Shoulder Surfing Detection

2 min

Overview

Shoulder Surfing Detection identifies if there is more than 1 face in the camera view. This is used to protect customer data which may be visible on the agent's screen from being seen by someone other than the agent. Note: camera doesn't need to be able to see the 2nd person's face -- only a shoulder or other body part is required.

We recommend only changing the "Common" settings without guidance from Tetherfi support.

Feature Settings - Common

TRS_SHOULDERSURFING_DETECTION_ENABLED

Enable detection of multiple faces/persons

Default: true

TRS_SHOULDERSURFING_VIOLATION_TIMEOUT

Timespan in seconds for which shoulder surfing is detected beyond which a multiple person event will be raised.

Default: 10

Feature Settings - Advanced

PERSON_DETECTOR_MODEL_THRESHOLD

Threshold for person detector

Default: 0.2

PERSON_DETECTOR_MODEL_CONFIDENCE

Confidence for person detector above which a positive identification is done

Default: 0.3

PERSON_DETECTOR_MODEL_RUNFREQUENCY

Run frequency for person detector in milliseconds

Default: 500

PERSON_DETECTOR_MODEL_SWAPRGB

1 = enable feature, 0 = disable feature

TRS_USE_PERSON_DETECTOR_FOR_MULTIFACE - If this is set to true this will use Person detector model.