Count people anywhere in real-time. No doorway required.

About

Radar based counting. No impact from lighting, weather, temperature or reflections. No ceiling mount required.

Xandar is the first company to pioneer the concept of people counting with radars – going back as early as 2014. After numerous successful development years and receiving over 6 related patents in the field, the solution is now finally available commercially for various applications. To understand the basic concept, people counting by Xandar learns and maps out the “average motion” a single body generates in a given space. This information is constantly learned via AI and analyzed in real-time provide people counting data.

Why is it unique?

Vs. Monovision:

Unlike what you see in hollywood movies, people counting via cameras is extremely difficult because of the limitations of 2D imaging characteristics. If a group of people are standing around far away making little motion, 2D image processing will not be able to identify them as people. They would look like a background.  

Vs. 3D vision:

This is the reason why companies sometimes uses 3D stereo vision cameras in order to obtain depth information. Unfortunately the range is severely limited for depth calculation and therefore it is most often used above a doorway from the ceiling looking down. Due to this limitation, only a maximum range of 2.5M x 2.5M (standard doorway) can be covered.

Vs. Thermal Vision:

Another method of people counting is by using thermal lens / cameras. As people generate heat, a general body shape created by heat signatures can identify humans from objects. This has been a popular choice of people counting for supermarkets as it would avoid counting shopping carts as people. Unfortunately, accuracy of thermal counting systems can be affected if a group of people walk through at the same time or if weather (temperature) conditions are too extreme. Hot weather = blending with 36.5 degrees body temperature. Cold weather = people wearing thermal jackets can be missed as not enough heat signature can be read.

Vs. Hybrid Vision:

New method of using thermal + depth cameras have emerged recently that combines the pros of 3D vision (Depth) and privacy protection of thermal imaging. These systems are however, again used only over doorways and also face some of the same limitations as other camera based systems.

Other Radar Based Counting solutions:

Some of the newer competitors in the field of radar-based people counting is at the early stages of development (where Xandar Kardian was 6 years ago). One of the limiting factors to this is the need to have people spread out evenly from each other. In a lab, this may be ok, but in the real world, people’s proximity is never uniform in a given area.

Use cases

The list of advantages that Xandar’s people counting solution provides unique use case opportunities.

No impact from weather / lighting:

This allows Xandar to count people in areas such as outdoor bus stops or taxi que stands. Transportation services can now know how many people are waiting at which locations.

No camera:

Count people in a meeting room without risk of privacy or security breach. Confidential materials and password keystrokes can all be compromised! Even if the systems do not provide HD resolution, the camera itself can be replaced. With Xandar, since it can be even installed behind a ceiling panel, there is absolutely no way a camera can be hacked into it.  Furthermore, hotels and lodging industries can now safely know about the number of occupants in a given room without having privacy concerns.

10M range:

A standard doorway is around 2 to 2.5M. For locations with larger entrances such as shopping malls and stadiums, this would mean that you would need to place multiple cameras to cover multiple doors. 1 Xandar radar can cover 3 to 4 camera ranges providing better ROI.

Install anywhere:

Not every real estate space has adequate ceiling heights or even ceilings at all for camera based solutions. This is why even in some shopping malls only a “snapshot” of certain zones are able to get people flow data. Airports, train stations, shopping malls, schools, etc., all have locations with high ceilings. In these areas, you are able to use Xandar’s solution against a wall to get people counting in areas that wasn’t possible before.

Count via zones:

Xandar’s solution can be easily tuned to count people in a zone. For example, in a supermarket scenario, doorway counters can provide information about how many TOTAL number of people entered the store. However, you are not able to get more specific information about how many of them actually walked into a specific aisle. With Xandar, that is made possible. Knowledge of total people count vs. people in a specific zone (wine aisle) vs. POS data (how many wine sold) can give important converate rate data.

Related patents (granted)

Method and Apparatus for Foot Traffic Counting and Moving Direction Recognition using UWB Radar Sensor – KR 10-1928417
Method and Apparatus for moving target detection – KR 10-1876809
Systems and methods for background differentiation – KR 10-1726690
Method and Apparatus for counting the number of people using Radar – KR 10-1678313
Method and Apparatus for Counting number of object using UWB Radar – KR 10-1582404

*Above patents have been filed under Hanyang University (XK is a spin off from the university lab) and have since been acquired by WRT LAB – a subsidiary of Xandar Kardian Inc. ALL of the above patents have been filed in EP and USA as well and are currently patent-pending.

Contact

If you are interested in talking to us about a new project,
please send us a message.

Contact Us