How to calculate True Postive, False Positive and False Negative values

I used the trainFasterRCNNObjectDetector function to train a detector with 7 classes.

Now I have a test set with 180 images and the respective labels and bouding boxes and I want to calculate the True Postive, False Positive and False Negative values. However I don't know how to do this, taking into account all labels detected, and not only the ones chosen by the function

[bboxes, scores, label] = detect (detector, Image);

How can I count all the True Postives, False Positives and False Negatives?