Open Access

Table 2.

Sub-dataset distribution after splitting and downsampling.

Dataset Classification Quantity X M C NF FX FM FC
Training CMX Average number of image 218 1918 11 824 10 665 35 344.0 2917
Standard deviation 13.35 81.36 154.79 205.05 9.32 27.81 90.19
Percentage of the dataset 0.78% 6.87% 42.35% 38.19% 0.13% 1.23% 10.45%
MX Average number of image 207 1955 370 2201 7 69 596
Standard deviation 12.1 62.45 18.03 50.88 2.9 11.03 27.76
Percentage of the dataset 3.84% 36.16% 6.84% 40.72% 0.13% 1.28% 11.02%
Validation CMX Average number of image 25 277 1578 8409 21 255 2467
Standard deviation 12.12 60.63 145.56 228.39 40.64 71.73 194.46
Percentage of the dataset 0.2% 2.12% 12.11% 64.52% 0.17% 1.96% 18.93%
MX Average number of image 37 240 1405 8773 23 315 2466
Standard deviation 15.58 37.56 165.69 322.65 38.83 129.88 353.82
Percentage of the dataset 0.28% 1.82% 10.6% 66.18% 0.18% 2.37% 18.58%
Test CMX Average number of image 26 271 1562 8949 27 264 2467
Standard deviation 12.51 60.41 145.08 393.78 29.12 88.07 288.46
Percentage of the dataset 0.2% 2.01% 11.53% 65.94% 0.2% 1.95% 18.17%
MX Average number of image 26 270 1489 8726 45 317 2533
Standard deviation 11.51 53.77 107.64 500.91 47.86 146.73 189.15
Percentage of the dataset 0.2% 2.02% 11.12% 65.04% 0.34% 2.39% 18.9%

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.