Worldwide evolutions and events show that monitoring and managing crowds has become crucial. Just think of increasing urbanisation, the rise of mega-events and the Covid-19 pandemic. At the same time, there is a growing concern for privacy, resulting in ethical debates and, ultimately, appropriate legislation. CrowdScan balances both.
CrowdScan originated at the University of Antwerp and the Belgian R&D hub imec and took off in 2020. The company provides a wireless crowd density system that predicts size and density of large crowds. This information can be highly valuable for large-scale event organizers, public transportation companies and cities.
But this innovative start-up offers much more than a unique solution for counting and monitoring people. Thanks to its highly accurate and real-time data dashboards, customers can make smarter decisions and fine-tune policies where needed.
“We believe that crowds can be accurately and precisely measured while respecting the privacy of the individual.”
– BEN BELLEKENS CEO, CROWDSCAN
Kicking off at Tomorrowland festival
CrowdScan’s unique sensor technology was invented by three academic researchers: Ben Bellekens, Stijn Denis and Maarten Weyn. At the world famous dance festival Tomorrowland they were allowed to test the technology on a large scale. “We had equipped the Freedom Stage and the VIP area with our sensors, so we could analyze the size of the crowd and compare it with the organization’s own counts and cameras. Our sensors worked, and with the permanent real-time monitoring, we discovered that we had a unique security solution on our hands”, says Ben Bellekens who is now CrowdScan’s CEO.
With Anton Dierickx, who previously worked for the Antwerp Police department, the fourth founder came in. As he was managing the safety of big city events, his field experience was crucial. “He immediately saw the added value of our application,” Bellekens explains, “especially when it comes to safety.”
CrowdScan’s system is capable of detecting critical situations in which dangerously large amounts of people are present. Needless to say that this is indeed extremely valuable in preventing potential lethal accidents. Conventional crowd monitoring systems often rely on a computer vision approach. Also, other technologies like smartphone, GPS and Bluetooth have been proposed for estimation of crowd location, density, and dynamics during major events. These existing techniques may raise data privacy concerns and might become obsolete with the emerging data privacy laws.
Ben Bellekens states: “Our measurement methodology does not make use of camera images and operates entirely in a device-free manner in which the influence of the physical presence of human individuals on low-energy radio frequency signals in the environment is used. Using this approach, it is impossible to determine the identity of individual crowd members, which therefore results in a system that is inherently privacy-conscious.”
Over the last five years, experiments have been performed in real-life large-scale environments containing thousands of human individuals and they indicate that the system can obtain very high crowd size estimation accuracy, with median errors ranging between 5% and 10%. The combination of a gateway with sensors ensures accurate measurement and real-time processing, tailored to each situation and location.
The sensor nodes with which (a part of) the magic happens. These battery powered devices can be installed in a very flexible way. CrowdScan is the first and only company to use sensors and radio frequencies to measure crowd density
Added value for public transportation
In 2021, in order to realize an investment round, the founding team had to rethink which markets were useful. An exercise that paid off, because in June 2021 the company was able to successfully finalize its first investment round.
“We discovered that our technology also offers a lot of advantages for the public transportation market. Avoidance of crowding situations in stations is crucial to foster sustainable mobility and also improve daily operations”, says Bellekens. “The current video-based systems do not allow to estimate the number of present people with sufficient accuracy, due to possible obstructions, clutter, and poor light or weather conditions, and involve high deployment costs in complex environments. Moreover, they require electrical power supply and may incur in privacy problems.”
In 2022, CrowdScan is ready to put maximum effort into finding the right partners to enter different markets. Various international tech magazines are writing about CrowdScan as a very promising internet-of-things solution. Recently the 13th IoT Innovation World Cup® at HANNOVER MESSE named the company as one of the three top innovators in the category of smart cities.