Edge Detection

Implementation of  All Edge Detection techniques. 

Theory

An edge represents the dividing line between two areas with noticeably different gray-level characteristics. It can be described as a series of connected pixels outlining the separation between two distinct regions. The assumption is that these regions are consistent enough in their gray levels, allowing boundaries to be detected by sharp intensity changes. Edge detection is a method used to segment an image by identifying regions of discontinuity. 

Edge Detection Operators are of two types: 

• Gradient – based operator which computes first-order derivations in a digital image like, Sobel operator, Prewitt operator, Robert operator

• Gaussian – based operator which computes second-order derivations in a digital image like, Canny edge detector, Laplacian of Gaussian.

Description

The most commonly used edge detection techniques, along with their descriptions:

1. Sobel Edge Detection

It is a discrete differentiation operator. It computes the gradient approximation of image intensity function for image edge detection. At the pixels of an image, the Sobel operator produces either the normal to a vector or the corresponding gradient vector. 

2. Prewitt Edge Detection

The Prewitt operator is similar to the Sobel operator but uses different convolution kernels to detect edges. It also computes the gradient in horizontal and vertical directions but uses simpler filters. The Prewitt operator is computationally less expensive than Sobel

3. Roberts Cross Edge Detection

The Roberts cross operator is one of the earliest edge detection methods. It calculates the difference between neighboring pixels along the diagonals. The Roberts operator is effective for detecting edges with sharp transitions and is applied using two small 2x2 kernels.

4. Laplacian of Gaussian (LoG) Edge Detection

 It is a gaussian-based operator which uses the Laplacian to take the second derivative of an image. This really works well when the transition of the grey level seems to be abrupt. It works on the zero-crossing method i.e when the second-order derivative crosses zero, then that particular location corresponds to a maximum level. It is called an edge location.