Skip to main content
. 2018 Jan 11;8:538. doi: 10.1038/s41598-017-18993-5

Figure 5.

Figure 5

Influence of the use of the window descriptor features using classification. Above: the mean error using each window descriptor on each dataset. Below: the time needed for extracting 10,000 observations using each window descriptor. Grey bars represent the results obtained by using only the best resolution found during a 10-cross validation process.