Table 5. Semantic category-wise comparative analysis of recall with state-of-the-art CBIR techniques on the Corel-1000 image collection (bold values indicate category-wise best performance).
Semantic categories | Zeng et al. [34] | Douik et al. [20] | Feng et al. [40] | ElAlami et al. [39] | Mehmood et al. [15] | Proposed technique |
---|---|---|---|---|---|---|
Africa | 14.50 | 15.20 | 10.50 | 16.10 | 14.61 | 16.57 |
Beach | 13.04 | 09.00 | 08.20 | 19.30 | 14.92 | 13.57 |
Buildings | 14.12 | 11.80 | 07.40 | 19.10 | 16.05 | 16.13 |
Buses | 17.84 | 19.60 | 09.60 | 12.60 | 19.17 | 19.49 |
Dinosaurs | 20.00 | 20.00 | 12.00 | 10.90 | 19.59 | 19.72 |
Elephants | 14.10 | 12.80 | 08.10 | 16.30 | 17.53 | 14.00 |
Flowers | 18.96 | 19.20 | 10.60 | 12.90 | 17.03 | 20.00 |
Horses | 18.36 | 18.60 | 12.00 | 14.40 | 17.26 | 19.48 |
Mountains | 14.45 | 11.80 | 06.60 | 18.60 | 16.49 | 14.21 |
Food | 15.76 | 12.20 | 08.90 | 14.80 | 15.79 | 18.80 |
Avg. Recall | 16.11 | 15.00 | 9.39 | 16.10 | 16.84 | 17.19 |