Research Article
A Novel Two-Stage Spectrum-Based Approach for Dimensionality Reduction: A Case Study on the Recognition of Handwritten Numerals
Table 4
Recognition rate corresponding to different feature vectors for the English dataset MNIST and Arabic/Farsi dataset Hoda.
| | Feature set | Number of features in feature vector | Feature selection method | Accuracy % | Proposed method | PCA method | RP method |
| English dataset MNIST | Initial_S | 133 | — | — | — | 91.93 | E- | 103 | * | | | 93.17 | E-S 2 | 79 | * | | | 94.88 | E- | 103 | | * | | 93.82 | E- | 79 | | * | | 90.71 | E- | 103 | | | * | 93.51 | E- | 79 | | | * | 88.39 |
| Arabic/Farsi dataset Hoda | Initial_S | 133 | — | — | — | 90.41 | A/F- | 94 | * | | | 92.60 | A/F-S 2 | 58 | * | | | 95.12 | A/F- | 94 | | * | | 94.04 | A/F- | 58 | | * | | 89.00 | A/F- | 94 | | | * | 91.07 | A/F- | 58 | | | * | 83.66 |
|
|