Free Access
Table 5.
Comparison between smoothing techniques.
Method | Sen | Spe | Acc |
---|---|---|---|
(%) | (%) | (%) | |
Median + Lerr | 81 | 99 | 90 |
Cosine + Lerr | 92 | 90 | 91 |
Hamming + Lerr | 93 | 81 | 87 |
Gaussian + Lerr | 92 | 88 | 90 |
No prefilter + Lerr + Ltv | 94 | 97 | 96 |
Notes. The first four rows show performance results obtained by applying smoothing (with different kernels) before the CNN using Lerr to train the network. The last row presents the performance metrics of CNN with the loss function (Eq. (4)) without smoothing. Data set 1 was used.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.