Skip to main content

Table 2 MAP performance comparison of the proposed technique with recent CBIR techniques on the image benchmark of the Corel-1 K

From: Scene search based on the adapted triangular regions and soft clustering to improve the effectiveness of the visual-bag-of-words model

Class name

Proposed technique with step size = 10

[43]

[42]

[41]

[28]

[44]

Africa

70.22

73

65

63.50

74.60

58.73

Beach

78.95

72

70

64.20

37.80

48.94

Buildings

80.89

79

75

69.80

53.90

53.74

Busses

97.87

100

95

91.50

96.70

95.81

Dinosaurs

99.29

97

100

99.20

99

98.36

Elephants

94.65

75

80

78.10

65.90

64.17

Flowers

89.11

86

95

94.80

91.20

85.64

Horses

97.04

82

90

95.20

86.90

80.31

Mountains

81.84

69

75

73.80

58.50

54.27

Food

82.25

90

75

80.60

62.20

63.14

MAP

87.22

82.30

82.00

81.07

72.67

70.31