IVS Camera Features: Line Crossing and Intrusion Zones

An ivs camera is a camera that runs intelligent video analysis rules on top of plain motion detection: instead of alerting whenever enough pixels change, it alerts only when a moving object crosses a virtual line you draw (line crossing, also called a tripwire) or enters and stays inside an area you mark (intrusion detection). Used well, these rules can cut many of the alerts caused by passing cars, swaying trees and shadows. Not every model has them, so the first job is to check whether yours does, and the second is to draw the line or area in the right place.
How IVS rules differ from a normal motion zone
A normal motion zone, covered in detail in the motion detection settings guide, asks one question: did something change inside this box? A leaf, a moth, headlights on a wall and a burglar all give the same answer.
IVS rules ask more specific questions about how something moves:
- Line crossing (tripwire). You draw a line across the picture, for example across a gate or the edge of a driveway. The camera tracks moving objects and alerts only when one crosses that line. Most implementations let you choose a direction: A to B, B to A, or both. A person walking along the pavement parallel to your fence never crosses the line, so no alert.
- Intrusion area (region entrance or perimeter). You draw a shape, often with several corners rather than a simple rectangle. The camera alerts when an object enters the shape and, on many models, only after it has stayed inside for a short time you set. That delay is what filters out a bird flying through.
- Related rules found on some cameras and recorders include region exit, loitering, and object left behind or removed. Names and availability vary widely.
Many IVS cameras combine these rules with object classification, so a line or area can be set to react to people only, or to people and vehicles. That combination is what removes much of the noise: a rule that needs a person-shaped object to cross a specific line in a specific direction is very hard to trigger by accident.
One thing IVS does not change is the underlying sensor. If your camera wakes up from a heat sensor, as many battery models do, the analysis can only run on what the camera records after it wakes. The PIR sensor explainer covers why those cameras miss some events entirely, which no drawn line can fix.
Step 1: Check whether your model supports IVS
Support depends on the exact model, its firmware and the app version, not on the brand name on the box. Many home Wi-Fi cameras offer motion detection, a detection area and human detection, but no line or intrusion rules. Check in this order:
- The manual and the specification list. Look for words such as IVS, intelligent analysis, smart detection, line crossing, tripwire, cross-line, intrusion, region detection or perimeter.
- The camera's settings in the app. Open the camera, tap the settings icon, and look under sections such as Alarm Settings, Smart Detection, Intelligent Analysis or Event Settings. If you only see sensitivity, a detection area grid and a human detection switch, your model most likely has no line-crossing rule. Menu names may differ slightly by app version.
- Firmware updates. In the device information page, check whether a firmware update is offered. An update occasionally adds detection features, but do not count on it.
- The seller. If the listing is vague, ask the shop whether your exact model supports line crossing or intrusion rules, and where they are configured. Ask before buying if the feature matters to you.
The ONVIF route
Some models support ONVIF, the open standard that lets cameras and recorders from different makers work together. If yours does, a network video recorder or recording software can receive the camera's video and, depending on the recorder, either receive the camera's own analysis events or run its own analysis on the recorder side. This is why you see searches for an "intelligent camera onvif" setup: the intelligence can live in the camera, in the recorder, or in both.
The practical consequence is that a camera with no line-crossing option in the app might still get line-crossing alerts through a recorder that offers it. If you want to explore that, start with how to add a Yoosee camera to an NVR and then check the recorder's own manual for its analysis features. Whether rules created on a recorder are passed back to the phone app is something to confirm for your specific combination.
Step 2: Fix the view before drawing anything
A tripwire is drawn on a flat picture of a three-dimensional scene, so the camera angle matters even more than it does for plain motion zones.
- Aim so people cross the frame sideways where possible. Analysis tracks movement more reliably when a person moves across the image than when they walk straight towards the lens and only grow larger.
- Keep the area you care about in the middle distance. Too far away and a person is only a few pixels tall, too small to classify. Too close below the camera and you see the top of a head, which also confuses person classification.
- Check the night view. Switch to infrared mode, or wait until dark, and look at the same scene. Shadows, reflections from a window or wet ground can make an object hard to follow at night even if it is obvious by day.
- Remove obvious noise sources such as a hanging plant right in front of the lens.
If repositioning means a ladder or touching a wired power point, switch off the power first, have someone steady the ladder, and use a qualified electrician for any mains wiring.
Step 3: Draw a line-crossing rule that works
Exact screens vary, but the steps are similar on most cameras and recorders that offer the feature:
- Open the rule and enable it. Look for Line Crossing, Tripwire or Cross Line, then switch it on.
- Draw the line on a real boundary. Drag the two end points over the live picture so the line sits on something a person must cross: the threshold of a gate, the edge of a driveway at your property line, the bottom step of a porch.
- Make the line long enough. It should run across the whole route, with a little extra on both ends. A short line can be walked around.
- Keep it away from frame edges. Objects entering at the very edge are partly out of view, so tracking may start too late to see the crossing. Place the line where a person is fully visible on both sides.
- Choose the direction. If you only care about people coming in, pick the direction that points into your property. Leaving the house will then not trigger an alert every morning.
- Set the target type to people, or people and vehicles, if offered.
- Save, then walk it. Cross the line at a normal pace in both directions, then walk near it without crossing. Check which events appear in the app.
A common mistake is drawing the line along the ground where the pavement is visible but your property is not. A person walking past on that pavement may cross the line in the image even though they never set foot on your land, because the camera sees a flat picture. Move the line further into your property until only people actually entering cross it.
Step 4: Draw an intrusion area that works
- Enable the rule, often called Intrusion, Region Intrusion, Perimeter or Area Detection.
- Outline the zone by its corners. Follow the edge of the space you want to protect, such as the courtyard, the back garden or the area in front of a shuttered shop door. Leave out the street, neighbours' windows and tree tops.
- Set the time threshold if offered. This is how long an object must stay inside before an alert. A short delay filters out birds and leaves blown through, while still catching someone who walks in. There is no universal value: try a low setting, review a day of events, and adjust.
- Set the target type and sensitivity. Start in the middle and change one setting at a time.
- Test with a walk-through at the far edge of the area, where a person looks smallest, and again after dark.
Line or area? Use a line when there is a clear entry point, such as a gate, a door or a driveway entrance. Use an area when people can arrive from several directions, such as an open yard. You can combine them on many models: a line at the gate plus an area around the back door.
Step 5: Link the rule to alerts, recording and deterrents
A rule that fires but does nothing useful is easy to miss. After setting it up:
- Check recording. Make sure an event recording is saved when the rule triggers, either to the memory card or to the storage you use. Look in playback for clips marked with the event.
- Check phone alerts. If the event list shows the trigger but your phone stays silent, that is a delivery issue, not a detection one. Work through why app notifications stop arriving.
- Add a siren or light if you have one. On models with a built-in siren or spotlight, linking it to a precise rule such as line crossing, rather than to all motion, means it fires for people at the gate and not for leaves at noon. The siren setup guide explains how to schedule it so it does not disturb neighbours.
- Use a schedule. Many cameras let a rule run only at certain hours, for example when the shop is closed.
Reading the symptoms
- No alerts at all, even when you cross the line. The rule may be saved but switched off, the alarm master switch may be off, or the camera may be in a schedule window where detection is disabled. Also confirm the person was fully visible on both sides of the line.
- Alerts for people on the pavement. The line or area reaches into the public space in the image. Pull it further inside your boundary.
- Alerts for cars only. The target type may include vehicles, or headlights at night are being tracked. Set the target to people, or move the rule away from the driveway entrance.
- Misses at night only. The scene is too dark or too noisy in infrared. Add lighting near the entry point, clean the lens, and check for reflections from glass.
- Misses people walking straight at the camera. Re-aim so the route crosses the frame, or move the line closer to where people pass sideways.
- Works for a while, then becomes noisy. The camera has shifted slightly, or plants have grown. Redraw the rule after any change of angle.
When to get help
Contact the seller or a professional installer if your manual lists line crossing or intrusion but you cannot find the setting after updating the app and firmware, if rules keep resetting after a restart, or if you want to combine several cameras with a recorder that runs its own analysis. Bring the exact model name from the label on the camera, the firmware version from the device information page, and a short description of what you drew and what happened. That makes it much faster to tell a missing feature from a setting problem.
Frequently asked questions
What does IVS stand for on a camera?
Usage varies: makers expand it as intelligent video surveillance, intelligent video system or similar. In practice the label means the camera or recorder can apply movement rules such as line crossing and intrusion areas, so the term on a spec sheet is less important than the specific rules listed beside it.
Is human detection the same as IVS?
Not quite. Human detection is a filter that decides whether a moving shape looks like a person, and it usually applies to the whole detection area. IVS rules add where and how the person moves, such as crossing a line in one direction. Many cameras with human detection have no line or intrusion rules.
Can I add line crossing to a camera that does not have it?
Not inside the camera itself, unless a firmware update adds it, which you should not count on. If your model supports ONVIF, a recorder or recording software with its own analysis may provide line crossing on its side. Check the recorder's manual and confirm compatibility with the seller first.
Does line crossing use more memory card space?
Usually it uses less than plain motion detection, because the camera records fewer pointless events. The space used still depends on resolution, clip length and how often the rule triggers. If you record continuously, the rule changes alerts but not how fast the card fills.
Will animals trigger an intrusion area?
A large animal can, especially at night or if the target type is set to all objects. Setting the target to people and adding a short time threshold helps, but no classification is perfect, so expect the occasional dog or cat in your event list.
