Multi-sensor cameras are a boon to wide-area surveillance. You can install one camera where four may have been required for a 360° view in the past, saving not only the cost of additional cameras, but the cost of wiring, power, and installation too.
Because multiple camera streams flow down a single ethernet cable, you can avoid purchasing larger network switches and reduce the number of per-camera licenses required for a video management system (VMS). Beyond cost savings, multi-sensor cameras can deliver better image quality than fish-eye lenses for the same coverage area.
As the size of these once-bulky cameras has gotten smaller and their capabilities continue to expand, demand for them is building—for indoor as well as outdoor installations. In this period of rapid innovation in multi-sensor cameras, there’s a wide range of options available, with an equally wide range of quality and capabilities. To make the most of your multi-sensor camera purchase, here are five things to look for when considering your options.
Size and Flexibility
Early multi-sensor cameras were a bit of an eyesore. Their bulky boxes relegated them to outdoor installations only. Newer models are available now that are extremely compact, no thicker than a credit card, but still maintain weather- and vandal-resistance, so they can be installed anywhere indoors or outdoors.
If you intend to mount the cameras in lower positions, make sure the lenses can tilt upwards for increased coverage near stairwells or down hallways. They should support vertical tilt angles from at least 10° to 100°.
Also, consider a screwless design, so you can easily move individual sensor positions around the housing, and make sure each camera has an independent zoom, tilt, twist, and yaw adjustment. Some cameras offer motorized aspects that move the lenses around, but as you will generally set up the camera for the coverage you need and leave it, these features may add unnecessary costs.
Also look closely at the marketing around frames per second (FPS) and megapixel (MP) calculations. Some manufacturers list the MPs per sensor as the total, which makes the camera appear to deliver higher quality than it does. You need to understand the total MP and the MP per sensor.
When comparing FPS, consider how you will set up/use the camera to achieve smooth motion capture. The marketing specs may list FPS on a per-image basis, but you may be sharing that FPS between multiple sensors. The resolution and/or features you enable can change what you may be able to achieve within a given FPS.
While artificial intelligence (AI) features such as advanced analytics and image optimization are now common in security cameras, not all implementations are the same. Algorithms based on machine and deep learning are useful for recognizing vehicles and humans and detecting attributes such as vehicle/clothing colors, objects, hats, bags and more. Having this additional information can make motion analytics like line crossing, direction, and loitering almost entirely error free.
When evaluating AI functionality, consider the number of attributes that can be described for objects being recognized since this directly impacts search success. Make sure the camera you purchase supports AI analytics “per sensor” without restrictions for the type of data captured. Some manufacturers limit their AI functionality to object recognition in certain lines, and charge a premium to detect the attributes belonging to those objects. Consider a camera that provides full AI support so you can make use of all AI parameters.
For AI functions to work at night, colors that define objects must be maintained. To keep all sensors in color mode as long as possible, make sure the camera you choose delivers superior low light performance. There are cameras on the market today that deliver low light support up to .3 Lux.
Some cameras also use AI to inform the image processing system about which objects in an image, such as a face, are important to capture with the highest clarity. Look for a camera that uses AI to automatically adjust shutter speed, auto exposure, gamma correction, and noise reduction to enhance images.
Finally, make sure the camera you choose offers analytics across multiple sensors—people detection on one, vehicle detection on another, and so on.
The AI capabilities in a new camera are of no use without a VMS able to interpret and utilize the vast amounts of metadata being supplied. Your camera plug-in should provide a user interface that integrates directly within popular VMSs and delivers full access to all the AI capabilities for either real-time proactive notifications/alarms or forensic searching. It should be open for customization with other third-party analytics as well.
A few of the new multi-sensor cameras support an automatic handoff to pan, tilt, zoom (PTZ) cameras with AI-based auto-tracking so that whenever it sees a particular image—a human, face, or vehicle, etc.—it can trigger the PTZ to navigate to that area for a closer look. With both multi-sensor and PTZ cameras having fully embedded AI, the ability to seamlessly auto-track people or vehicles is fast and seamless to make sure you get the best possible coverage for any events within range.
An open platform also helps you keep pace with fast-evolving technology. To minimize cybersecurity risk and future-proof security cameras, make sure your camera enables you to add capabilities in the field easily with regular updates and enhancements. The multi-sensor AI camera you buy should be easy to update remotely and host third-party analytics plug-ins through an open platform and SDK/API. Check to see that you can add features such as automated license plate recognition (ALPR) or advanced business and operations analytics without replacing the camera.
The last thing you want is to have new camera technology expose your wider security system—or the entire network—to cyberattacks. To help ensure the security and integrity of video data, make sure the multi-sensor camera you purchase includes FIPS 140-2 level 3 encryption and level 3 compliance, vandal-resistant technology, identity-based authentication, and an extra layer of security that protects critical security parameters on the device. The cameras must be completely NDAA and possibly TAA compliant as well, especially if they’re used in high-security applications within the government, law enforcement, healthcare, and education.
Ease of Installation
Mounting your multi-sensor camera should be easy. Look for cameras that hang with hooks or snaps from a mount, allowing one person to easily attach with simple tools. Even something as mundane as the power requirements can have a direct impact on installation costs and complexity.
Many multi-sensor cameras can require high power over ethernet (PoE) demanding as much as 60 watt-power injectors. This might require separate power injectors over a typical PoE or PoE+ network switch. Consider a multi-sensor camera that can be run over PoE+ for maximum compatibility and efficiency, and make sure to budget for your power needs before installation.
Multi-sensor cameras come in many sizes and configurations, with a wide range of features. Whether you’re upgrading, expanding or installing a new system, it will serve you to understand the nuances of these powerful cameras so you can deploy camera technology that serves you best today and well into the future.
Adam Ring is senior manager of tech services and sales engineering at i-PRO Americas Inc.