Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Trajectories processing procedure with their use in agent-based pedestrian model.

More »

Fig 1 Expand

Fig 2.

Raw trajectories of vaccination center waiting room with red polygon for transformation matrix estimation (left) and perspective transformation of trajectories (right).

More »

Fig 2 Expand

Fig 3.

K-distance graph for estimation of optimal value of Eps.

More »

Fig 3 Expand

Fig 4.

Clustering of start (dot) and end points (cross) of all 408 trajectories.

More »

Fig 4 Expand

Fig 5.

Data visualization for individual classes.

More »

Fig 5 Expand

Fig 6.

Visualization of cluster analysis results with t-SNE.

More »

Fig 6 Expand

Fig 7.

Trajectories from training dataset.

More »

Fig 7 Expand

Fig 8.

Pairplot of downsampled training dataset.

More »

Fig 8 Expand

Fig 9.

Pairplot of SMOTE training dataset.

More »

Fig 9 Expand

Fig 10.

Covariance matrix of training data set.

More »

Fig 10 Expand

Table 1.

Comparison of classification algorithms.

More »

Table 1 Expand

Fig 11.

Graph of the dependence of accuracy and training time on the number of trees.

More »

Fig 11 Expand

Table 2.

Parameters of random forest classifier.

More »

Table 2 Expand

Fig 12.

Confusion matrix of undersampled train set (left) with accuracy: 0.836 and recall: 0.7481, undersampled trainset mixed with pseudo labelled data 90% (middle) accuracy: 0.7626 and recall: 0.8217 and undersampled trainset mixed with pseudo labelled data 99% (right) with accuracy: 0.8463 and recall: 0.7558.

More »

Fig 12 Expand

Fig 13.

Graph of accuracy score (or recall score) dependence on the size of the training set.

The gray area shows when only data from the original training set was used. The accuracy score for the training set was determined by 10-fold cross validation.

More »

Fig 13 Expand

Fig 14.

Confusion matrix of the model using SMOTE data set with accuracy: 0.8471 and recall: 0.7016.

More »

Fig 14 Expand

Fig 15.

Graph of the dependence of accuracy and recall scores on the size of the training set.

The accuracy score for the training set was determined by 10-fold cross-validation.

More »

Fig 15 Expand

Table 3.

Results of model comparisons.

More »

Table 3 Expand

Fig 16.

View corrected trajectories from the test dataset (left column) and their comparison with manually defined erroneous points (right column). The green curve indicates the corrected trajectory.

More »

Fig 16 Expand

Table 4.

Model parameters used for calibration.

More »

Table 4 Expand

Fig 17.

Waiting times for 1/1 waiting (left), 1/2 waiting (middle), 2/2 waiting (right).

More »

Fig 17 Expand

Fig 18.

Waiting points location and qualitative validation of trajectory shapes of cleaned trajectories from pedestrian traffic detector, model A–basic, and model B–advanced.

More »

Fig 18 Expand