24-08-2012, 05:02 PM
A New Proposed Method for Image Segmentation Based on Gray Scale Morphological Transformations
gray scale.pdf (Size: 199.79 KB / Downloads: 56)
Abstract
Image segmentation is one of the most important steps leading
to the analysis of processed data. Its main goal is to divide an
Image into parts that have a strong correlation with Areas of
the Real world contained in the Image .Image Segmentation by
Mathematical Morphology is a Methodology based upon the
notions of Reconstruction and Gradient method of an Image.
Reconstruction is a very useful operator for Image Filtering,
Segmentation, and Feature Extraction. In this paper a new
Method is proposed based on the notion of regional maxima
and makes use of Sequential reconstruction algorithm and
Morphological Gradient .The present paper has two main
sections, first is reconstruction of original Image from blurred
Image by eliminating noise. Second is segment the Image by
applying Morphological gradient method this method produced
good result over conventional methods.
Introduction
Reconstruction is a very useful operator provided by mathematical
morphology [1, 2]. Although it can easily be defined in itself,
it is often presented as part as a set of operators known as
geodesic ones [3]. In this paper Reconstruction is applied for
grayscale Images.
Definition: Let A and B be two grayscale images defined on
the same domain Da and such that A ≤ B. The grayscale
reconstruction ρa (B) of A from B is obtained by iterating
grayscale geodesic dilations of B “under" A until stability is
reached.
Sequential reconstruction algorithm:
This algorithm reduces the number of scannings required for
the computation of an image transform, sequential or recursive
algorithms have been proposed [6]. They rely on the following
two principles:
1) The image pixels are scanned in a predefined order, generally
raster or anti-raster,
2) The new value of the current pixel, determined from the
values of the pixels in its neighborhood, is written directly in the
same image, so that it is taken into account when determining
the new values of the as yet unconsidered pixels.
Here, unlike for parallel algorithms, the scanning order is
essential. This type of algorithm was first introduced for the
computation of distance functions [6] and then extended to a
number of morphological transformations [7, 8]. Among others,
grayscale reconstruction can be obtained sequentially by using
the following algorithm, where information is first propagated
downwards in a raster scanning and then upwards in an antiraster
scanning.
Expermental Results
In this work now turn to several experiments made with the
algorithm introduced above. For all tests, in this study use
a 8-neighborhood system of order1. For example in this study
diffident cloth textures, bark textures, lincon and Monalisa
Images are tested. The experimental results are shown
below. The results are better than gradient Images on original
Images. All Images are 64X64 sizes.
Conclusion
The present study on Image processing is a collection of
techniques that improve the quality of the given image in some
sense. The techniques developed are mainly problem oriented.
In this paper Morphological approach is made, the edges in the
images are clearly marked and are better visible than that of
primitive operations. The Reconstruction Algorithm described
in present study has a potentiality to generate new concepts
in design of enhanced Images. A new algorithm for image
segmentation has been implemented using morphological
transformations. The gradient operator illustrates that it can
be useful to consider edges as two-dimensional surface. This
allows the combination of gradient direction and Reconstruction
Algorithm is useful for extraction of phase regions. It does have
much effect, when implemented on reconstructed Images.
This algorithm has been tested on various images and verified
the result.