced during the signal acquisition stage. There are two types of impulse noise, they are salt and pepper noise and random valued noise. This is based on the noise values. The noise which is easier to restore is called salt and pepper noise and the noise more difficult random valued is called impulse noise [2]. 6.2 Salt and Pepper Noise: Impulse noise in images is present due to bit errors in transmission or introduced during the signal acquisition stage. There are two types of impulse noise, they are salt and pepper noise and random valued noise. Salt and pepper noise can corrupt the images where the corrupted pixel takes either maximum or minimum gray level. Several nonlinear lters have been proposed for restoration of images contaminated by salt and pepper noise [1]. Among these standard median lter has been established as reliable method to remove the salt and pepper noise without damaging the edge details [2].

6.3 Disadvantages of Salt and Pepper Noise: Salt and pepper noise can corrupt the images where the corrupted pixel takes either maximum or minimum gray levels. Several nonlinear filters have been proposed for restoration of images contained by salt and pepper noise.

6.4 Mean Filter: Mean filtering is a simple, intuitive and easy to implement method of smoothing images, i.e. reducing the amount of intensity variation between one pixel and the next. It is often used to reduce noise in images. The idea of mean filtering is simply to replace each pixel value in an image with the mean (`average') value of its neighbors, including itself. This has the effect of eliminating pixel values which are unrepresentative of their surroundings. Mean filtering is usually thought of as a convolution filter. Like other convolutions it is based around a kernel, which represents the shape and size of the neighborhood to be sampled when calculating the mean. Often a 33 square kernel is used, as shown in Figure 6.1, although larger kernels (e.g. 55 squares) can be used for more severe smoothing.

Fig 6.1 33 averaging kernel often used in mean filtering

The two main problems with mean filtering which are A single pixel with a very unrepresentative value can significantly affect the mean value of all the pixels in its neighbourhood. When the filter neighbourhood straddles an edge, the filter will interpolate new values for pixels on the edge and so will blur that edge. This may be a problem if sharp edges are required in the output. Both of these problems are tackled by the median filter, which is often a better filter for reducing noise than the mean filter, but it takes longer to compute. 6.5 Median Filter: The median filter is normally used to reduce noise in an image, somewhat like the mean filter. However, it often does a better job than the mean filter of preserving useful detail in the image. This class of filter belongs to the class of edge preserving smoothing filters which are non-linear filters. This means that for two images A(x) and B(x) .This is shown by the below equation 6.1.

Median [A(x) + B(x)]

Median [A(x)] + Median [B(x)]


These filters smoothes the data while keeping the small and sharp details. The median is just the middle value of all the values of the pixels in the neighbourhood [7]. The median has half the values in the neighbourhood larger and half smaller. The median is a stronger "central indicator" than the average. In particular, the median is hardly affected by a small number of discrepant values among the pixels in the neighbourhood. Consequently, median filtering is very effective at removing various kinds of noise.

Fig 6.2 Illustrates an example of median filtering Like the mean filter, the median filter considers each pixel in the image in turn and looks at its nearby neighbors to decide whether or not it is representative of its surroundings. Instead of simply replacing the pixel value with the mean of neighboring pixel values, it replaces it with the median of those values.

Fig 6.3 calculating the median value of a pixel neighborhood

The median is calculated by first sorting all the pixel values from the surrounding neighborhood into numerical order and then replacing the pixel being considered with the middle pixel value. (If the neighborhood under consideration contains an even number of pixels, the average of the two middle pixel values is used.) Figure 7.3 illustrates an example calculation [7]. 6.5.1 Advantages of Median Filter: By calculating the median value of a neighborhood rather than the mean filter, the median filter has two main advantages over the mean filter

The median is a more robust average than the mean and so a single very unrepresentative pixel in a neighborhood will not affect the median value significantly.

Since the median value must actually be the value of one of the pixels in the neighborhood, the median filter does not create new unrealistic pixel values when the filter straddles an edge. For this reason the median filter is much better at preserving sharp edges than the mean filter.

6.5.2 Disadvantage of the Median Filter: Although median filter is a useful non-linear image smoothing and enhancement technique. It also has some disadvantages. These are The median filter removes both the noise and the fine detail since it can't tell the difference between the two. Anything relatively small in size compared to the size of the neighbourhood will have minimal affect on the value of the median, and will be filtered out. In other words, the median filter can't distinguish fine detail from noise.

a)Original image;

b)Added Impulse Noisy at 10%

Fig 6.4 The original image and the same image of median filter

As Figure 6.4 shown below are the original image and the same image after it has been corrupted by impulse noise at 10%. This means that 10% of its pixels were replaced by full white pixels. Also shown are the median filtering results using 3x3 and 5x5 windows three iterations of 3x3 median filter applied to the noisy image and finally for comparison, the result when applying a 5x5 mean filter to the noisy image.

a)3x3 Median Filtered

b)5x5 Median Filtered

a)3x3 Median Filtered applied 3 times

b)5x5 Average Filter

Fig 6.5 Comparison of the nonlinear Median filter and the linear Mean filter.

6.5.3 Comparison between the median filter and the average filter: Sometimes we are confused by median filter and average filter, thus lets do some comparison between them [7]. The median filter is a non-linear tool, while the average filter is a linear one. In smooth, uniform areas of the image, the median and the average will differ by very little. The median filter removes noise, while the average filter just spreads it around evenly. The performance of median filter is particularly better for removing impulse noise than average filter.

6.6 Removing Of Salt and Pepper Noise Using Different Filters: The restoration of gray scale, and colour images that are highly corrupted by salt and pepper noise. The pixel values of 0s and 255s are present in the selected window. The noise pixel is replaced by mean value of all the elements present in the selected window. Different types of filters are used to remove the salt and pepper noise and to give a better Peak signal to noise Ratio and Image Enhancement Factor. 6.6.1 Standard Median Filter: The standard median filter has been established as reliable method to remove the salt and pepper noise without damaging the edge details. It returns the median value of the pixels in a neighborhood is non linear. It is similar to a uniform blurring filter which returns the mean value of the pixels in a neighborhood of a pixel unlike a mean value filter the median tends to preserve step edges [13].

Fig 6.6 Standard median filter The Major drawback of standard median filter is The filter is effective only at low noise densities. When the noise level is over 50%, the edge details of the original image will not be preserved by the standard median filter.

6.6.2 Adaptive Median Filter: The adaptive median filtering has been applied widely as an advanced method compared with standard median filtering [7]. The Adaptive Median Filter performs spatial processing to determine which pixels in an image have been affected by impulse noise. The Adaptive Median Filter classifies

Fig 6.7 Adaptive Median Filter

The pixels as noise by comparing each pixel in the image to its surrounding neighbour pixels. The size of the neighbourhood is adjustable, as well as the threshold for the comparison. A pixel that is different from a majority of its neighbours, as well as being not structurally aligned with those pixels to which it is similar, is labelled as impulse noise. These noise pixels are then replaced by the median pixel value of the pixels in the neighbourhood that have passed the noise labelling test. The purpose of the adaptive median filter is the Remove impulse noise. Smoothing of other noise. Reduce distortion like excessive thinning or thickening of object boundaries.

Comparison of standard median filter over adaptive median filter

The standard median filter does not perform well when impulse noise is greater than 0.2, while the adaptive median filter can better handle these noises. The adaptive median filter preserves detail and smooth non-impulsive noise, while the standard median filter does not. 6.6.3 Tolerance Based Switched Median Filter: The decision is based on a predefined threshold value. The major drawback of switching median filter is defining a robust decision, it is difficult. The edge may not be recovered satisfactorily especially when the noise level is high. To overcome the above drawback we can use the decision based algorithm [5].

Fig 6.8 Tolerance Based Switched Median Filter

6.6.4 Decision Based Algorithm (DBA): The image is de-noised by using a 3 X 3 window. If the processing pixel value is 0 or 255 it is processed or else it is left unchanged. At high noise density the median value will be 0 or 255 which is noisy [6]. In such case, neighbouring pixel is used for replacement. This repeated replacement of neighbouring pixel produces streaking effect. To overcome this problem decision based unsymmetric trimmed median filter is used.

Fig 6.9 Decision Based Algorithm

6.6.5 Unsymmetric Trimmed Median Filter: The crux behind the above filter is to eliminate the outliers inside the current window. Certain type of non linear filters such as Alpha trimmed mean filter (ATMF), Alpha trimmed midpoint (ATMP) etc., works on the above principle. These filters use a parameter called which decides the number of pixels to be eliminated. It was found that when is increased, the filter fared well. For high noise densities it does not preserve the image information due to the elimination of outlier values [11].

The idea behind a trimmed filter is to reject the noisy pixel from the selected 3 X 3 window. Alpha Trimmed Mean Filtering (ATMF) is a symmetrical filter where the trimming is symmetric at either end. In this procedure, even the uncorrupted pixels are also trimmed. This leads to loss of image details and blurring of the image. In order to overcome this drawback, an Un-symmetric Trimmed Median Filter (UTMF) is proposed. In this UTMF, the selected 3 3 window elements are arranged in either increasing or decreasing order. Then the pixel values 0s and 255s in the image (i.e., the pixel values responsible for the salt and pepper noise) are removed from the image. 6.6.6 Decision Based Un-symmetric Trimmed Median Filter: Digital images are contaminated by impulse noise during image acquisition or transmission due to malfunctioning pixels in camera sensors, faulty memory locations in hardware, or transmission in a noisy channel. Salt and pepper noise is one type of impulse noise which can corrupt the image, where the noisy pixels can take only the maximum and minimum gray values in the dynamic range. The linear filter like mean filter and related filters are not effective in removing impulse noise. Non-linear filtering techniques like Standard Median Filter (SMF), Adaptive Median Filter (AMF) are widely used to remove salt and pepper noise due to its good de-noising power and computational efficiency. SMF is effective only at low noise densities. Several methods have been proposed for removal of impulse noise at higher noise densities. The window size used in these methods is small which results in minimum computational complexity. However, small window size leads to insufficient noise reduction. Switching based median filtering has been proposed as an effective alternative for reducing computational complexity. Recent methods like Decision Based Algorithm (DBA), Modified Decision Based Algorithm (MDBA), are one of

the fastest and efficient algorithms capable of impulse noise removal at noise densities as high as 80%. A major drawback of this algorithm is streaking effect at higher noise densities. To overcome this drawback, Modified Decision Based Unsymmetric Trimmed Median Filter (MDBUTMF) is used to remove salt and pepper noise at very high densities as 80 -90%. In this algorithm, at high noise density, the processing pixel is replaced by the mean value of elements within the window. This will lead to blurring of fine details in the image. To avoid this problem, we have introduced fuzzy thresholding is used to preserve the edges and fine details in this paper. These filters are removing the salt and pepper noise at medium noise variance 50- 60%. 6.6.7 Modified Decision Based Un-symmetric Trimmed Median Filter: At high noise densities, if the selected window contains all 0s or 255s or both then, trimmed median value cannot be obtained. So this algorithm does not give better results at very high noise density that is at 80% to 90%. The proposed Modified Decision Based Un-symmetric Trimmed Median Filter (MDBUTMF) algorithm removes this drawback at high noise density and gives better Peak Signal-to-Noise Ratio (PSNR) and Image Enhancement Factor (IEF) values than the existing algorithm. In the proposed method first the noisy image is read then based on some decision salt and pepper noise detection takes place. At the end of the detection stage the noisy and noise-free pixels get separated. The noise-free pixel is left unchanged and the noisy pixel is given to the Modified Decision Based Unsymmetric Trimmed Median Filter (MDBUTMF). The MDBUTMF produces an image as its throughput that is a partially noise removed one. And it is further processed by Fuzzy Noise Reduction Method (FNRM). Finally the FNRM provides a restored image that is fully

free from noise. The Modified Decision Based Un-symmetric Trimmed Median Filter provides the final output image with higher PSNR value and less Mean square error [16].

Fig 6.10 Modified Decisions Based Un-symmetric Trimmed Median Filter


7.1 Photo shop: Photoshop files have default file extension as PSD, which stands for "Photoshop Document." A PSD file stores an image with support for most imaging options available in Photoshop. These include layers with masks, transparency, text, alpha channels and spot colors, clipping paths, and duotone settings. This is in contrast to many other file formats (e.g. .JPG or .GIF) that restrict content to provide streamlined, predictable functionality. A PSD file has a maximum height and width of 30,000 pixels, and a length limit of 3 Gigabytes. Photoshop files sometimes have the file extension .PSB, which stands for "Photoshop Big" (also known as "large document format"). A PSB file extends the PSD file format, increasing the maximum height and width to 300,000 pixels and the length limit to around 4 Exabyte. The dimension limit was apparently chosen arbitrarily by Adobe, not based on computer arithmetic constraints (it is not close to a power of two, as is 30,000) but for ease of software testing. PSD and PSB formats are documented. Because of Photoshop's popularity, PSD files are widely used and supported to some extent by most competing software. The .PSD file format can be exported to and from Adobe's other apps like Adobe Illustrator, Adobe Premiere Pro, and After Effects, to make professional standard DVDs and provide non-linear editing and special effects services, such as backgrounds, textures, and so on, for television, film, and the web. Photoshop's primary strength is as a pixel-based image editor, unlike vector-based image editors. Photoshop also enables vector graphics editing through its Paths, Pen tools, Shape tools, Shape Layers, Type tools, Import command, and Smart Object functions. These tools and commands are convenient to combine pixel-based and vector-based images in one Photoshop document, because it may not be necessary to use more than

one program. To create very complex vector graphics with numerous shapes and colours, it may be easier to use software that was created primarily for that purpose, such as Adobe Illustrator or CorelDraw. Photoshop's non-destructive Smart Objects can also import complex vector shapes. 7.2 Satellites: In the context of spaceflight, a satellite is an object which has been placed into orbit by human endeavour. Such objects are sometimes called artificial satellites to distinguish them from natural satellites such as the Moon. The world's first artificial satellite, the Sputnik 1, was launched by the Soviet Union in 1957. Since then, thousands of satellites have been launched into orbit around the Earth. Some satellites, notably space stations, have been launched in parts and assembled in orbit. Artificial satellites originate from more than 50 countries and have used the satellite launching capabilities of ten nations. A few hundred satellites are currently operational, whereas thousands of unused satellites and satellite fragments orbit the Earth as space debris. A few space have been placed into orbit around other bodies and become artificial satellites to the

Moon, Mercury, Venus, Mars, Jupiter, Saturn, and the Sun. Satellites are used for a large number of purposes. Common types include military and civilian Earth observation satellites, communications, navigation satellites, weather satellites, and research satellites. Space stations and

human spacecraft in orbit are also satellites. Satellite orbits vary greatly, depending on the purpose of the satellite, and are classified in a number of ways. Well-known (overlapping) classes include low Earth orbit, polar orbit, and geostationary orbit.

Satellites are usually semi-independent computer-controlled systems. Satellite subsystems attend many tasks, such as power generation, thermal control, telemetry, attitude control and orbit control. 7.3 Medical line process: A prevention or preventive measure is a way to avoid an injury, sickness, or disease in the first place, and generally it will not help someone who is already ill (though there are exceptions). For instance, many babies in developed countries are given a polio vaccination soon after they are born, which prevents them from contracting polio. But the vaccination does not work on patients who already have polio. A treatment or cure is applied after a medical problem has already started. A treatment treats a problem, and may lead to its cure, but treatments often ameliorate a problem only for as long as the treatment is continued, especially in chronic diseases. For example, there is no cure for AIDS, but treatments are available to slow down the harm done by HIV and delay the fatality of the disease. Treatments don't always work. For example, chemotherapy is a treatment for some types of cancer. In some cases, chemotherapy may cause a cure, but not in all cases for all cancers. When nothing can be done to stop or improve a medical condition, beyond efforts to make the patient more comfortable, the condition is said to be untreatable. Some untreatable conditions naturally resolve on their own, others do not. Cures are a subset of treatments that reverse illnesses completely or end medical problems permanently. Many diseases that cannot be cured are still treatable.


The performance of the proposed algorithm is tested for various levels of noise corruption and compared with standard filters namely standard median filter (SMF), Adaptive median filter (AMF) and Modified decision based algorithm (MDBA). Each time the test image is corrupted by salt and pepper noise of different density ranging from 10 to 90 with an increment of 10 and it will be applied to various filters. In addition to the visual quality, the performance of the developed algorithm and other standard algorithms are quantitatively measured by the following parameters such as peak signal-to- noise ratio (PSNR), Mean square error (MSE) and Image Enhancement Factor (IEF) [1]. All the filters are implemented in MATLAB 7.5 mon a PC equipped with 2.4 GHz CPU and 1 GB RAM memory for the evaluation of computation time of all algorithms.


Fig: 8.1 project execution first step

Fig 8.2 Select the video sample BROWSE VIDEO:

Fig 8.3 Browsing the video sample


Fig 8.4 Frame separation


Fig 8.5 Frame selection

Fig 8.6 selecting the frame



Fig 8.7 Generating the noise in first level


Fig 8.8.Generating the noise in second level


Fig 8.9 Generating the noise in third level


Fig 8.10 Generating the noise in fourth level


Fig 8.11 Generating the noise in fifth level


Fig 8.12.The output of the standard median filter


Fig 8.13.The output of the adaptive median filter


Fig 8.14.The output of the tolerance based switched median filter


Fig 8.15.The output of the decision based algorithm

THE OUTPUT OF EXISTING METHODS: Existing Methods SMF AMF TSAMF DBA PSNR 8.57982 9.56272 8.87427 8.82034 MSE 58.2655 29.5491 47.542 49.3464

Fig 8.16.The output values of the existing methods


Fig 8.17.The output of the decision based unsymmetric trimmed median filter


Fig 8.18.The output of the decision based unsymmetric trimmed median filter for video Restoration

THE MODIFIED METHOD: Proposed Method TSAMFT PSNR 11.6698 MSE 6.89324

Fig 8.19.The output value of the modified method


The developed algorithms are tested using 512X512, 8bits/pixel image Lena (Gray), Parrot (colour), Barbara (colour). The performance of the proposed algorithm is tested for various levels of noise corruption and compared with standard filters namely standard median filter (SMF), adaptive Median filter (AMF) and decision based Algorithm (DBA). Each time the test image is corrupted by salt and pepper noise of different density ranging from 10 to 90 with an increment of 10 and it will be applied to various filters [2]. In addition to the visual quality, the performance of The developed algorithm and other standard algorithms are quantitatively measured by the following parameters such as peak signal-to-noise ratio (PSNR), Mean square error (MSE) and Image Enhancement Factor (IEF). All the filters are implemented in MATLAB 7.5 on a PC equipped with 2.4 GHz CPU and 1 GB RAM memory for the evaluation of computation time of all algorithms [2].


BIBLOGRAPHY [1] K. S. Srinivasan and D. Ebenezer, A new fast and efcient decision based algorithm for removal of high density impulse noise, IEEE Signal Process. Lett., vol. 14, no. 3, pp. 189192, Mar. 2007.

[2] W.K.Pratt, Digital Image Processing (New York: Wiley 1978). [3] H. Hwang and R. A. Haddad, Adaptive median filters: New algorithms and results, IEEE Trans. Image Process., Vol. 4, No. 4, pp. 499502, Apr. 1995. [4] T. C. Lin and P. T. Yu, Adaptive two-pass median filter based on support vector machine for image restoration, Neural Computation, Vol. 16, 2004, pp. 333-354. [5] T. Chen and R. W. Hong, Application of partition-based median type filters for suppressing noise in images, IEEE Transactions on Image Processing, Vol. 10, 2001, pp. 829-836. [6] E. Abreu, M. Lightstone, S. K. Mitra, and K. Arakawa, A new efficient approach for the removal of impulse noise from highly corrupted images, IEEE Transactions on Image Processing, Vol. 5, 1996, pp. 1012-1025. [7] S. J. Ko and Y. H. Lee, Center weighted median filters and their applications to image enhancement, IEEE Transactions on Circuits Systems, Vol. 38, 1991, pp. 984-993. [8] T. Chen and H. R. Wu, Adaptive impulse detection using center-weighted median filters, IEEE Signal Processing Letters, Vol. 8, 2001, pp. 1-3.

[9] A. Beghdadi and A. Khellaf, A noise-filtering method using a local information measure, IEEE Transactions on Image Processing, Vol. 6, 1997, pp. 879-882. [10] K. C. Lee, H. J. Song and K. H. Sohn, Detection-estimation based approach for impulsive noise removal, Electronics Letters, Vol. 34, 1998, pp. 449-450. [11] J. S. Kim and H. W. Park, Adaptive 3-D median filtering for restoration of an image sequence corrupted by impulse noise, Signal Processing: Image Communication, Vol. 16, 2001, pp. 657-668. [12] T. Sun and Y. Neuvo, Detail-preserving median based filters in image processing, Pattern Recognition Letters, Vol. 5, 1994, pp. 341-347. [13] T. Chen, K. K. Ma and L. H. Chen, Tri-state median filter for image denoising, IEEE Transactions on Image Processing, Vol. 8, 1999, pp. 1834-1838. [14] E. Abreu and S. K. Mitra, A signal-dependent rank ordered mean (SD-ROM) filter: a new approach for removal of impulses from highly corrupted images, in Proceedings of the IEEE International Conference on Acoustics, Speech and SignalProcessing, 1995, pp. 2371-2374. [15] Z. Wang and D. Zhang, Progressive switching median filter for the removal of impulse noise from highly corrupted images, IEEE Transactions on Circuits and Systems- II: Analog and Digital Signal Processing, Vol. 46, 1999, pp. 78-80. [16] J. H. Wang and L. D. Lin, Improved median filter using minmax algorithm for image processing, Electronics Letters, Vol. 33, 1997, pp. 1362-1363.