10-04-2017, 08:50 PM
Sobel Edge Detection Algorithm
B.Tech Seminar report
by
Ninto Anto K
Department of Computer Science And Engineering
Government Engineering College, Thrissur
December 2010
B.Tech Seminar report
by
Ninto Anto K
Department of Computer Science And Engineering
Government Engineering College, Thrissur
December 2010
[attachment=7714]
Abstract
Sobel edge detection algorithm is a method to find the edge pixels in an image.it s
a gradient approach in edge detection. Edges are pixels which carry important in-
formation in an image. For example boundaries, shape of models and so on. Thus
edge detection is a process which is of significant importance. Sobel edge detection
is one among the methods which make use of the change in intensity with respect to
neighbouring pixels. That s the property of edge pixels. An edge is a pixel with high
intensity contrast with it s neighbours. As a result we obtain first derivative (gradi-
ent) of pixels and evaluvate it. Sobel method has got a pair of horizontal and vertical
operators. Which on operation obtain gradient of the whole image. we compare the
gradient with a threshold value and judge whether the pixel is part of an edge.
Contents
1 Introduction
1.1 Organization of the Report
2 Edge And Edge Detection
2.1 Edge .
2.2 Edge Detection
2.3 Applications of Edge detection
3 Sobel s Edge Detection Algorithm
3.1 Gradient
3.2 Sobel operator
3.3 Sobel edge detection
3.4 Modified Algorithm .
4 Advantages and Disadvantages of Sobel algorithm
4.1 Advantages .
4.2 Disadvantages
5 Comparison with different Algorithms
References
Chapter 1
Introduction
An edge can be described as a boundary between two regions seperated by distinc-
tion in intensity values of pixels. Changes may be in surface orientation, illumination
etc. An edge detection is a process of identifying those pixels with intensity varia-
tions compared to neighbouring pixels. Edge detection finds important place in object
identification, image compression, object tracking, watermarking etc.
There are mainly two approaches in detecting edge.
Gradient Approach
Laplace Approach
In gradient method we obtain first derivative of pixels property and find pixel with
higest derivative. In Laplace method second derivative of pixel property is measured
and pixels with high first derivative will have second derivative zero.
Sobel edge detection method comes under gradient approach. First derivative is
measured using Sobel Operator. We set threshold value to compare with and decide
whether the pixel is an edge or not.
1.1 Organization of the Report
Rest of the report is organized as follows.
1. Chapter 2 details about edge, edge detection and application of edge detection.
2. Chapter 3 explains about Sobel s Edge Detection Algorithm.
3. Chapter 4 describes advantages and disadvantages of Sobel algorithm.
4. Chapter 5 compares with different Algorithms.
