Page 190 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications

Chapter 6 Plant leaf disease classification based on feature selection 179

4.3.2.2 Gray wolf optimization
GWO that was first introduced by Mirjalili et al. [19] is an algo-
rithm basing on the behavior of the wolves in catching their prey.
This algorithm also focuses on the social dominant hierarchy of
the wolf.
The alphas are responsible for the leaders who take a role of
making decisions about hunting, sleeping place, time to wake,
and so on. However, in some democratic behavior, the alpha
follows the other wolves in the pack. These alpha wolves are
only allowed to mate in the pack. These alphas are not necessary
to be the strongest but the best one in managing the pack
(Fig. 6.22).
The following rank of wolf’s hierarchy is beta. These wolves
are the subordinate, who help the alpha in making decision or
other pack actives. The main role of the beta is giving the advice
to the alpha and discipliner for the pack. These wolves support
the alpha’s commands throughout the pack and give feedback
to the alpha.

Figure 6.22 Position updating in GWO. GWO, gray wolf optimization.