Methods | Highlights | Sensitivity under different FPI values | Scores | ||||||
---|---|---|---|---|---|---|---|---|---|
\(\frac{{1}}{{8}}\) | \(\frac{{1}}{{4}}\) | \(\frac{{1}}{{2}}\) | 1 | 2 | 4 | 8 | |||
Chudzik [31] | Fully convolutional neural networks (FCN) | 0.187 | 0.246 | 0.288 | 0.365 | 0.449 | 0.570 | 0.641 | 0.392 |
Orlando [42] | CNN-based features, RF | 0.06 | 0.09 | 0.10 | 0.29 | 0.37 | 0.49 | 0.67 | 0.294 |
Habib [38] | Tree ensemble | − | − | − | 0.18 | 0.20 | 0.38 | 0.58 | 0.2109 |
Adal [25] | Semi-supervised learning | 0.024 | 0.033 | 0.045 | 0.103 | 0.204 | 0.305 | 0.571 | 0.184 |
Proposed method | DLC feature, NB | 0.013 | 0.026 | 0.052 | 0.104 | 0.209 | 0.400 | 0.669 | 0.210 |