Protecting remote facilities is a difficult task, in terms of budget, manpower and system complexity. Fortunately, technology advances continue to provide exciting capabilities and more affordable deployment. One such technology available to system integrators and owners of remote facilities is Camera Auto Follow.
Camera Auto Follow is the capability to automatically control a PTZ camera’s pan, tilt or zoom in an attempt to keep a specified object centered in the camera’s field of view. This algorithm is a closed loop control that continually identifies the target, its current position in the camera’s field of view and its direction of travel. As the target moves from the center of the image, it automatically adjusts the camera’s physical position using pan, tilt or zoom commands to keep the target centered in the camera’s field of view. This continues until the target can no longer be detected, a user takes manual control or after a set time out period. The automatic control of a PTZ camera has considerable value by acting as a force multiplier, allowing the first responder to react to a situation without the need to also control the camera. By automatically following an intruder, it can also document activities in and around a facility without the need for operator intervention.
Flexible Deployment Options
Camera Auto Follow capability can physically reside in three places: the camera, an edge device or a server. This allows a great deal of flexibility when designing a new site, or protecting an existing one. To add auto follow capability to a site, existing cameras may be replaced or new cameras with embedded auto follow technology installed. Another option is to retrofit existing cameras with the addition of an edge device or a server equipped with the capability. Each type of implementation has its pros and cons. The good news is you don’t need to be designing a brand new system or replace all your existing cameras to acquire this functionality.
Aid in Identification
Although “detection” is important with remote facilities, the time required to respond to an event often makes it difficult to apprehend the intruder on the spot. In these cases, it is important to be able to “identify” the perpetrator. It is a male or female, hair color, clothing, etc. To aid in this effort, most camera auto follow algorithms include the ability to control zoom. Controlling zoom means, that in addition to actually keeping the intruder centered in the camera view as they move, the system also zooms in to fill the camera view with the intruder’s image. The extent of zoom depends on many factors, including target speed and camera reaction time, but the result is a more detailed image, allowing a better identification of the intruder.
Each facility has different security challenges. What is considered an event at one location may not be of interest at another. To that respect, today’s camera auto follow algorithms allow the user to specify the types of objects that are of interest. Therefore, if “human” intruders are of interest, but you don’t want to follow maintenance trucks or wildlife, the algorithms allow specifying targets that fit into the security mission of a facility, or a specific site. This added level of intelligence, referred to as classification, allows for security customization per site, and further reduces operator workload, with the system only responding to defined target types.
Another valuable feature Camera Auto Follow brings to the remote facility operator is a variety of ways to utilize the capability within a security system. Three popular designs include Always On, Manual by Scene, Manual by Target or Slew to Cue.
“Always On/Off” is the simplest implementation. Auto Follow is either on or off and triggers based on the camera scene. Once enabled, the algorithm will analyze the entire scene, or part of the scene, when an object of interest is detected, it will begin following until it loses the target, a time out is reached, or a user takes manual control.
“Manual by Target” is the ability for the user to manually invoke auto follow and then select the specific target of interest by clicking directly on the target within the live video view. Normally a security operator would need to continue to control the camera to maintain visual contact with the intruder, while at the same time take further actions to react the event, such as notifying local authorities, or dispatching other security personnel to the location. The ability to select a target for automatic following frees security personnel to take action in response to the event. This has further benefits in situations where there may be more than one target in the scene and the operator wants to select a specific one to follow.
Finally, there is engagement via “Slew to Cue”. Similar to Manual by Target, Slew to cue involves the initial steering of the camera to a location based on input from another sensor, such as a door alarm, fence sensor or another camera enabled with video analytics. Once redirected, the camera searches this new scene to identify the target of interest and begins automatic follow.
Camera Auto Follow is a very powerful feature to consider as an addition to your security arsenal. The ability to track an intrusion hands-free, allows your security personnel freedom to perform other tasks in support of the event. It can also help provide additional details to identify the perpetrator, is flexible in terms of installation options, can discriminate by object classification and can be used stand alone, or in conjunction with other sensors. If you design security systems for remote facilities, or manage the security for a company that includes remote locations, camera auto follow is a useful technology to consider.