In many security environments, continuous observation is not the goal. That’s why line crossing video analytics are so important —They notify us of incidents that need our attention by creating actionable events like an alerts.
This article explains how line crossing analytics work, how they are integrated within Milestone XProtect and when they provide the right level of capability for your operational needs.
Line crossing is a rule-based video analytic that detects when an object passes over a virtual line drawn within a camera’s field of view. The line is digitally configured within the image, and when an object intersects that boundary, the camera generates a structured event.
When it comes to XProtect, this processing occurs at the edge. The camera itself runs the analytic and produces metadata describing the event, such as the time of crossing and direction of travel. That metadata is then transmitted to the video management system.
Depending on the camera and analytic engine, line crossing can support direction-based detection, counterflow detection, object filtering and counting functionality. For example, a system can be configured to trigger only when a person crosses from left to right, or when a vehicle enters a restricted area.
An area-based analytic configured within the camera detects vehicles within a defined zone, generating metadata and events that are ingested and operationalized within XProtect.
Because this analytic is commonly embedded in modern network cameras, it provides an efficient and cost-effective way to monitor clearly defined boundaries without requiring dedicated analytics servers or additional infrastructure.
Once a line crossing event is generated at the camera, the next consideration is how that event is managed and operationalized within the video management system.
While the camera performs the detection, XProtect turns that detection into coordinated action.
Through validated integrations with leading camera manufacturers, XProtect ingests video analytics metadata and converts it into XProtect-native events via the Milestone Event Server, where those events can trigger rules and generate structured alarms in Alarm Manager.
When a line crossing event is received, it does not appear as a generic notification. Instead, it becomes part of a structured workflow. Operators can see the alarm in context, assign priority, acknowledge it, add notes and jump directly to the exact timestamp of the event. Because the metadata is preserved, investigators can also search and filter events later based on direction or object type, when supported by the camera.
You can see this in the image below:
Top left: The analytic is configured at the camera, with a defined detection area and rule parameters.
Top right: Live video shows the monitored scene with the defined zone applied.
Bottom left: Triggered events appear as a structured list of alarms in XProtect, providing operators with visibility into all detections.
Bottom right: Selecting an event immediately pulls up the associated video, allowing operators to review the exact moment of detection with object tracking and overlays.
This orchestration layer is where the real value emerges. Detection happens at the edge. Response, coordination and investigation happen within XProtect.
The operational coordination provided by XProtect relies on structured video analytics metadata. Because line crossing is typically generated at the edge, that metadata must be accurately translated and exposed to the Milestone Event Server in order to trigger rules and generate structured alarms.
Milestone supports this through validated integrations with leading camera manufacturers, ensuring that line crossing events remain searchable, actionable and fully integrated into system workflows.
AXIS Optimizer
AXIS Optimizer for XProtect is a validated integration suite that ensures Axis edge video analytics are fully usable within the Milestone environment. Delivered as a free, one-time installation with automatic updates, it is designed specifically to optimize the interaction between Axis devices and XProtect and supports current XProtect editions.
When an Axis camera generates a line crossing event, AXIS Optimizer enables the associated metadata to be made available to the Milestone Event Server and used throughout the system. Operators can create rules, generate alarms in Alarm Manager and search events using available object-level metadata such as direction of travel.
Because the analytic processing remains within the camera, this approach supports efficient boundary monitoring without requiring additional analytics servers, while still enabling structured workflows inside XProtect.
With AXIS Optimizer, line crossing analytics are configured directly on the camera and delivered into XProtect as structured events, enabling operators to quickly review detections with associated metadata such as direction of travel.
Bosch Analytics Service
Bosch cameras running built-in video analytics ( Bosch IVA/EVA) generate metadata-based events directly at the camera. The Bosch Analytics Service for XProtect acts as a dedicated integration component that translates those analytics events into Milestone-native events and requires XProtect 2017 R1 or later with metadata support enabled.
When a Bosch camera detects a line crossing event, the Analytics Service consumes the associated metadata stream and makes the event available to the Milestone Event Server. From there, administrators can create rules, generate alarms in Alarm Manager and use available metadata for search and investigation within Smart Client.
Depending on the camera configuration, line crossing events can include additional contextual metadata such as object trajectory or classification. This ensures that Bosch edge analytics remain fully usable within XProtect’s operational workflows without requiring separate analytics infrastructure.
Hanwha Vision Plug-in
Hanwha Vision AI cameras generate analytics events at the edge, including boundary-based detections such as line crossing. The Hanwha Vision Plug-in integrates these AI-generated events with all XProtect variants and requires XProtect 2022 R3 or later.
The integration includes a server plug-in installed on the Milestone Event Server and a client plug-in for Smart Client. When a Hanwha camera detects a line crossing event, the associated AI metadata is delivered to the Milestone Event Server, where it becomes available for rule configuration, alarm generation in Alarm Manager and metadata-driven search within Smart Client.
Because analytics processing remains within the camera, this approach supports scalable boundary detection while maintaining centralized workflow management inside XProtect.
Edge-based, line crossing analytics are often the right solution when monitoring requirements are clearly defined and limited in scope. For example, protecting a single perimeter, enforcing directional movement in a corridor or monitoring access to a restricted area can typically be handled effectively within the camera.
Because the analytic runs at the edge and integrates through the Milestone Event Server, this approach keeps infrastructure simple while still enabling rule-based alarms and searchable metadata within XProtect.
However, as operational complexity increases, so do analytics requirements. Deployments that combine multiple analytic types across many cameras may benefit from a more advanced analytics platform, particularly when investigations require deeper forensic filtering or faster review of large volumes of recorded video.
In these cases, organizations should consider shifting away from edge-based analytics to the robust offerings of server-based video analytics like BriefCam. While edge-based analytics generate line crossing events directly in the camera, BriefCam approaches this differently by analyzing movement patterns, direction and object behavior within recorded video. This allows organizations to replicate the outcomes of line crossing analytics while expanding beyond fixed boundaries to support more flexible, attribute-driven analysis.
For instance, the video below shows how law enforcement used BriefCam to identify a drug house. In step 6 of the demonstration, the operator draws a directional path into the house to identify individuals entering the premises. While this resembles the concept of a rule-based line drawn within a camera’s field of view, it is not a traditional line crossing analytic. Instead, it demonstrates how movement patterns and direction can be used to replicate and extend boundary-based detection using more flexible, attribute-driven analysis.
This image gives a closer look at path filter that was drawn for this scenario.
For more information about BriefCam analytics, check out this webinar.
If you have any questions about line crossing for your security deployment, schedule a call with one of our experts.