3D Sensor, Camera, Infrared and Lidar or Wi-Fi: Which Technology Counts People How Accurately?
Every evaluation starts with the same question: which technology should go on the ceiling? It is usually answered with a percentage the vendor supplies, and that answer is usually wrong.
This article works through the four technologies in use today. For each it gives the operating data an IT department will ask for anyway, and at one point it does a calculation almost nobody does: what a small counting error does to your occupancy figure.
You choose the technology once. You live with the consequences for ten years.
Why accuracy claims are not comparable
Practically every vendor quotes accuracy above 98 or 99 percent. Those numbers sit side by side as if something could be inferred from them. Nothing can, because they were produced under different conditions.
Accuracy in people counting is always stated per crossing and always under defined conditions. Measuring 98 percent while individuals walk through a wide door at a calm pace, and measuring 98 percent while a school class pushes through that same door at once, are two entirely different measurements.
The questions that make an accuracy figure usable:
- At what traffic density? People per minute per metre of door width. The error does not rise linearly with density but disproportionately, as soon as people start occluding one another.
- How was the reference established? Manual counting by two independent people with reconciliation, or video re-counting? A reference that carries its own error cannot prove 99 percent.
- Over what period? A thirty minute measurement at midday says nothing about closing time.
- Broken out by entry and exit? This is the most important question, and it is almost never asked. Why, is in the next section.
- Were groups, prams, wheelchairs and shopping trolleys counted? And if so, as what?
Without a test protocol, a percentage is marketing. With one, it is a commitment you can verify at acceptance.
The calculation almost nobody does
If you only count at doors, you do not know occupancy directly. It is computed: all entries minus all exits. That subtraction hides a problem that appears in no brochure.
A uniform deviation is harmless. If a system counts both entries and exits two percent low, the difference is two percent low as well. With 200 people present that is 196. You can work with that.
The trouble starts as soon as the deviation differs between entrance and exit. And that is the normal case, not the exception: people arrive in groups and leave individually. The entrance is wider than the exit. Morning light differs from evening light. There is a queue at the exit and none at the entrance.
An example with realistic numbers. A floor records 5000 entries and 4800 exits by 6pm. Actually present: 200 people.
| Counter deviation | measured occupancy | error |
|---|---|---|
| 2 % at entrance, 2 % at exit | 196 | -2 % |
| 2 % at entrance, 1 % at exit | 148 | -26 % |
| 2 % at entrance, 3 % at exit | 244 | +22 % |
| 5 % at entrance, 3 % at exit | 94 | -53 % |
A single percentage point of difference between entrance and exit turns a two percent deviation into a twenty-six percent one. The reason is elementary: the error grows with total traffic, while occupancy is a small difference between two large numbers.
And it gets worse the larger the floor. At 20 000 entries and 19 500 exits, meaning 500 people present, the same two-versus-one percent constellation already produces a 41 percent deviation.
Occupancy is a small difference between two large numbers. Small errors in the large numbers are large errors in the result.
Two practical consequences follow.
First: an accuracy figure not broken out by entry and exit says nothing about occupancy measurement. Ask for the breakdown.
Second: if you genuinely need occupancy, dwell time or utilisation, cover the floor rather than just the doors. A system that sees who is currently on the floor computes no difference and therefore accumulates no drift. It simply counts what is there.
Technology 1: 3D stereo vision
Two lenses at a fixed distance deliver two slightly offset images of the same scene. From the shift between them, a processing unit inside the device calculates the distance to every pixel. The same principle by which two eyes see in three dimensions.
The gain is the third dimension. The sensor does not see a surface with blobs on it, but a height map. Two heads side by side are two maxima and therefore two people. An adult and a child differ in height. A shopping trolley has no head profile. Exactly the cases where simpler methods fail are resolved by the depth information itself.
All image processing happens inside the device. What leaves it is text: a count, a coordinate, a timestamp. No video stream.
Operating data your IT department will ask for
- Power: Power over Ethernet to IEEE 802.3af, class 0, nominal 48 V. Consumption between 7.5 and 13 W depending on model. No separate mains supply at the ceiling, no socket, no electrical work at the mounting point.
- Cabling: one cable per sensor, shielded Cat-5e or better, up to 100 m, Gigabit Ethernet over RJ45.
- Network: IPv4 and IPv6, DHCP, HTTPS and, usually decisive for approval on a corporate network, IEEE 802.1x. Without port authentication a device will not be admitted to the network in many organisations.
- Data delivery: REST interface with Swagger documentation, plus active data push over HTTP(S), FTP(S), SFTP, MQTT(S), TCP or UDP. MQTT is the reason such sensors attach to a building management system without bespoke development.
- On-device storage: up to three years of counting data, depending on the number of configured counters. A network outage therefore does not create a gap in the record.
- Light: from 2 lux. That is less than dimmed museum lighting. A method that needs brightness is ruled out in an exhibition.
- Mounting height: depending on model 2 to 6 m for ordinary rooms, 6 to 20 m for halls, atria and station buildings. The height determines the field of view and therefore how many devices a floor needs.
- Mounting tolerance: roughly 15 degrees of tilt in one axis, 5 degrees in the other. A ceiling need not be perfectly level, but it cannot be arbitrarily skewed either.
- Environment: 0 to 45 degrees, 20 to 80 percent relative humidity non-condensing, ingress protection IP40. Which means unambiguously: indoors. In this form factor the technology is not intended for outdoor use.
- Reliability: calculated MTBF of roughly 73 years at 40 degrees ambient, to IEC 61709. At 65 degrees it falls to about 29 years. Heat is the governing factor, not running hours.
Limits, stated plainly: large floors need correspondingly many devices, and the mounting points have to be planned. This is not a device you hang somewhere and hope for the best. Covering a floor without gaps requires overlapping fields of view, and those follow from ceiling height and layout.
Technology 2: camera with facial recognition
A video camera records, software evaluates. With facial recognition, people can be uniquely identified and recognised again, which in theory also yields dwell time and repeat visits.
Three points argue against it.
The perspective. Cameras usually hang at an angle, not vertically. Anyone standing behind someone else is occluded. Precisely at high density, where the data would be worth most, the gaps appear.
The infrastructure. Video has to cross the network and land on storage. With several cameras over months this becomes an installation of its own, with its own budget, its own maintenance and its own failure risk. The data load is orders of magnitude above a device that sends only text.
The legal framework. Facial features are biometric data. Under the GDPR they fall under Article 9 as a special category of personal data; in Switzerland they are sensitive personal data under the revised FADP. Processing them requires a legal basis, a data protection impact assessment, a deletion concept and notice to the people affected. That is not a formality, it is a second project alongside the counting project.
Storing faces solves a counting problem by purchasing a data protection problem.
Technology 3: light and laser, meaning infrared and lidar
Both emit light and measure the time until it returns. That yields a distance, and many distances yield a shape.
Infrared is the classic in the door frame: often already installed, inexpensive, proven for decades. The price is accuracy. If the system works with single beams, a larger person occludes a smaller one completely, and two side by side become one. For a daily total at a quiet door that is fine. For an occupancy calculation, per the maths above, it is not.
Lidar uses a focused laser, reaches further and measures considerably more precisely. A high quality sensor covers a substantial area, but costs accordingly per unit.
Both share a sensitivity to their surroundings: indoor humidity, and rain, snow or fog outdoors, distort the time-of-flight measurement. Additional attributes such as view direction generally cannot be retrofitted.
A genuine advantage: nobody can be uniquely identified from time-of-flight. In data protection terms these methods are unproblematic.
Technology 4: Wi-Fi and Bluetooth
Sensors pick up the radio signals of devices people carry and try to recognise them again by MAC address. The method has been overtaken from several directions at once.
The sample is wrong. Anyone carrying two phones, work and private, is counted twice. Children and many older people carry none and are missing entirely. What gets counted is devices, not people, and the ratio between the two is neither known nor stable.
The basis was deliberately removed. Operating systems now randomise the MAC address on every connection. That is a necessary security measure by the manufacturers, and it renders re-identification useless. An installation built on it loses value with every system update, without anyone having done anything wrong.
And in data protection terms it is delicate. A MAC address is a unique identifier, and datasets exist that link such identifiers to individuals.
The four side by side
| 3D stereo vision | Camera | Lidar | Wi-Fi / Bluetooth | |
|---|---|---|---|---|
| Accuracy under crowding | very high | moderate | high | low |
| Floor coverage possible | yes | limited | yes | no |
| Real-time data | yes | yes | yes | no |
| Network data load | low (text only) | high (video) | high | moderate |
| Light and shadow | from 2 lux, insensitive | sensitive | insensitive | no effect |
| Weather, humidity | indoor, IP40 | depends on housing | sensitive | no effect |
| Personal data | none | biometric | none | unique identifier |
| Future-proof | yes | yes | yes | no |
| Lifetime cost | low | high | low | considerable |
Data protection is an architectural decision, not an afterthought
Switzerland has had a revised Federal Act on Data Protection since 1 September 2023; the EU has the GDPR. Both follow the same idea: personal data may not be processed without justification, and biometric features count as particularly sensitive.
For a tender this creates an uncomfortable asymmetry. If a method can identify people, you have to keep explaining why you use it, how you protect the data, how long you keep it and how you inform those affected. If a method cannot identify anyone in the first place, that discussion disappears entirely: in the project, with the works council, in audit, and toward your visitors.
The most robust data protection concept is the one where personal data never comes into existence.
That is why 3D sensors do without facial recognition. Not out of technical inability, but because the evaluation happens inside the device and the image stays there. What leaves is a position and a timestamp. Nobody can be reconstructed from that, not later, and not with better software.
Checklist for your tender
Seven questions that separate a commitment from a percentage:
- Accuracy broken out by entry and exit, with test protocol and stated traffic density.
- Is the floor covered, or only the door line? That determines whether dwell time and occupancy drift.
- What data leaves the device? Image, video, or text only?
- Does the device support IEEE 802.1x? Without it, approval on the corporate network fails.
- How many devices does your layout need at your ceiling height? That answer belongs in the quotation, not in the rollout.
- How long does the device store data itself if the network drops?
- What interface is there, and is it documented? A REST interface with Swagger is an integration project of days; an undocumented one is a project of weeks.
How we do it
The ANALYSIT Counting System uses 3D sensors and covers the floor, not just the door line. You see in one-minute resolution how many people are present, how they distribute across the zones of your floor and how long they stay where, looking back over 730 days. Because it counts what is there rather than subtracting, the drift error from the calculation above does not arise.
Counting is anonymous. No images, no facial features and no device identifiers are created, and therefore no personal data that would have to be protected, reported or deleted.
If you want to know how many sensors your floor needs and what can be measured on it, talk to us. We look at the layout and the ceiling height and tell you both before the quotation, not after.