From: Scene search based on the adapted triangular regions and soft clustering to improve the effectiveness of the visual-bag-of-words model
Total number of retrieved images
Proposed technique
Spatial level-2 technique [51]
Standard VBoW technique
Top-05
0.3601
0.3726
0.3415
Top-10
0.4829
0.5178
0.4674
Top-15
0.6901
0.7050
0.6589
Top-20
0.8513
0.8882
0.8213
Top-25
1.0243
1.0599
0.9913