# Introduction dge detection [1][2] is a fundamental problem of computer vision and image processing. It has been a major concerning issue in image segmentation [3][4][5][6][7] and for the researchers. The purpose of image segmentation is to partition an image into meaningful regions with respect to a particular application where edges in digital images are areas with strong intensity contrasts and a jump in intensity from one pixel to the next can create major variation in the picture quality and image segmentation. For computer vision and image processing systems to interpret an Image, they first must be able to detect the edges of each object in the image [8][9][10][11]. There are several edge detection operators available for image segmentation and object boundary extraction of digital images. Each operator is designed to be sensitive to certain types of edges. Among them Sobel, Roberts, Prewitt, LoG, and canny is major concerning operators. The geometry of the operator determines a characteristic direction in which it is most sensitive to edges. The presence of noise is a problem for image segmentation. Images are very much prone to be affected by a verity of noise like Gaussian noise, Rayleigh noise, Impulse noise and Speckle noise. It has been found from the parent study that in presence of noise the Canny edge detection [12][13][14] operator has yielded the best subjective segmented view of the test image in respective of appearance and object boundary localization then Sobel, Roberts, Prewitt, and LoG. The entropy which is a statistical measure of randomness that can be used to characterize the texture of the input image is studied along with peak signal to noise ratio (PSNR), mean square ratio (MSE) and execution times are also studied in this paper. The objective of the present study is to compare various edge detection operators and analyze their performance and also performances of such techniques is carried out for an image by using MATLAB software. In this literature the section 2 introduces comprehensive theoretical and mathematical background for edge detection and explains different computing approaches to edge detection. Section 3 presents the proposed approach. Section 4 provides the experimental results and discussion and section 5 contains a quick discussion about the conclusion. # II. # Traditional edge detectors a) Sobel The sobel edge detector computes the gradient by using the discrete differences between rows and columns of a 3X3 neighborhood. The sobel operator is based on convolving the image with a small, separable, and integer valued filter. Canny edge detection is a multistage algorithm to detect a wide range of edges in images. This detector finds edges by looking for local maxima of the gradient of f(x, y). The gradient is calculated using the derivative of a Gaussian filter. The method uses two thresholds to detect strong and weak edges and includes the weak edges in the output only if they are connected to strong edges. # III. # Proposed approach The flowchart of the proposed approach is given below. In proposed approach at very beginning a colored image is chosen and inserted into the Mat Lab software for processing. The image is converted into gray scale in the immediate step. A gray scale image is mainly combination of two colors, black and white. It 1. Peak signal to noise ratio (PSNR) is the ratio between the maximum possible power of a signal and the power of corrupting noise that affects the fidelity of its representation. It is the logarithmic function of the peak value of the image and the mean square error. Its value must be high. It have been observed that that the Canny edge detector produces higher accuracy in detection of object edges with higher entropy, PSNR, MSE and execution time compared with Sobel, Roberts, Prewitt, Zero crossing and LOG. On the other hand Roberts edge detector has the minimum entropy with PSNR, MSE and execution time compared with others. The statistical analyses for all the edge detectors are shown in table 1. # V. Conclusion Since edge detection is the initial step in object boundary extraction and object recognition, it is important to know the differences between different edge detection operators. In this paper an attempt is made to review the edge detection techniques which are based on discontinuity intensity levels. The relative performance of various edge detection techniques is carried out with two images by using MATLAB software. It have been observed that that the Canny edge detector produces higher accuracy in detection of object edges with higher entropy, PSNR, MSE and execution time compared with Sobel, Roberts, Prewitt, Zero crossing and LOG. ![b) Prewitt Prewitt operator edge detection masks are the one of the oldest and best understood methods of detecting edges in images The Prewitt edge detector uses the following mask to approximate digitally the first derivatives G x and G y . c) Roberts In Robert edge detection, the vertical and horizontal edges bring out individually and then put together for resulting edge detection. The Roberts edge Volume XII Issue XIII Version I detector uses the following masks to approximate digitally the first derivatives as differences between adjacent pixels. d) Laplacian of Gaussian (LOG) This detector finds edges by looking for zero crossings after filtering f(x, y) with a Laplacian of Gaussian filter. In this method, the Gaussian filtering is combined with Laplacian to break down the image where the intensity varies to detect the edges effectively. It finds the correct place of edges and testing wider area around the pixel. e) Canny Edge Detector](image-2.png "") ![information where, black have the low or weakest intensity and white have the high or strongest intensity. In final step different edge detection operators are applied to detect the object boundaries and edges.IV.Experimental resultThis section presents the relative performance of various edge detectors. Five edge detection operators have been chosen to carry out for edge detection and image segmentation. There are namely, Sobel, Prewitt, Roberts, LoG and Canny. The original image is shown in figure1(a) and figure 2(a) and the segmented images have been shown in figure 1(b) to figure 1(f) and from figure 2(b) to figure 2(f) respectively. The entropy, PSNR, MSE and execution times of the segmented images have been calculated and are shown in Table](image-3.png "") 12![Fig. 1 : (a) Trisha (original image), (b) Sobel, (c) Prewitt, (d) Roberts, (e) Laplacian of Gaussian, (f) canny](image-4.png "Fig. 1 :Fig. 2 :") IIMAGEENTROPY PSNRMSEEXECUTIONTIMETrisha1.282011.4067 4.7034e+0031.052911withseconds.SobelTrisha1.279211.3928 4.7185e+0030.878266withseconds.PrewittTrisha1.230617.1396 1.2564e+0030.831094withseconds.RobertsTrisha1.435411.2313 4.8973e+0030.978503with LoGseconds.Trisha1.570110.9043 5.2803e+0031.014961withseconds.CannyDiya with1.27229.93656.5983e+0030.851769Sobelseconds.Diya with1.27079.93746.5969e+0030.855519Prewittseconds.Diya with1.24939.92126.6215e+0030.818108Robertsseconds.Diya with1.43189.95966.5633e+0030.856581LoGseconds.Diya with1.54779.69826.9705e+0031.040114Cannyseconds. © 2012 Global Journals Inc. (US) Global Journal of Computer Science and Technology © 2012 Global Journals Inc. (US) * A Novel Edge Detection Based Segmentation Algorithm for Polarimetric Sar Images JieYanga RanYanga ShigaoLib SShoujing Yina QianqingQina The International Archives of the Photogrammetry, Remote sensing and Spatial Information Beijing 2008 XXXVII, Part B7 * Techniques for edge detection EArgyle Proc. IEEE IEEE 1971 59 * Image Segmentation by Histogram Thresholding Using Fuzzy Sets JOrlando Tobias & RuiSeara IEEE Transactions on Image Processing 11 12 2002 * A Study of Image Segmentation and Edge Detection Techniques PunamThakare International Journal on Computer Science and Engineering 3 2 2011 * CRafael RichardEGonzalez LWoods & Steven Eddins Digital Image Processing Using MATLAB Singapore Pearson Education Ptd. Ltd 2004 * Medical Images Edge detection Based on mathematical Morphology GuiWeiZhao Yu Quian ChenZhenHua TangCheng LiLingJing Tian Yun Proceedings of the 2005 IEEE the 2005 IEEE * Robust Estimation for Range Image Segmentation and Reconstruction XYu TDBui IEEE trans. Pattern Analysis and Machine Intelligence 16 5 1994 * A study of Edge Detection Techniques for Segmention Computing Approaches SLakshmi .VDr Sankaranarayanan IJCA special issue on " Computer Aided Soft Computing Techniques for imaging and Biomedical Applications 20 * On the Edge Location Error for Local Maximum and Zero-Crossing Edge Detectors JKoplowitz IEEE Trans. Pattern Analysis and Machine Intelligence 16 12 Dec.1994 * A Study of edge detection techniques for segmentation computing approaches S & VLakshmi Sankaranarayanan Computer Aided Soft Computing Techniques for Imaging and Biomedical Applications 2010 * Theory of edge detection D & E.Marr Hildreth Proc. Royal Society of Royal Society ofLondon, B 1980 207 * TAMohmoud * Edge -Detected Guided Morphological Filter for Image sharpening S Hindawi Publishing orporation EURASIP Journal on image and video Processing 2008 * A computational approach to edge detection JCanny IEEE Transaction on Pattern Analysis and Machine Intelligence 8 1986 * RCGonzalez REWoods Digital Image Processing Prentice Hall 2002 2nd ed.