Edge Enhancement of Digital Images Using Edge Profile Analysis

Dhinush Jayasuriya, N. D. Kodikara

ABSTRACT

Edge enhancement is one of the ways of sharpening a blurred image. Edges in a blurred digital image are more closely modelled as having a ramp profile. With the analysis and correction of edge profiles, edges can be sharpened. The slope of the ramp is inversely proportional to the degree of blurring. If the steepness of the slope can be increased then the blurring can be reduced. Using the local analysis of gradients, ramp pixels can be detected. Since an edge is necessarily characterized by an abrupt change in intensity values, regions on either side of an edge should have pixels with the difference in their intensity values being higher than a threshold value. The extent of the ramp is determined by the value of its first derivative. Once the range is detected, one can interpolate the ramp pixels with the intensity values of region pixels relevant to each side of the profile. The result would be a conversion of ramp profiles into step edge profiles.