C-WERK®: pioneering face recognition from GRUNDIG Security

Face recognition is a method of identifying - or confirming the identity of - a person based on his or her face. This method is one category within biometric security. C WERK's face recognition function can be used, for instance, to search for people in video recordings. The function can also be used to alert in real time to specific individuals, such as registered hooligans before entering the stadium or known gambling addicts at a casino entrance.

The face recognition function enables the recognition and identification of faces. In addition, it is possible to match the recognized face with an existing photo in order to perform further automated actions.

With C-WERK® GRUNDIG Security provides comprehensive face recognition solutions. Two types of analytics are to be distinguished:

a) Analytics in the camera (Edge Analytics). With this option, face recognition takes place in the camera. The recognized face is compared with a "blacklist" or a "whitelist". Actions, such as sending an e-mail, alerting via a mobile app, etc., are performed directly from the camera. The advantage: no additional servers are required. However, with this type of analysis, aggregation can be solved less elegantly in a system with many cameras for face recognition. In addition, the renewal of the face lists cannot be done centrally. These disadvantages can be overcome by using server-based analytics.

b) Server-based analytics. Here, face recognition from many image sources is performed simultaneously on a single server. The faces are analyzed in real time and compared with the entries in a database. The maintenance of this database is done centrally. Further advantages: multiple inputs and outputs can be monitored simultaneously. This C-WERK solution also offers a counting function, dashboard display and extensive reporting. In addition, the location of the detected face can also be displayed on a site plan.

Identified faces are visualized in the site plan.

With GRUNDIG Security almost all face recognition solution requirements can be implemented both efficiently and cost-effectively. The face recognition can be freely combined with the many other analysis functions of C WERK.