This differentiates it from some forms of camera auto follow whereby the p/t/z camera requires a fixed camera continually detects and provides location information for the target being tracked. This type of follow capability really doesn’t have a good name. If it did it would probably be something like “camera auto follow – slaved to a fixed camera,” but that’s a mouthful, which is probably why it’s not used.
RELATED: Your Crash Course on Creating a Vehicular Perimeter Security System
This type of camera auto follow is not necessarily bad. By default, the fixed camera needs to have geospatial awareness. Meaning, it can translate the object detected in the fixed camera video to latitude/longitude information that can then be presented to the p/t/z camera performing the auto follow, which will also require this geospatial capability.
Geospatial recognition is a good thing and it provides lots of other advantages throughout a perimeter security system. Nevertheless it’s important to understand that the p/t/z camera slaved to a fixed camera scenario does have limitations that a single
camera auto follow does not.
Here, there are three factors to consider: the need to have at least one fixed and one p/t/z camera; the need to have geospatial capability on both cameras; and the understanding that once the target leaves the fixed camera’s FOV, the p/t/z camera with auto follow will no longer be able to follow the target. The result can be a fairly limited area in which auto follow can work, namely that of the FOV of the fixed camera.
When looking to add camera auto follow to a security system, there are other related attributes that are worth exploring as well. In terms of detection, some algorithms follow pixels, while others follow classified targets. Why does this make a difference? A classic scenario is what happens when the camera is following a person, and then a car moves through the FOV. Pixel-based detection will most likely change course and follow the car as it is a moving target with many more pixels.
On the other hand, following algorithms based on classification will track the object by type. So if the auto follow was initiated for a “human,” when a “car” is introduced into the scene, it will attempt to stay with the object classified as human.
Terms for Different Methods of Initiating Camera Follow
Another valuable feature related to camera auto follow is how the feature is initiated. This typically falls into a few categories: always on, manual by scene, manual by target or slew to cue.
Always on/off is pretty simple. Auto follow is either on or off and triggers within a certain area of the scene viewed by the auto follow camera.
Manual by scene means the operator or a third-party sensor invokes the camera auto follow based on the current scene. So upon enabling this function the detection portion of the software evaluates the scene and starts to follow the object that is typically closest to the center of the view. Manual by scene can also involve a camera preset where a third-party sensor, such as an access control or a proximity sensor, trips and directs the camera to a preset scene. Once it redirects to the preset scene, it can be set up to engage camera auto follow based on the scene.
