An introduction to People Counting with 3D Time-of-Flight technology
This article is based on the webinar “An introduction to People Counting with 3D Time-of-Flight”, hosted by Clement Peltrault, product manager for people counting solutions at Tredecile.
Why use a people counting system?
If you’re a business owner, having detailed information on how people are using your space can provide a number of advantages for resource maximization. Shop managers, for example, can use people counting data to gain valuable insights into trends in customer behavior and ensure that staffing and operations are optimized to give the best possible experience to their customers.
As Clement Peltrault points out in the webinar, “Where customer traffic increases, businesses can put more people on the welcome desk, for example, and make sure that the shop’s customers have a great experience. All that will create a better occupant and user experience inside your building, your office or your shop.”
Similarly, if you’re a facility manager in a corporate building, determining how many workstations are in use at various times, whether conference rooms are being used efficiently, or how often the washrooms or cafeterias are occupied can have big implications for building management. People counting systems can be used to allocate space more effectively and, in cases, even reduce real estate expenditure and utility bills. Countless companies pay to heat, cool and illuminate unoccupied rooms every day – but people counting data enables heating, lighting and ventilation systems to be automated in response to room occupancy.
The first step to better understand the use of space is to count the number of people who go in and out of a given area. Technological solutions such as Time-of-Flight sensors have made people counting much easier and more accurate.
Time-of-Flight basics
The Time-of-Flight principle is a method for measuring the distance of an object from a sensor. A device emits a pulse of infrared light and measures the amount of time taken for the pulse to reflect off an object and travel back to a sensor. This time interval – the “time of flight” of the pulse – can then be used to calculate the distance the pulse has traveled, and hence the distance between the sensor and the object.
The ToF principle lies at the heart of Tredecile’s people counting systems. In the example below, Tredecile Evo 60m is a single pixel sensor that enables only one dimension.
Going further, a device like People Counting L-XL enables third dimensional data collection by using infrared LEDs and an 80 x 60 array of photodetectors to collect depth information at multiple points several times per second. Where the pixels in a conventional sensor capture measurements of color and brightness, each “pixel” in the People Counting L-XL measures distance, enabling the system to build up a 3D image of the surrounding environment.
“The difference with this kind of sensor is that we have a wider field of view and then a wider emission,” says Clement Peltrault. “This means that we can light a larger scene, and we can start seeing more objects and more information.”
For instance, if we place a sensor onto a ceiling and one person passes below it, the sensor starts by illuminating the scene with an infrared light which is invisible to the naked eye (1). The yellow triangle is the Field of View, what the device sees. The signal is first emitted on the arms, then the shoulder and the head of the person. Bouncing back to the sensor (2), it calculates the time differences between all the height levels and helps identify any kind of object, from a person to a box, a chair, or an animal.
Time-of-Flight sensor capturing 3D images:
Clement Peltrault continues, “In the picture (on the left), the color scale is exactly what you can get from the depth sensor. The red color indicates a very short distance from the sensor to the object, whereas the blue object is far from the sensor.”
With this raw information and the different colors, we are able to distinguish various shapes from head (red), shoulders (orange) to arms (yellow and green) or any other kind of objects. The onboard processing then extracts useful information from this 3D image. Tredecile algorithms are used to recognize individual features such as shoulders and arms while ignoring background objects, enabling people counting data to be extracted efficiently.
Why Time-of-Flight technology is ideal for people counting?
While a number of different people counting technologies exist, Time-of-Flight technology makes people counting easier, more accurate and more secure. Using an active infrared emission, the 3D ToF sensors from Tredecile are able to detect people, objects and the rest of the scene without being affected by the color of the ground or people’s clothing. For instance a person wearing white clothes and passing on a white floor will still be detected effectively.
Clement Peltrault: “Tredecile’s ToF sensors are easy to install – simply position a sensor in a doorway or corridor to count people entering or exiting a space – and work regardless of ambient lighting conditions with very low computational demand.”
One of the most noteworthy advantages of Tredecile’s ToF people counting systems is that they are intrinsically GDPR compliant. Due to the relatively low spatial resolution of the sensors, no identifying information about individuals can be captured. In addition, since all image processing is carried out on board, the only data that is output from the device is counting logs. These are much less sensitive than image data from a GDPR perspective.
Tredecile People Counting solutions
Tredecile people counting systems are designed to be as easy to use as possible for system integrators, IoT solution providers or service providers. The devices host a local web interface that can be easily accessed by any WiFi-enabled device, giving immediate access to all relevant parameters for easy configuration and set-up.
Once installed, people counting devices can either be connected via Tredecile’s IoT Platform or easily integrated into existing platforms via a direct data transmission in HTTP or MQTT. This enables multiple spaces and entrance or exit points to be monitored together.
Tredecile’s ToF people counting solutions include the People Counting L-XL, suitable for wide doorways and larger spaces, and the People Counting M, which is optimized for single doorways and narrow passages.
To find out more about how a Tredecile people counting system could make your business more efficient, get in touch with a member of the Tredecile team today.
