Research Article
Color Face Recognition Based on Steerable Pyramid Transform and Extreme Learning Machines
Table 2
Performance comparison in different hybrid color spaces of LCSPT-ELM face recognition algorithm: the AR database.
| ā | ex1 | ex2 | ā | HS | | 0 | | | HS | | 0 | | |
| | 97.72 | 98.41 | 97.02 | 96.33 | 97.02 | 96.78 | 97.08 | 96.28 | 94.40 | 97.08 | | 95.25 | 96.61 | 94.72 | 95.64 | 94.69 | 93.51 | 94.58 | 97.58 | 95.59 | 93.98 | | 95.63 | 98.41 | 97.71 | 94.94 | 98.41 | 95.88 | 98.26 | 97.07 | 94.39 | 97.07 | | 96.33 | 97.02 | 97.72 | 95.64 | 96.33 | 95.88 | 97.07 | 97.67 | 95.88 | 96.78 | | 94.24 | 95.63 | 96.33 | 90.77 | 96.33 | 93.50 | 94.99 | 94.39 | 90.23 | 94.99 | | 94.02 | 99.01 | 97.02 | 97.41 | 98.41 | 96.78 | 98.16 | 97.37 | 95.98 | 98.08 | [5] | 93.75 | 96.82 | 97.71 | 95.64 | 95.69 | 94.99 | 97.85 | 97.37 | 94.99 | 97.07 | [5] | 93.55 | 97.02 | 97.72 | 95.64 | 97.72 | 94.99 | 95.59 | 95.59 | 92.91 | 95.29 | [10] | 94.25 | 97.02 | 97.72 | 95.64 | 97.72 | 94.69 | 98.56 | 97.37 | 94.10 | 97.97 | [8] | 94.94 | 97.71 | 97.71 | 96.33 | 98.41 | 94.39 | 96.18 | 95.88 | 93.80 | 95.88 | [9] | 96.33 | 97.72 | 97.72 | 95.64 | 98.41 | 94.69 | 96.28 | 96.18 | 93.80 | 96.18 | [5] | 97.02 | 97.72 | 97.02 | 94.94 | 97.02 | 96.18 | 96.48 | 97.37 | 94.10 | 98.26 | | 96.33 | 99.11 | 99.80 | 97.72 | 99.11 | 96.78 | 98.86 | 96.26 | 96.78 | 99.45 | | 95.64 | 99.11 | 99.80 | 97.02 | 99.11 | 97.07 | 99.16 | 98.56 | 96.78 | 99.16 | | 95.64 | 99.11 | 99.80 | 98.41 | 99.11 | 96.18 | 99.16 | 98.56 | 96.78 | 99.16 |
|
|
Bold values specify the hybrid color spaces and their subbands with the highest correctness.
|