The top image shows the flexible image streams possible when using 4 camera heads to cover a 180-degree street scene.

There’s an increase in the popularity of multi-head cameras – but how hard is it for such cameras to display their full view – should they be thought of as a delivering panoramic views, or considered to be a CCTV system comprising 2-4 separate cameras?

Multi-head cameras are camera systems that comprise 2 or 4 camera heads in a single housing, allowing them to cover a huge area while eliminating the pixel spread seen with hemispherical cameras. Typically, multi-head cameras feature resolutions of 2MP or higher and lenses with angles of view around 90-100 degrees, minimising pixel spread and barrel distortion.

This balance allows them to offer excellent depth of field, stronger performance in low light and more complete situational awareness. There are challenges, however. Each camera must be set up carefully and it’s important to plan the way images will be displayed in monitoring centres to ensure the best possible performance is delivered to operators.

From the perspective of an operator, the secret to a multi-head camera is realising that it’s a system that delivers a 4-way split from a single camera point that gives enormous coverage with some overlap using up to 5MP sensors and 90-degree angles of view.

As an operator you don’t realise how much you want to see a pedestrian front, back and sides during or after an incident until you have the ability to do so. The beauty of this is that not only are you getting a person or vehicle on approach and departure, you’re getting faces across 90 degrees from 4 different directions, which is excellent when you’re collating admissible evidence.

But how you deliver this capability to a video wall is something that will require collaboration with security integrators and VMS suppliers.

According to Lee Shelford of Genetec, multi-headed cameras offering wide views have benefits when they are supported by clever software.

“Multiheaded cameras are useful if there is some intelligence behind them, for example if the system can stitch the images together well on screen or if clicking at a point will enable an onboard PTZ to focus in further,” Shelford says.

Over at BGW Technologies, Mark Shannon agrees.

“Multi-head cameras are increasing in popularity,” he says. “Some are better as panoramic views, and some are better as individual views. The key point to these cameras is to understand what their purpose is and then work out where and how they best fit into a VMS display.

“For example, a 180-degree camera might be very well suited for a panoramic viewing set up but a 360-degree variant might be better by showing individual streams for 4 different directions or even two 180-degree combined panoramic streams.

“Like most of these products, however, while they are increasing in sales, they are still only a very small percentage of the surveillance market.”

Tony Luce of Network Optix says multi-head cameras are a great way to capture more data with less software licenses.

“In Nx Witness multi-sensor cameras are stitched and treated as a single object by default,” he says. “Users can treat each sensor as a separate camera, as well, by pulling in individual RTSP streams from each sensor.

“It’s really up to the end-user or integrator as to how these cameras will best fit in a customer’s system. Some may prefer to treat the camera as a single wide-view device. Others may prefer individual sensors to be used as individual cameras. Nx Witness accommodates both.”

Meanwhile, Brett Hansen of Milestone says multi-head cameras should be considered if the application suits them as they can also maximise budget.

“These cameras should be deployed in certain applications where they allow the optimal use of budgets,” Hansen says. “It’s worth bearing in mind that this strategy is only effective with functional de-warping technology.”

Something end users and integrators need to make sure of is that the cameras they chose work well with their VMS solution.

Andrew Cho of EOS says management of multi-head cameras is a strength of Digifort.

“We have already integrated many models of multi-sensor cameras,” Cho explains. “We don’t do the stitching from our end, but we integrate the stitched image from the camera if they provide that.

“Then from a surveillance client, we have the ability to create a screen style to whatever shape or size to fit that panoramic image exactly as it intended without stretching or squashing.”

#securityelectronicsandnetworks.com