IMAGE ENHANCEMENT USING FUZZY TECHNIQUE: SURVEY AND OVERVIEW

Similar documents
Image Enhancement Using Fuzzy Morphology

A Hybrid Approach using Fuzzy Logic and Neural Network for Enhancement of Low Contrast Images

A Modified SVD-DCT Method for Enhancement of Low Contrast Satellite Images

SCIENCE & TECHNOLOGY

Image Recognition using Bidirectional Associative Memory and Fuzzy Image Enhancement

Image Enhancement Using Fuzzy Logic

Low Contrast Image Enhancement Using Adaptive Filter and DWT: A Literature Review

Image Contrast Enhancement in Wavelet Domain

Fuzzy Inference System based Edge Detection in Images

Adaptive Wavelet Image Denoising Based on the Entropy of Homogenus Regions

IMAGE COMPRESSION USING HYBRID QUANTIZATION METHOD IN JPEG

Comparison between Various Edge Detection Methods on Satellite Image

IMAGE DE-NOISING IN WAVELET DOMAIN

Feature Based Watermarking Algorithm by Adopting Arnold Transform

Analysis of Image and Video Using Color, Texture and Shape Features for Object Identification

OCR For Handwritten Marathi Script

An Optimal Gamma Correction Based Image Contrast Enhancement Using DWT-SVD

Color image enhancement by fuzzy intensification

PRINCIPAL COMPONENT ANALYSIS IMAGE DENOISING USING LOCAL PIXEL GROUPING

Design of Fuzzy Inference System for Contrast Enhancement of Color Images

Denoising and Edge Detection Using Sobelmethod

Image Processing Lecture 10

ADDITIVE NOISE REMOVAL FOR COLOR IMAGES USING FUZZY FILTERS

Comparative Study of Linear and Non-linear Contrast Enhancement Techniques

An ICA based Approach for Complex Color Scene Text Binarization

International ejournals

Image Enhancement: To improve the quality of images

Skin Infection Recognition using Curvelet

Texture Image Segmentation using FCM

Color image enhancement by fuzzy intensification

SCALED WAVELET TRANSFORM VIDEO WATERMARKING METHOD USING HYBRID TECHNIQUE: SWT-SVD-DCT

International Journal for Research in Applied Science & Engineering Technology (IJRASET) Denoising Of Speech Signals Using Wavelets

Distinguishing the Noise and image structures for detecting the correction term and filtering the noise by using fuzzy rules

A New Approach to Compressed Image Steganography Using Wavelet Transform

HYBRID TRANSFORMATION TECHNIQUE FOR IMAGE COMPRESSION

Sparse Component Analysis (SCA) in Random-valued and Salt and Pepper Noise Removal

Volume 2, Issue 9, September 2014 ISSN

INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET)

Hybrid filters for medical image reconstruction

Similarity Measures of Pentagonal Fuzzy Numbers

ADVANCED IMAGE PROCESSING METHODS FOR ULTRASONIC NDE RESEARCH C. H. Chen, University of Massachusetts Dartmouth, N.

A Novice Approach To A Methodology Using Image Fusion Algorithms For Edge Detection Of Multifocus Images

CoE4TN3 Medical Image Processing

Image Enhancement in Spatial Domain. By Dr. Rajeev Srivastava

An Introduc+on to Mathema+cal Image Processing IAS, Park City Mathema2cs Ins2tute, Utah Undergraduate Summer School 2010

IMAGE COMPRESSION USING HYBRID TRANSFORM TECHNIQUE

IMAGE DIGITIZATION BY WAVELET COEFFICIENT WITH HISTOGRAM SHAPING AND SPECIFICATION

Digital Image Steganography Techniques: Case Study. Karnataka, India.

Vivekananda. Collegee of Engineering & Technology. Question and Answers on 10CS762 /10IS762 UNIT- 5 : IMAGE ENHANCEMENT.

Image Enhancement Techniques for Fingerprint Identification

Structural Analysis of Aerial Photographs (HB47 Computer Vision: Assignment)

Structural Similarity Optimized Wiener Filter: A Way to Fight Image Noise

Hybrid Algorithm for Edge Detection using Fuzzy Inference System

Generalized Fuzzy Clustering Model with Fuzzy C-Means

A Color Image Enhancement based on Discrete Wavelet Transform

Novel Intuitionistic Fuzzy C-Means Clustering for Linearly and Nonlinearly Separable Data

INTENSITY TRANSFORMATION AND SPATIAL FILTERING

IMAGE FUSION PARAMETER ESTIMATION AND COMPARISON BETWEEN SVD AND DWT TECHNIQUE

Introduction to Fuzzy Logic and Fuzzy Systems Adel Nadjaran Toosi

International Journal of Computer Engineering and Applications, Volume XII, Issue I, Jan. 18, ISSN

CHAPTER 4 FREQUENCY STABILIZATION USING FUZZY LOGIC CONTROLLER

Texture classification using fuzzy uncertainty texture spectrum

Image Denoising Using wavelet Transformation and Principal Component Analysis Using Local Pixel Grouping

Noise Suppression using Local Parameterized Adaptive Iterative Model in Areas of Interest

Implementation of Magnified Edge Detection using Fuzzy-Canny Logic

Fuzzy Mathematical Approach for the Extraction of Impulse Noise from Muzzle Images

An Improved Performance of Watermarking In DWT Domain Using SVD

ISSN (ONLINE): , VOLUME-3, ISSUE-1,

COMPARATIVE STUDY OF IMAGE FUSION TECHNIQUES IN SPATIAL AND TRANSFORM DOMAIN

A Image Comparative Study using DCT, Fast Fourier, Wavelet Transforms and Huffman Algorithm

Comparison of Wavelet Based Watermarking Techniques for Various Attacks

Fuzzified Denoising Technique for Directional Total Variation Minimization on Color Image

Fuzzy Transform for Low-contrast Image Enhancement

Modified Directional Weighted Median Filter

SPEECH WATERMARKING USING DISCRETE WAVELET TRANSFORM, DISCRETE COSINE TRANSFORM AND SINGULAR VALUE DECOMPOSITION

Robust Image Watermarking based on DCT-DWT- SVD Method

International Journal of Engineering Research-Online A Peer Reviewed International Journal Articles available online

Robust color segmentation algorithms in illumination variation conditions

Genetic Tuning for Improving Wang and Mendel s Fuzzy Database

Irregular Interval Valued Fuzzy Graphs

Lecture 4 Image Enhancement in Spatial Domain

American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) ISSN (Print) , ISSN (Online)

FACE DETECTION USING CURVELET TRANSFORM AND PCA

Novel Iterative Back Projection Approach

Spatial, Transform and Fractional Domain Digital Image Watermarking Techniques

Cursive Handwriting Recognition System Using Feature Extraction and Artificial Neural Network

A Novel method for image enhancement by Channel division method using Discrete Shearlet Transform and Genetic Algorithm

Statistical Image Compression using Fast Fourier Coefficients

A Novel NSCT Based Medical Image Fusion Technique

Robust Image Watermarking based on Discrete Wavelet Transform, Discrete Cosine Transform & Singular Value Decomposition

Image Enhancement 3-1

SURVEY ON IMAGE PROCESSING IN THE FIELD OF DE-NOISING TECHNIQUES AND EDGE DETECTION TECHNIQUES ON RADIOGRAPHIC IMAGES

Enhancing the Image Compression Rate Using Steganography

Comparative Analysis in Medical Imaging

ISSN: Page 320

Classification with Diffuse or Incomplete Information

AN EFFICIENT VIDEO WATERMARKING USING COLOR HISTOGRAM ANALYSIS AND BITPLANE IMAGE ARRAYS

Random spatial sampling and majority voting based image thresholding

BIG DATA-DRIVEN FAST REDUCING THE VISUAL BLOCK ARTIFACTS OF DCT COMPRESSED IMAGES FOR URBAN SURVEILLANCE SYSTEMS

Application of partial differential equations in image processing. Xiaoke Cui 1, a *

Transcription:

IMAGE ENHANCEMENT USING FUZZY TECHNIQUE: SURVEY AND OVERVIEW Pushpa Devi Patel 1, Prof. Vijay Kumar Trivedi 2, Dr. Sadhna Mishra 3 1 M.Tech Scholar, LNCT Bhopal, (India) 2 Asst. Prof. LNCT Bhopal, (India) 3 Professor, LNCT Bhopal, (India) ABSTRACT Image enhancement is a process of improving the quality of image by improving its feature. Contrast enhancement is one of the challenging and interesting area of image processing. Image enhancement plays a vital role in medical images, aerial image, microscopic images, even real life photographs, all these suffers from poor contrast and noise. Uncertainty present in the images can be handled and managed effectively by using fuzzy set theory. Different applications require different types of algorithm for enhancing the quality of an image. The performance parameters of contrast enhancement algorithm are subjective as well as objective. This paper focuses on different fuzzy image enhancement algorithm, which maps elements from image pixel plane to fuzzy plane and to transformed plane by using fuzzy technique and gives better approach for future research. But the problem of existing contrast enhancement techniques are over enhancement and under enhancement. By using nonlinear membership function of fuzzy set theory drawback of over and under enhancement of images can be rectified. Keywords: Image Enhancement, Fuzzy Sets, Fuzzy Image Processing, Fuzzy Operator. I. INTRODUCTION Image processing is defined as the process of converting an image into digital image and performing certain operations on it. There are various operations that are performed, so that enhanced image can be obtained from the original image or we can extract some useful and meaningful information from the original image [1]. The output of the image processing can be an image or the characteristics that are associated with input image. Usually image processing considers images such as 2-D signals, where we apply certain operation on them. With the invention of new technology we can manipulate images with certain systems. Manipulation of the images can be categorized into three types namely: 1. Image Processing 2. Image analysis 3. Image understanding Image Processing[1] can be defined as the systems where the input is in the form of images where processing methods have been applied that produces output in the form of images. Image processing involves operations such as to denoising, color contrast enhancement and image edge detection etc. Image Analysis[1] is defined as the system that accepts image as input where certain methods have been applied that provides the measurements of the image as an output. These measurements includes parameters such as the size of the pixel, the size of the image, pixel length etc. Image understanding [1] provides a system where input image is processed with certain methods and the outputs determine the high level description of the image. This high level description helps in 154 P a g e

more understanding of the image and also helps in extracting the more useful information from the given input image. Many contrast enhancement algorithms are existing, but the development of new algorithm which would produce better images than the existing one is a challenging problem. Several algorithms have been proposed to overcome the uncertainties encountered during transmission and acquisition of the images. Sometimes these uncertainties or vagueness are caused by low contrast in the images. So it is necessary to represent and resolve uncertainty effectively to improve the contrast. Because of the ability to handle and manage the imprecision encountered with images effectively, applying fuzzy set theory becomes a strong in road image processing areas like contrast enhancement. Many research works are still going on in this area to make improvements in the existing techniques [2, 3]. Use of intensification operator makes a boom in the field of enhancement of images [3]. Fuzzy sets introduces the way for modelling imprecision in the biomedical as weil as natural images. A fuzzy control filter to sharpen image edges, without noise magnification uses exponential membership function. Different algorithms uses different membership functions like triangular, Gaussian, exponential, triangular, S- function, or trapezoidal to map pixel image to fuzzy image. An image contrast enhancement algorithm was developed using some trigonometric membership function. One of the main shortcomings of image enhancement is under enhancement. That is enhanced image becomes darker than the original image. Rest of the paper is organized as follows. Section 2 describes briefly about Fuzzy set theory and Fuzzy operators. Description about the performance metrics are given in Section 3. Description about fuzzy image enhancement technique is presented in section 4. Section 5 finally concludes this paper. II. FUZZY SET THEORY Fuzzy set theory is a mathematical tool for handling imprecision or vagueness [4]. A particular problem that is to be solved is represented in the form of human language can be solved easily with the help of fuzzy sets. So the extension of classical set theory-'fuzzy set' becomes robust and flexible. In a fuzzy set values are between 0 to 1 rather than 0 or 1 as in the case of classical set. This varying degree of values shows the degree of possessiveness of elements to its fuzzy set or simply it shows the fuzziness and is known as membership function. Depending on the values, these functions are classified as Gaussian, triangular, exponential, polynomial etc. Selection of particular function is user defined. Let X= {x 1, x 2,... x n }be the universal set, a fuzzy set A defined as ordered pair where the membership function µ A (x i ) having positive values in the interval (0, 1) denotes the degree to which an event x i may be a member of A. 2.1 Fuzzy Image Processing (FIP) Ability to represent the concept of partial truthness, makes fuzzy suitable for image processing. Even if solution for solving a particular problem is not clear, fuzzy set can easily manage if that solution can be represented in the form of linguistic terms. Main steps in fuzzy image processing [4] are: 1. Fuzzification 2. Inference Engine 3. Defuzzification Fuzzification is nothing but assigning required membership function to map images from pixel plane to fuzzy plane. For an 8x8 image of pixel value 0-255 becomes 0 to 1 to represents the fuzziness of pixels. In the fuzzy 155 P a g e

plane, with the help of enhancement / transform operator, images can be modified to get better ones. This plane is sometimes called inference plane. In the defuzzification plane, images from fuzzy plane are taken back to pixel plane with inverse transform. Fig1 shows the block diagram of FIP[4]. Fig1: Fuzzy Image Processing 2.2 Fuzzy Image Plane For any image processing application images from pixel domain are taken to fuzzy domain. Let I be an image of size M N, and I(i,j) represents the intensity element or simply pixels which has to be mapped to fuzzy characteristic plane. Image I can be expressed as follows: Here [µ I(i,j) /I(i,j)] represents each pixel and µ I(i,j) is the degree of intensity level of the image with values between zero and one. So in the fuzzy image plane, pixels from 0-255 are mapped to 0-1. Different transformation functions are applied in this plane to enhance the image. 2.3 Fuzzy Operator Following three operators are used for reducing the fuzziness of a fuzzy set [5]: 1. Dilation(DIL): It increases the degree of membership of all members by spreading out the curve. 2. Concentration(CON): It decreases the degree of membership of all members. 3. Contrast Intensification(INT): It reduces the fuzziness of set A by increasing the µa(x) which are above the 0.5 and decreasing those are below it. III. PERFORMANCE METRICS 156 P a g e

Several parameters [1] are to be considered to evaluate the performance. Performance evaluation of enhanced image is subjective [6] as well as objective. To evaluate the enhanced image, following parameters are used in this paper. 3.1 Measure of Luminance Index (MLI) In this case Iuminance index is used as a measure of intensity. It is a type of similarity based approach. This metric is defined as the ratio of mean (MI) of enhanced image I e to the mean of original image I o, For a good quality image MLI must be a high value. It is defined as follows. 3.2 Contrast Difference (CD) Contrast difference is the difference between maximum (I max ) and minimum (I min ) pixel intensity of an image. High value of contrast difference stands for enhancing brighter (high gray level) pixel to brightest and darker (pixel with low value) to darkest pixel. It is defined as: 3.3 Mean Squared Difference (MSD) and Peak Enhanced to Original Image Ratio (PEOIR) Two commonly used objective measures to check the quality of image are Mean Squared Difference (MSD) and Peak Enhanced to Original Image Ratio (PEOIR) defined as: Where, Imax is the maximum intensity of enhanced image I e is the enhanced image and I o is original image intensity. In order to get a high quality contrast enhanced image, the intensity difference between both enhanced and original image must be very high. Hence the quality of processed image will be higher for high values of MSD, which means that there is a large difference between the intensities of enhanced and original image. Therefore PEOIR will be small for high values of MSD. IV. FUZZY IMAGE ENHANCEMENT TECHNIQUES The objective of image processing is to improve the visual appearance of the image, or to provide a better transform representation for future automated image processing, such as analysis, detection, segmentation and recognition. Moreover, it helps in analyzing background information that is essential to understand object behaviour without requiring Due to low contrast, we cannot clearly extract objects from the dark background. Image enhancement methods are categorised into two broad categories: 1. Spatial based domain image enhancement 2. Frequency based domain image enhancement. Spatial based domain image enhancement operates directly on pixels. The main advantage of spatial based domain technique is that they are conceptually simple to understand and the complexity of these techniques is 157 P a g e

low which favours real time implementations. Frequency based domain image enhancement is a term used to describe the analysis of mathematical functions or signals with respect to frequency and operate directly on the transform coefficients of the image, such as Fourier transform, discrete wavelet transform (DWT), and discrete cosine transform (DCT). The basic idea in using this technique is to enhance the image by manipulating the transform coefficients. The basic limitations including are it cannot simultaneously enhance all parts of image very well and it is also difficult to automate the image enhancement procedure. 4.1 Image Enhancement Using Fuzzy Sets In 1980, Pal S. K. et. al. [2] proposed a method of fuzzy technique for extracting the fuzzy properties corresponding to pixels in the spatial domain of the image and successively applied the fuzzy contrast intensification operator for modification of pixels intensity in fuzzy property plane of the image. The performance of this method is evaluated with different indexes of fuzziness for English script input image. 4.2 Image Enhancement Using Smoothing with Fuzzy Sets In 1981, Pal S. K et. al. [3] proposed A model for grey-tone image enhancement using the concept of fuzzy sets is suggested. It involves primary enhancement, smoothing, and then final enhancement. The algorithm for both the primary and final enhancements includes the extraction of fuzzy properties corresponding to pixels and then successive applications of the fuzzy operator "contrast intensifier" on the property plane. The reduction of the "index of fuzziness" and "entropy" for different enhanced outputs is demonstrated for an English script input. 4.3 A Novel Fuzzy Logic Approach to Contrast Enhancement In 1999, H. D. Cheng et. al. [7] proposed a novel adaptive direct fuzzy contrast enhancement method based on the fuzzy entropy principle and fuzzy set theory. They used S-function as membership function and edge gradient operator to find the edge value of an image. 4.4 A New Fuzzy Logic Filter for Image Enhancement In 2000, Forbiz H. R. et. al.[8] proposed a new fuzzy-logic-control based filter with the ability to remove impulsive noise and smooth Gaussian noise, while, simultaneously, preserving edges and image details efficiently. 4.5 Image Enhancement Based On Fuzzy Logic In 2009, Harish kundra et. al. [9] proposed a filter which remove the noise and improve the contrast of the image. To achieve this goal fuzzy-logic-control based approach is used. The filter is tested on the colored images. 4.6 Fuzzy Inference System Based Contrast Enhancement In 2011, Balasubramaniam et. al. [10] proposed a fuzzy inference system based contrast enhancement of gray level images. They have used triangular fuzzy membership function for converting pixel domain to fuzzy property domain. They proposed a new method of generating the fuzzy if-then rules specific to a given image based on the local information available to be used by a fuzzy inference system. To this end, only generate a partial histogram and not a complete histogram thus saving on computational costs 4.7 Image Contrast Enhancement Using Fuzzy Technique 158 P a g e

In 2013, Reshmalashmi C et. al. [11] proposed a method that deals with a new contrast enhancement algorithm, which maps elements from pixel plane to membership plane and to enhancement plane. Shortcomings of existing contrast enhancement techniques are rectified with the help of a mathematical tool called 'Fuzzy set'. This paper proposes an alternative membership function to map elements defined on the cross over point of fuzzy set. They have achieved PEOIR about 11.37. 4.8 An Approach for Image Enhancement Using Fuzzy Inference System for Noisy Image In 2013, Jaspreet Singh Rajal [12] proposed a fuzzy gray scale enhancement technique for low contrast image corrupted by Gaussian noise. The degradation of the low contrast image is mainly caused by the inadequate lighting during image capturing and thus eventually resulted in non uniform illumination in the image. The proposed method has better performance than available methods in the enhancement of noisy images and has been validated by the performance measures like MSD, PEOIR etc. They have achieved 20.11 PEOIR value. V. CONCLUSION There are various classical contrast enhancement technique in spatial domain such as point processing operation, histogram processing operations, spatial filtering operations etc. All these classical enhancement technique have the drawback of under enhancement and over enhancement of an image. By using nonlinear membership function of fuzzy set theory drawback of over and under enhancement of images can be rectified. Other fuzzy approaches used for contrast enhancement are minimization of fuzziness, equalization using fuzzy expected values, fuzzy hyperbolization, fuzzy rule based approach etc. This paper focused on different fuzzy image enhancement algorithm, which maps elements from image pixel plane to fuzzy plane and to transformed plane by using fuzzy technique and gives better approach for future research. REFERENCES [1] Gonzalez R. C. and Woods R. E., Digital Image Processing, 3'd ed. Upper Saddle River, NJ Prentice- Hall, 2006. [2] Pal S K, King R A. Image enhancement using fuzzy sets IEEE Electronics Letters, 1980, Vol. 16 No 10 pp. 376-378. [3] Pal S K, King R A. Image enhancement using smoothing with fuzzy sets IEEE Trans. Systems, Man & Cybernetics, 1981, Vol. 11 No 7 pp. 494-501. [4] Klir. G. J and Yuan B. Fuzzy sets and Fuzzy logic. Upper Saddle River,NJ: Prentice - Hall, 1955. [5] S. K. Pal and D. Datta Majumdar Effect of fuzzificaon on the plosive cognition system Electronics and communication science unit, Indian statistical Institute, Calcutta, 1977, pp.873-886 [6] Tizhoosh H.R. and Michaelis B. Subjective, and Fuzzy technique: A new approach to image enhancement IEEE 1999, pp. 522-526. [7] H. D Cheng, H Xu novel fuzzy logic approach to contrast enhancement pattern recognition society Elsevier 1999, pp. 809-819. [8] Farbiz F, Menhaj M B, Motamedi S A, and Hagan M T. A new fuzzy logic filter for image enhancement IEEE Trans. Systems, Man & Cybernetics, 2000, Vol. 30 No. 1 pp. 110-119. 159 P a g e

[9] Mr. Harish Kundra, Er. Aashima, Er. Monika Verma Image Enhancement Based On Fuzzy Logic IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.10, October 2009, pp. 141-145. [10] Balasubramaniam Jayaram, Kakarla V.V.D.L. Narayana,V. Vetrivel Fuzzy Inference System Based Contrast Enhancement Aix-les-Bains, France Published by Atlantis Press 2011 pp. 311-318. [11] Reshmalakshmi C, Sasikumar M. Image Contrast Enhancement using Fuzzy Technique IEEE International Conference on Circuits, Power and Computing Technologies [ICCPCT] ISSN 978-1-4673-4922, 2013, Vol. 2 no pp. 861-865. [12] Jaspreet Singh Rajal An Approach for Image Enhancement Using Fuzzy Inference System for Noisy Image Journal of Engineering, Computers & Applied Sciences (JEC&AS) ISSN No: 2319 5606, 2013, Volume2,No.5,pp.4-11. 160 P a g e