Intelligent AnalyticsStrengthen on-premises security with deep learning-powered algorithms and detection profiles designed for Synology DVA series NVRs. Watch video Track and identifySmarter monitoring for important perimeters, entry points, and passages.People and vehicle detectionTrigger immediate or threshold-based alerts using idle times within defined zones for vehicles or people. License plate recognitionIdentify license plates and set up accompanying triggers based on existing allow or blocklists. Learn more Intrusion detectionSet up bidirectional or one-way virtual borders to issue alerts for people or vehicles trespassing. Monitor on-site occupancyKeep the number of visitors under control to avoid crowds and regulate movement.People countingGauge occupancy levels with aggregated counting of people entering and leaving an area, with automatic alerts when reaching capacity.Congestion alertsAvoid overcrowding in sensitive areas by detecting when the number of people passing through a predefined zone crosses a set threshold.Visitor trendsGenerate and export statistics to create informed infrastructure or traffic flow optimizations.Personnel detectionFast identification and authentication of individuals enable smoother day-to-day operations.

Face recognitionManage on-prem access easily with detection support for up to 10,000 personnel profiles with >97% accuracy.

Identify and reviewQuickly find unauthorized personnel and enable security teams to cross-check recordings and already approved individuals.

Face covering detectionIdentify whether your visitors are in compliance with local rules or policies and alert personnel on site to take appropriate actions.

Notes:Features described here are available on Synology DVA series NVRs only.License plate recognition is only available in select countries. Please check here for a list of countries supporting it.Testing was performed by the National Institute of Standards and Technology (NIST) in their Facial Recognition Vendor Test (FRVT) under its WILD category (using faces extracted from real-world footage).