Image Processing Using Edge Detection Filters

DOI : 10.17577/IJERTCONV2IS04038

Download Full-Text PDF Cite this Publication

Text Only Version

Image Processing Using Edge Detection Filters

Snehal M. Kognule1, Rahul R. Talawadekar 2, Monica S. Jadhav3, Swapnesh S. Surve. 4

1,2,3,4 Department of Computer Engineering, P.V.P.P.C.O.E, Sion, Mumbai

Abstract:-This project is a new approach to Image Processing Using Edge Detection Filters. This project briefly introduces the development of three recent algorithms using wavelet transform for Edge Detection. Then it has worked on the methods & processes required to help us in the completion of our major project which is Image processing using edge detection filters.

The classical edge detectors work fine with high-quality pictures, but often are not good enough for noisy pictures because they cannot distinguish edges of different significance and working only in spatial domain. The proposed wavelet based edge detection algorithms working in spatial as well as frequency domain, therefore the accuracy will be high to detect the edges. All Algorithms are develop in the MATLAB Environment using Wavelet and Image Processing Toolbox.

Keyword- Edge Detection, wavelet transform, classical edge detectors.


    Edges are significant local changes of intensity in an image. Edges typically occur on the boundary between two different regions in an image. Edge Detection is very important in medical science. We are try to detect the edges using various filters like Robert, Prewitt, Sobel, Laplacian Of Gaussian (LOG) operators. Canny operators. Major focus on the wavelet transform algorithm because it works in spatial as well as frequency domain and gives response in less time as compared to other classical operators like Robert, Prewitt, Sobel, Laplacian Of Gaussian (LOG) operators. Canny operators. The following goals will achieve to use wavelet transform for edge detection-(1) Correctly detect edge.(2) Higher accuracy to edge location.(3) Single pixel detection echo.(4) More accurate echo to different scale edge and lower ratio of missed detection.(5) Higher capability of anti-noise. For more accuracy we will implement three recent algorithms with wavelet transform and also provide the classical edge detector for user simplicity. The main application for this project is to find the tumor in the x-ray images i.e. it is useful for medical purpose and the advantage is that it reduces the time of the output.


    Some recent papers have addressed some of the problems by the classical edge detectors. The authors are use multi scale Wavelet transform for detects the edges in frequency

    domain and DWT for image processing and other various algorithms depending on wavelet transform. Lei Zhai and Shouping Dong and Honglian Ma presents the IEEE paper on the recent methods Applications on Image Edge Detection with wavelet transform.

    Wanpeng Cao et al[1] .proposed illumination-independent edge detection and fuzzy enhancement algorithm based on wavelet transform to extract edges out of the non-uniform weak illumination image.

    Jie Hou et al. [1] presented a new edge detection method which combines the merit of canny edge detection algorithm with the wavelet transform.

    Xianmin Ma [1] proposed a revised algorithm based on wavelet transform to suppress the speckles in the gangue images, to enhance edges. Using these three algorithms we are try to implement the software in mat lab.


    The study of wavelet analysis shows that this particular field of mathematics would not have been developing so quickly were it not for the vast applications that required such a mathematical tool. In France, Yves Meyer of the University of Paris-Dauphina found himself talking to astronomers about how these new techniques could be used to study the large-scale structure of the universe. The technique of edge detection has long been studied by researchers in electrical engineering and in mathematics. It was not until the work of Mallet and his colleagues that wavelet was introduced into such studies. Mallet and his colleagues put their emphasis on decomposing and reconstructing digital signals from their edge information. Here, we make improvements based on some of the previous results, and develop a wavelet approach for the problem of edge detection. The concept of wavelet analysis has been developed since the late 1980s. However, its idea can be traced back to the Littlewood- Paley technique and Calderón-Zygmund theory in harmonic analysis. Wavelet analysis is a powerful tool for time-frequency analysis. The transform was designed to represent edges and other singularities along curves much more efficiently than traditional transforms, i.e. using significantly fewer coefficients for a given accuracy of reconstruction. Later and based on a frequency partition technique, the same authors proposed a considerably simpler second-generation wavelet transform.


    Edge detection is the name for a set of mathematical methods which aim at identifying points in a digital image at which the image brightness changes sharply or, more formally, has discontinuities. The points at which image brightness changes sharply are typically organized into a set of curved line segments termed edges. The same problem of finding discontinuities in 1D signals is known as step detection and the problem of finding signal discontinuities over time is known as change detection. Edge detection is a fundamental tool in image processing, machine vision and computer vision, particularly in the areas of feature detection and feature extraction.

    Edge detection is a fundamental tool in image processing and computer vision, particularly in the areas of feature detection and feature. The edges extracted from a two-dimensional image of a three-dimensional scene can be classified as either viewpoint dependent or viewpoint independent. The Edge Detection block outputs a binary image with the edges shown in white. This output is displayed in the Edges window. Edges are boundaries between different textures. Edge also can be defined as discontinuities in image intensity from one pixel to another.. The edges for an image are always the important characteristics that offer an indication for a higher frequency.

    Detection of edges for an image may help for image segmentation, data compression, and also help for well matching, such as image reconstruction and so on. There are many methods to make edge detection. The most common method for edge detection is to calculate the differentiation of an image. The first-order derivatives in an image are computed using the gradient, and the second- order derivatives are obtained using the Laplacian. Another method for edge detection uses Hilbert Transform. And we have proposed a new method called short response Hilbert transform (SRHLT) that combines the differentiation method and the Hilbert transform method. Edge detection is a terminology in image processing that refers to algorithms which aim at identifying edges in an image. It is encountered in the areas of features election and feature extraction in Computer Vision. An edge detector accepts a digital image as input and produces an edge map as output. The edge map of some detectors includes explicit Information about the position and strength of the edges and their orientation[10].


    Types of Edge Detection

    1. Canny Edge Detection

    2. Sobel Edge Detection

    3. Prewitt Edge Detection

    4. Laplacian of Gaussian Edge Detection


    1. Canny Edge Detection

      BW = edge(I,'canny')

      BW = edge(I,'canny ,thresh)

      BW = edge(I,'canny',thresh,sigma) [BW,threshold] =edge(I,'canny',…)

    2. Sobel Edge Detection

      BW = edge(I)

      BW = edge(I,'sobel')

      BW = edge(I,'sobel',thresh)

      BW =edge(I,'sobel',thresh,direction) [BW,thresh] = edge(I,'sobel',…)

    3. Prewitt Edge Detection

      BW = edge(I,'prewitt')

      BW = edge(I,'prewitt',thresh)

      BW =edge(I,'prewitt',thresh,direction) [BW,thresh] = edge(I,'prewitt',…);

    4. Laplacian of Gaussian Edge Detection

      BW = edge (I,'log')

      BW = edge (I,'log',thresh)

      BW=edge(I,'log',thresh,sigm) [BW,threshold]=edge(I,'log',.)


      Edge detection is a common operation issue in image analysis. Edges can be considered as transients in a signal or mathematically defined as local singularities. The Fourier transform is global and not well adapted to local singularities.

      Canny proposed the operators of different widths to obtain better signal to Noise ratios in the detection of patterns that appear at different scales in the image. Marr proposed a multi resolution scheme to detect primitives related to human visual system behavior. Bergholm used Cannys operator and was able to differentiate between shadow contours from perfect ones. All of them used multi resolution scheme but had difficulty in analyzing the information appearing at different scales. After the introduction of wavelet multi resolution analysis that was based on the Gaussian kernel, The results obtained for edge detection and classification of the edges appearing in the Image were better.


      Wavelets represent a departure from Fourier analysis. Actually wavelet analysis is a natural extension to Fourier analysis through its mathematical relationship dated all the Way back to the work of Joseph Fourier in the nineteenth century. Fourier laid the Foundation for the frequency analysis, which proved to been or moistly important. Fourier asserted that any2-periodicfunctionf(x) is the sum of its Fourier series.

      a0 +(ak coskx+bksinkx) k=1

      Thecoefficientsa0,ak and bk are calculated by 1 2 12 12

      a0 = 2 f(x)dx,ak = 0

      f(x)cos(kx)dx,b k =0


      dx 0

      Mathematicians gradually turned their attention from frequency-based analysis to scale-based analysis due to the fact that measuring average fluctuations at different scales might prove less sensitive to noise. The first mention of wavelets appeared in an appendix to the thesis of A. Haar (1909).One property of the Haar wavelet is that it has compact support, which means That It vanishes outside of a finite interval. Unfortunately, Haar wavelets are not Continuously differentiable which somewhat limits their applications. By using a scale-varying basis function called the Haar basis function, PaulLevy, a 1930s physicist, investigated Brownian motion, a type of random signal. He found the Haar basis function superior to the Fourier basis functions for studying small complicated details In the Brownian motion. The computation produced different results if the energy was concentrated around a few points or distributed overall auger interval. Their work provided David Marr With an effective algorithm for numerical image processing using wavelets in the early 1980s. Between 1960 and 1980, the mathematicians Guido Weiss and Ronald R. Coifman studied the simplest elements of a function space, called atoms, with the Goal of finding the atoms for a common function and finding the" assembly rules" that Allow there construction of all the elements of the function space using these atoms. In 1980, Grossman and Morlet, broadly defined wavelets in the context of quantum physics. In 1985,Stephane Mallat gave wavelets an additional jump- start through his work In digital signal processing. He discovered some relationships between quadrature mirror filters, pyramid algorithms, and orthonormal wavelet bases. Inspired by Mallet's work, Yves Meyer constructed the first non-trivial wavelets. [1]


      Medical image enhancement technologies have attracted much attention since advanced medical equipments were put into use in the medical field. Enhanced medical images are desired by a surgeon to assist diagnosis and interpretation because medical image qualities are often deteriorated by noise and other data acquisition devices, illumination conditions, etc. Also targets of medical image enhancement are mainly to solve problems of low contrast and the high level noise of a medical image. Medical image enhancement technologies have attracted many studies, mainly on gray scale transform and frequency domain transform. Studies of frequency domain transform mainly concentrate on the wavelet transform. The wavelet transform is a time frequency analysis tool developed in

      the 1980s, which has been successfully applied in the image processing domain.[9].For to reduce the above problems in medical images and achieve the goals successfully we implementing the following three algorithms in our project. Before detecting the edge of gray image f(x,y) using wavelet transform, a two- dimension smoothing function (x,y) is firstly defined, and its first-order partial derivative in horizontal and vertical directions will act as two basic wavelets, respectively. The convolution of scale transformed wavelets of these two basic wavelets and the image function f(x,y) are defined as the horizontal and vertical component of wavelet transformation. Then the modulus and phase angle of wavelet transform can be obtained and the local modulus maximum of the wavelet transform along the phase angle direction is the edge of the image [10].

      Fig 1:Block diagram of DWT

      Wanpeng Cao et al. [2] proposed an illumination independent edge detection and fuzzy enhancement algorithm based on wavelet transform to extract edges out of the non-uniform weak illumination image. Through synthetic and real images experiments, this edge detection methods performance for the non-uniform weak illumination images is analyzed and compared with other two edge detection methods quantitatively and qualitatively. The experiment result proves the edge detection method works well for the uneven gray and low contrast images.

      Jie Hou et al. [3] presented a new edge detection method which combines the merit of Canny edge detection algorithm with the wavelet transform. The bounds of step- structure and Dirac-structure are distinguished by using modular-angle-separated wavelet transform, or combining algorithm, thus obtaining, the bounds of the step-structure. But the connection with this algorithm's edge and response times of the same edge hasn't solved well. To solve this problem, the edge detection method is proposed. Good effect is achieved in the image edge detection of the bound of step-structure. This algorithm is simple, useful and it's easy to realize.

      Xianmin Ma [4] proposed a revised algorithm based on wavelet transform to suppress the speckles in the gangue images, to enhance edges. Through traditional methods of image processing are efficient as edge detectors, a lot of false edge information will be extracted from a coal gangue image, because the coal gangue images are always contaminated with speckles of coal dust. The experiment results show that the recognition rate of coal gangues is increased by the wavelet transform method. At last the gangues and coal are separated according to their characteristics.


      MATLAB is software package for high performance numerical computation and visualization .There are several optional toolboxes available from the development of MATLAB, but here we use mainly two toolboxes of MATLAB.[6]

      1. Image processing Toolboxes

      2. Wavelet Toolboxes

    1. Image processing Tolboxes

      The Image Processing Toolbox software is a collection of functions that extend the capability of the MATLAB numeric computing environment. The toolbox supports a wide range of image processing operations.

      Image enhancement: It refers to accentuation, or sharpening, of image features such as boundaries, or contrast to make a graphic display more useful for display & analysis. This process does not increase the inherent information content in data. It includes gray level & contrast manipulation, noise reduction, edge christening and sharpening, filtering, interpolation and magnification, pseudo coloring, and so on.

      Image restoration: It is concerned with filtering the observed image to minimize the effect of degradations. Effectiveness of image restoration depends on the extent and accuracy of the knowledge of degradation process as well as on filter design. Image restoration differs from image enhancement in that the latter is concerned with more extraction or accentuation of image features.

      Image compression: It is concerned with minimizing the number of bits required to represent an image. Application of compression are in broadcast TV, remote sensing via satellite, military communication via aircraft, radar, teleconferencing, facsimile transmission, for educational & business documents, medical images that arise in computer tomography, magnetic resonance imaging and digital radiology, motion, pictures, satellite images, weather maps, geological surveys and so on.[7]

    2. Wavelet Toolboxes

    Wavelet is toolbox of MATLAB software .The word " Wavelet is a French word that means small waves used by Jean Morlet and Grossmann in early 1980s.The word wavelet has been used decades in DSP and Exploration Geophysics.[8]

    Everywhere around us are signals that can be analyzed. For example, there are human speech, engine vibrations, medical images, financial data, music, and many other types of signals. Wavelet analysis is a new and promising set of tools and techniques for analyzing these signals.

    Wavelet is basically use for time & frequency domain.


    Matlab gives the service to make gui for image processing snapshot of Our Project gui. It takes the input from the user and show the input as well as processed image. It takes input from the pc or from the camera and we provided various Operations for edge detection.

    Fig 2.GUI snapshot


    Edge detection is a very important problem in image processing. We studied the Problems faced by traditional edge detectors with noisy images. We proved that wavelets are more functional and better edge detectors both theoretically and experimentally. We briefly review the key contributions of this below. In this thesis we have to developed an algorithms with the wavelet edge detection. The comparison of wavelets and traditional edge detection techniques on Images in noisy environment are performed. The main DWT algorithm is implemented. A database of images in environment is subjected to Edge detection using wavelets and the results are tabulated. Although the wavelet transform is a good decorrelator for images, the wavelet Coefficients of natural images exhibit strong dependencies both across scales and between Neighbor coefficients with in a sub band, especially around image edges. This gave rise to several successful joint statistical models in the wavelet domain as well as improved Image compression schemes. The major drawback for wavelets in two-dimensions is their limited ability in capturing directional information.

    International Journal of Engineering ReseaICrcOh N&ETCeTch' n1o4loCgoyn(fIeJrEeRnTce) Proceedings

    ISSN: 2278-0181

    Vol. 3 Issue 2, February – 2014


    1. Correctly detect edge.

    2. Higher accuracy to edge location.

    3. Single pixel detection echo.

    4. More accurate echo to different scale edge and lower ratio of missed detection.

    5. Higher capability of anti-noise.

    6. It will detect the edges which is in spatial as well as frequency domain.

    7. Reduces the time


      1. This wavelet program is only use for square image .


    1. Wavelet transform is used for Edge detection.

    2. For CCTV camera.

    3. For removing noise.

    4. Face recognition, Iris recognition, pattern recognition.

    5. This is use for medical science.


    In recent times, wavelet transform and wavelet package based edge detection become a focus to many researchers. If wavelet generating function is properly selected, the results of extracted image edges will be ideal. Many researchers also estimated threshold using wavelet transform and associated with other algorithms, finally, ideal effectiveness is accomplished. Nowadays, the main difficulties on image project are medical images, infrared images, micro-images, remote sensing images. Those images characterized in higher noise and complicated image edge. Therefore, we use the concept multi-scale in edge extraction, however, wavelet transform appeared unusual superiority. Many researchers improved working efficiency of their algorithms. All of the methods of image edge detection have own advantages and disadvantages, so more research is needed to optimize algorithms in order to make the edge detection more useful not only for higher level image processing, but also for the Accuracy as well. The following studies are required to achieve this goal:(1) Correctly detect edge.(2) Higher accuracy to edge location.(3) Single pixel detection echo.(4) More accurate echo to different scale edge and lower ratio of missed detection.(5) Higher capability of anti-noise.


  1. Lei Zhai, Shouping Dong, Honglian Ma- Recent Methods and Applications on Image Edge Detection

  2. W.P. Cao, R.S. Che, and D. Ye, An illumination independent edge detection and fuzzy enhancement algorithm based on wavelet transform for non-uniform weak illumination images, Pattern Recognition Letters, 2008,29(3),pp.192-199.

  3. Jie Hou, Jin-Hua Ye and Sha-Sha Li, Application of canny combining and wavelet transform in the bound of step-structure edge detection, 2007 International

    Conference on Wavelet Analysis and Pattern Recognition, IEEE, USA,2007, pp.1635-1637.


  4. Xian-Min Ma, A revised edge detection algorithm based on wavelet transform for coal gangue image, Proceedings of 2007 International Conference on Machine Learning and Cybernetics, IEEE,USA,2007,pp.1639-

  5. Using MATLAB, Version 7, The Math Work, Inc., 2000.

  6. MATLAB Function Reference, The Math Work, Inc., 2000

  7. MATLAB Application Program Interfaces, Version 1,The Math Work, Inc,2000.

  8. Using Matlab Graphics Version 6, The Math work, Inc., 2000.


  10. L.Fei, X. De-sheng and Liu Qi-zhong, et al., Image Recognition for Light-Scattering of Air Bubbles in Water, Chinese Optics & Optoelectronic Technology. 2006,4(6),pp. 25- 27.


Leave a Reply