Table 3.
Results from the tests performed on synthetic images.
| Subset | Calibration | CLV compensation | Segmentation | Differences | R | ||
|---|---|---|---|---|---|---|---|
| Max abs. | Mean abs. | RMS | |||||
| 1 | Our method | Our method | NR | 0.003 | 0.001 | 0.001 | 0.999 |
| Uncalibrated | Our method | NR | 0.008 | 0.002 | 0.002 | 0.997 | |
| Uncalibrated | Imposed CLV | NR | 0.010 | 0.003 | 0.004 | 0.998 | |
| Uncalibrated | Priyal et al. (2014) | NR | 0.010 | 0.002 | 0.003 | 0.998 | |
| Uncalibrated | Worden et al. (1998) | NR | 0.034 | 0.006 | 0.008 | 0.981 | |
| Uncalibrated | Imposed CLV | Bertello et al. (2010) | 0.011 | 0.002 | 0.003 | 0.983 | |
| Uncalibrated | Imposed CLV | Chatterjee et al. (2016) | 0.039 | 0.008 | 0.012 | 0.955 | |
| 6 | Our method | Our method | NR | 0.029 | 0.003 | 0.005 | 0.942 |
| Uncalibrated | Our method | NR | 0.026 | 0.004 | 0.006 | 0.927 | |
| Uncalibrated | Imposed CLV | NR | 0.016 | 0.004 | 0.006 | 0.979 | |
| Uncalibrated | Priyal et al. (2014) | NR | 0.704 | 0.014 | 0.030 | 0.186 | |
| Uncalibrated | Worden et al. (1998) | NR | 0.037 | 0.008 | 0.011 | 0.890 | |
| Uncalibrated | Imposed CLV | Bertello et al. (2010) | 0.007 | 0.003 | 0.003 | 0.969 | |
| Uncalibrated | Imposed CLV | Chatterjee et al. (2016) | 0.023 | 0.008 | 0.011 | 0.953 | |
| 8 | Our method | Our method | NR | 0.025 | 0.001 | 0.002 | 0.989 |
| Uncalibrated | Our method | NR | 0.034 | 0.002 | 0.003 | 0.993 | |
| Uncalibrated | Imposed CLV | NR | 0.044 | 0.003 | 0.004 | 0.993 | |
| Uncalibrated | Priyal et al. (2014) | NR | 0.755 | 0.008 | 0.026 | 0.300 | |
| Uncalibrated | Worden et al. (1998) | NR | 0.035 | 0.006 | 0.009 | 0.972 | |
| Uncalibrated | Imposed CLV | Bertello et al. (2010) | 0.219 | 0.005 | 0.009 | 0.801 | |
| Uncalibrated | Imposed CLV | Chatterjee et al. (2016) | 0.046 | 0.007 | 0.011 | 0.937 | |
| Priyal et al. (2014) | Imposed CLV | NR | 0.038 | 0.001 | 0.002 | 0.992 | |
| Priyal et al. (2014) | Priyal et al. (2014) | NR | 0.751 | 0.007 | 0.021 | 0.434 | |
Notes. Columns are synthetic subset ID number, calibration, CLV compensation, and segmentation methods applied on the data, maximum absolute differences, mean absolute differences, RMS differences, and the Pearson coefficient between the disc fractions calculated in each case and those derived from the original Rome/PSPT images.
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.