Research Article
A Novel Image Retrieval Based on a Combination of Local and Global Histograms of Visual Words
Table 1
MAP of the proposed image representation on a vocabulary size of 200 visual words and pixel stride of 5.
| Vocabulary size & features% used | 20 | 50 | 100 | 200 | 400 |
| 10% | 77.40 | 80.27 | 81.98 | 84.01 | 82.57 | 25% | 77.74 | 80.92 | 82.02 | 84.12 | 83.16 | 50% | 77.85 | 81.06 | 82.53 | 84.25 | 83.01 | 75% | 78.08 | 81.16 | 83.07 | 84.33 | 83.24 | 100% | 78.36 | 81.32 | 83.27 | 84.38 | 83.29 | MAP | 77.88 | 80.95 | 82.57 | 84.21 | 83.05 | Standard deviation | 0.3609 | 0.4050 | 0.5899 | 0.1522 | 0.2905 | Confidence interval | 77.43–78.33 | 80.44–81.44 | 81.84–83.30 | 84.02–84.40 | 82.69–83.41 | Standard error | 0.1614 | 0.1811 | 0.2638 | 0.0680 | 0.1299 |
|
|