Information Technology Reference
In-Depth Information
True positive rate (TPR) (also referred as Recall) is defined as the proportion of
positive cases that were correctly classified as positive. TPR can be calculated using
following formula.
TP
TPR
(Re
call
)
=
(9)
TP
+
FN
Predicted positive rate (PPR) (also referred as Precision) is defined as the propor-
tion of the predicted positive cases that were correct. PPR can be calculated using
following formula.
TP
PPR
(Pr
ecision
)
=
(10)
FP
+
TP
Values of all performance measures have been shown in Table 6, for the proposed
method and all other state-of-the-art methods.
Table 6.
Performance measure values
Average
Classification
Accuracy
(%)
TPR
(Recall)
(%)
PPR
(Precision)
(%)
Methods Name
The Proposed method with Zernike Moment as a
feature
81.00
75.00
77.00
The Proposed method with DTCWT
(Level-1) as a feature (Khare et al. [2])
90.00
90.00
90.00
The Proposed method with DTCWT
(Level-2) as a feature (Khare et al. [2])
92.00
90.19
91.00
The Proposed method with DTCWT
(Level-3) as a feature (Khare et al. [2])
95.00
94.06
94.50
The Proposed method with DTCWT
(Level-4) as a feature (Khare et al. [2])
95.00
94.06
94.50
The Proposed method with DTCWT
(Level-5) as a feature (Khare et al. [2])
97.00
94.17
95.50
The Proposed method with DTCWT
(Level-6) as a feature (Khare et al. [2])
97.00
97.06
98.00
The Proposed method with DTCWT
(Level-7) as a feature (Khare et al. [2])
99.00
99.00
99.00
The Proposed method with combination of
DTCWT (Level-1) and Zernike moment as a
feature
95.00
90.47
92.50
The Proposed method with combination of
DTCWT (Level-2) and Zernike moment as a
feature
95.00
93.12
94.00
The Proposed method with combination of
DTCWT (Level-3) and Zernike moment as a
feature
96.00
93.21
94.50
Search WWH ::
Custom Search