Reactive Access Control System

Area access based on real-time facial recognition

Biometric face recognitionConcept

In integrated on-deck surveillance system employs object detection and recognition algorithms to detect and analyse intrusions in an observed area. Video feeds are processed in real time providing a reactive system to control the access level to certain areas of the ship with remote controlled locks.

It performs face or object recognition and tracking in real time. Depending on the mode selected it executes one or both of the functions: motion detection and tracking algorithm performs advanced detection of moving objects in the scene, their classification and tracking until they disappear, biometric face recognition algorithm tracks and identify multiple faces.

The detected faces are matched against the watchlist from an internal database. Recognized faces are immediately reported to the system. Critical event video data is recorded to be later reviewed at a remote station.

Determining movements and facial features Stream4s Solution

The technology is based on a proven background subtraction method to track multiple objects from a distance discerning objects in a variety of conditions

Object classification: After cameras are calibrated, it allows to perform classification based on the size and the movement speed of objects. For example, users can configure a surveillance system to determine if a tracked object is a vehicle, a single person or a group of people.

Restricted areas control: The algorithm can detect and report if people or objects enter, leave or stay in restricted areas. The events are triggered when people or objects cross predefined lines or enter polygon-shaped areas.

Tolerance to weather conditions: The algorithm ignores rain and snow, as well as trees and bushes, which are swayed by wind if used in external conditions.

Multiple face detection: Features extraction and template matching with the internal database in real time.

Face tracking: Performed in all successive frames from the video source within the camera field of view. The algorithm is able to continue tracking a face when it re-appears after being fully covered by occlusions (like walls, furniture).

Components of a Reactive Access SystemSystem Diagram

The full system is formed from these integrated components: Analogue / IP Cameras - providing video feeds to the video server, Video server - processing video data, registering access events and biometric data, Access Control + Intercom - functioning with several layers / modes of operation: reacting to events provided from the video server, normal mode of operation as an intercom with personalised access, lockdown in critical intrusion events, Alarm triggers - integrated in sensitive areas where movement is detected by motion detection algorithms from the video camera, Local / Remote command and control stations - used to monitor specific events, review livestream from cameras, customise access rules and personnel roles.>

40 Pixels min
between eyes

5 Individual faces

4kb face features

20fps Algorithm

<0.5s Face watch-list


Reduced burden of authorisation to secured areas if a person is identified from a video feed that matches the internal database watchlist
Possibility to lockdown doors reactively if movement is detected
Prioritised video live-streams in remote command and control stations where unauthorised people are detected
Operation logs and event reports with relevant video data
Enrollment of new faces from the video stream and adding them to the watchlist


Moving object detection and classification: object / person / person group / vehicle
Real time face detection and template extraction
Simultaneous multiple face or object tracking in live video
Object detection ignores rain and snow, as well objects swayed by wind if used in external conditions
Office in London, UK
Office in Kaunas, LT
Office in London, UK
Office in Kaunas, LT

Office in Kaunas, LT

Donelaičio g. 62, Kaunas, 50000


Office in London, UK

45 King William Street, London, EC4R 9AN