A Grey Wolf Optimizer for Modular Granular Neural Networks for Human Recognition

Table 10

Comparison of results (test #2, ORL).

Method	Best	Average	Worst
Azami et al. [43]	96.50%	95.91%	95.37%
Ch'Ng et al. [3]	96.5%	94.75%	94%
Sánchez et al. [38]	99%	98.30%	98%
Sánchez et al. [39]	98.43%	97.59%	94.55%
Proposed GWO	99%	98.50%	98%