The fitness function of a firefly, is the recognition accuracy of the test face images while the face recognition system is trained. The fitness value of a firefly is defined as below
\begin{equation*}
G^{F}= (Nc/Nt) X 100
\end{equation*}
Where
\\Nc = Total number of correctly recognized face images
\\Nt = Total number of test images
When the recognition accuracy is more, the fitness value of the firefly F is large therefore said to be a brighter firefly compared to the one that produces less recognition accuracy with its features. Each firefly acquires new fitness value while moving in the 'd'-dimensional space. A firefly position represents the selected features based on which the classifier works to obtain face recognition accuracy.
\begin{equation*}
G^{F}= (Nc/Nt) X 100
\end{equation*}
Where
\\Nc = Total number of correctly recognized face images
\\Nt = Total number of test images
When the recognition accuracy is more, the fitness value of the firefly F is large therefore said to be a brighter firefly compared to the one that produces less recognition accuracy with its features. Each firefly acquires new fitness value while moving in the 'd'-dimensional space. A firefly position represents the selected features based on which the classifier works to obtain face recognition accuracy.