Choosing Camera for Crowd Analytics

Matrixing delivers Full-Journey Analytics

Add an on-premise NUC processor to your environment and you can matrix dozens or even hundreds of sensors into a single continuous field-of-view. Matrixing lets you do full-journey tracking of almost any size space with proven camera sensors.

Excellent Economics

By configuring the right camera to the right space, you can pay just for the coverage you need. Camera is an economical solution for everything from single-door measurement to quite large spaces.

Optimized for Crowded Environments

The top-down, ceiling view of a stereoscopic camera minimizes the blockage and impact of crowds. If you have dense crowding, top down positioning will maximize counting accuracy.

Outstanding Reliability

These stereoscopic people-sensors have been in the field for a long time. They have very few moving parts and they almost never break. It’s a technology you can deploy with absolute confidence that it works and works reliably.

LiDAR Coverage

LiDAR coverage is NEVER what the manufacturer says. Reliable people detection and classification can usually be done at about half the range claimed. However, even this will vary wildly depending on line-of-sight blockage and crowds. The LiDAR needs to put enough beams on an object to recognize and track it. In addition, LiDAR sensors typically have a significant blind spot underneath the sensor and some wave-pattern gaps in coverage. From a practical standpoint, this means that the minimum LIDAR implementation is 2 sensors even in spaces where a single LiDAR would otherwise suffice.

Camera Coverage

Camera coverage is much simpler. In general, a top-down camera will give you the coverage claimed by the manufacturer based on ceiling height. When doing full journey, also keep in mind that cameras must have an overlapping area to track people across their field of view. This will reduce the actual measured area of each camera and is NOT the same for every sensor.

Choosing between LiDAR and Camera

Both provide excellent measurement, so the choice usually comes down to environment, privacy and, most of all, cost. If LiDAR is the only workable solution then your decision is easy. If LiDARs lack of video is a PII essential, ditto. But if neither of those things are true, it’s very much a matter of picking the best TCO alternative for you space. In smaller spaces, that’s almost always camera. In medium to larger spaces, the right answer will take careful evaluation of your site but may well be LiDAR.

Xovis

Xovis sensors shine in large-scale, complex implementations. Their cameras have the best coverage options for large spaces, minimize overlap requirements for matrixing, and, when deployed with their Spider processor, scale really well.

They also provide a clean, easy to use https push mechanism and support very near real-time data.

Hella

Hella sensors provide extremely accurate people-tracking at very reasonable price-point. They’ve greatly expanded their remote management capabilities and improved their matrixing – making Hella sensors an excellent choice for any door-counting / occupancy application and for many smaller and mid-size full-journey tracking applications.

While the SOAP interface is more complex to setup, it provides a ton of flexibility in configuring the data push.