Gray image enhancement using harmony search
الباحث الأول:
Zaid Abdi Alkareem Alyasseri
الباحثين الآخرين:
Al-Betar Mohammed, AhamadTajudinKhuder, and Mohammed A. AWADALLAH,
المجلة:
InternationalJournal of Computational Intelligence Systems
تاريخ النشر:
None
مختصر البحث:
Abstract For decades, image enhancement has been considered one of the
most important aspects in computer science because of its in
uence on a
number of elds including but not limited to medical, security, banking and
nancial sectors. In this …
Abstract For decades, image enhancement has been considered one of the
most important aspects in computer science because of its in
uence on a
number of elds including but not limited to medical, security, banking and
nancial sectors. In this paper, a new gray level image (edge preserving) en-
hancement method called the harmony search algorithm (HSA) is proposed.
HSA has been recently introduced stemmed by the musical improvisation pro-
cess when a group of musicians play the pitches of their instruments seeking
for pleasing harmony. Tremendous successful stories of HSA application to a
wide variety of optimization problems have been passed on at a large scale.
In order to evaluate the proposed HSA-based image enhancement method, 14
standard images from the literature have been used. For comparative evalu-
ation, the results of the 14 enhanced image produced by HSA are compared
with two classical image enhancement methods (i.e., Histogram Equalization
algorithm and Image Adjacent algorithm) and two advanced methods (i.e.,
genetic algorithm and particle swarm optimization). These methods, it should
be emphasized, used the same criteria ( number of edges in an gray scaled im-
ages, summation intensity of edges detected using a Sobel lter and entropy
measure) to evaluate HSA results. The HSA almost achieves the best results
in comparison with the other classical and advanced image enhancement al-
gorithms. Due to such achievements, we believe that the proposed method is
very promising and has a potential to provide a substantial addition to the
image enhancement domain.
