26-05-2012, 05:08 PM
A Cellular Neural Network- Based Model for Edge Detection
A Cellular Neural Network- Based Model for Edge Detection.pdf (Size: 240.35 KB / Downloads: 113)
Abstract.
This study employed the use of Cellular Neural Networks (CNN) for Edge Detection in images
due to its high operational speed. The process of edge detection is unavoidable in many image processing
tasks such as obstacle detection and satellite picture processing .The conventional edge detector models such
as Sobel Operator, Robert Cross, among which Canny is the best, have high computational time. The CNN
Model is a class of Differential Equation that has been known to have many application areas and high
operational speed. The work investigated four parameters: resolutions, processing time, false alarm rate, and
usability for performance evaluation. The CNN Model was modified by using hyperbolic tangent (tanh x)
and Von Neumann Neighborhood. The modified CNN Model and enhanced Canny Model were implemented
using MATLAB 7.0 running on Pentium III and 128 MB RAM Personal Computer. A series of images
served as input for both Canny and modified CNN Model. With several images tested, the overall results
indicated that the two models have similar resolutions with average computational time of 1.1078 seconds
and 2.293 seconds for CNN-based and Enhanced Canny Model respectively. The hyperbolic entry a22 of the
cloning template A made our work fully controllable since max(tanh x) = +1 and min(tanh x) = -1. A
consideration of the set of digital images showed that edge maps which result from Canny Model have
adjacent boundaries that tend to merge.
Introduction
In image processing, one of the most effective procedures used in sharpening the images is to improve
the contrast. The contrast is improved by increasing the difference across discontinuities of the image
components in order to improve the differences. Edge detection algorithms are designed to detect and
highlight these continuities. It was first developed for processing satellites pictures and later become well
known and widely used in digital image processing [5]. Detecting the edges of an image reduces the
amount of data and filters out useless information, while preserving the important structural properties in an
image. The two methods of edge detection are gradient and Laplacian [4]. The gradient method detects
the edges by looking for the maximum and minimum in the first derivative of the image while the Laplacian
method searches for zero crossings in the second derivative of the image to find edges. The common
examples of edge detection algorithms are: Sobel, Canny, Laplace, Robert Cross, Prewitt and Sussan Edge
Detectors. Roberts Cross, Prewitt and Sobel operators are gradient based edge detectors and they all have
kernel operators [2] that calculate the strength of the slope in directories which are orthogonal to each other,
commonly vertical and horizontal. Later, the contributions of the different components of the slopes are
combined to give the total value of the edge strength.
Background
An edge is defined as a pixel at which the image values undergo a sharp variation – pixels with large
element while edge detection or extraction is the act or process of finding pixels that belong to the borders of
the objects. The detection of edges is useful in the following areas: reduction of data dimension, preservation
of content information, inspection for missing parts and measurement of critical part dimensions using
gauging. Other areas include identification and verification of electronic user interface display, object
detection and tracking, structure from motion and distance calculation.
CNN stands for Cellular Neural networks and was invented in 1988 by Chua and Yang [1] at the
University of California regular (rectangular, hexagonal, etc) array of mainly identical dynamical systems
called cells which satisfy two properties: most interactions are local within a finite radius and all states
values are continuous valued signals.
Enhanced Canny Edge Detection Algorithm
The Canny algorithm (known as Optimal Edge Detector) [4] first requires that the image be smoothed
with a Gaussian mask, which cuts down significantly on the noise within the image. Then the image is run
through the Sobel algorithm. This process is hardly affected by noise. Lastly, the pixel values are chosen
base on the angle of the magnitude of that pixel and its neighbouring pixels.
Conclusion
In this work, we have proposed a new approach for edge detection by using Modified Cellular Neural
Network with the introduction of hyperbolic entry in the a22 of cloning template A. Some experimental
results of the proposed algorithm were compared with Enhanced Canny Model. It can be seen that the
results from Modified Cellular Neural Network are better than the results from Enhanced Canny Model. This
new approach ensures higher degree of controllability and stability. It is timely and economical to use the
new model for edge detection.