|
In the modern information era, digital images have been widely used in a
growing number of applications and the effort on edge enhancement has been focused
mostly on improving the visual perception of images that are unclear because of blur.
In general, the popular edge enhancement filtering is carried out with the help
of traditional filters (Nick et al., 1988; Bin et al., 1993; and Day-Fann et al., 2006).
But these filters do have some problems, especially while enhancing a noisy
image. Mainly focusing on the clarity of the image and the number of computations done
for enhancing the image, we developed a novel approach. The edge enhancement
done by smoothing filters decreases the complexity and also increases the quality of
the image (Jin, 1990). The basic aim of edge enhancement is to modify the appearance
of an image to make it visually more attractive or to improve the visibility of
certain features. The edge enhancement technique
enhances all high spatial frequency detail in an image, including edges, lines and points of high gradients. In this approach,
the details of edges in an image can be obtained by subtracting a smoothed image
from the original (Cheevasuvit et al., 1992). This subtractive smoothing method has
been used as the simplest way to obtain high spatial frequency image and this method
of edge enhancement makes the image brighter and real edges are detected.
Edge enhancement filters enhance the local discontinuities at the boundaries of
different objects (edges) in the image. All the basic edge enhancement filters are based on first
and second order derivatives (Ilya et al., 2000). Hence, this approach for edge
enhancement requires more computations.
Here, we present a new idea for edge enhancement, wherein image smoothing
filters are used for edge enhancement. In order to decrease the number of computations,
we replace the differentiation mechanism with the integration technique, The image to be detected for edges should be smoothened by applying
any image smoothing filter. From the original image, the smoothened image should
be subtracted. The image smoothing filters, viz., mean filter, median, mode, circular,
cone and pyramidal filters, are used in our study.
|
