The XXII start-up relies on firefighters to train its algorithms to detect fires

A start-up that approaches firefighters. XXII, a software publisher using artificial intelligence techniques and more particularly computer vision, approached the departmental fire and rescue service (SDIS) of the Meuse. The start-up founded in 2015 in Paris has developed a real-time video analysis solution and wants to offer a new fire detection feature from September. Ultimately, the idea is to connect the software to the cameras of a city so that it can analyze, detect and alert quickly in the event of a fire starting.

Learning in real conditions

XXII is still in the learning and validation phase and wants to ensure that its systems recognize lights, such as car lights. The start-up has already conducted fire start experiments but in other contexts such as in recycling companies. “But the fire is different, it is necessary to have other situations”says Damien Mulhem, Chief Data Officer and co-founder of XXII.

For now, since it is very rare to find burning cars, the start-up has worked with quite a bit of learning. But it is preferable to learn and understand the departures of fires on the vehicles directly on the ground. “Our algorithms must be fed with knowledge”simply explains Damien Mulhem.

XXII therefore naturally approached the firefighters in order to join the training exercises carried out by the SDIS de la Meuse on the night of June 27 to 28 and during the day of June 28. This rapprochement was made easily because the XXII teams have a volunteer firefighter who made the link between the two. The SDIS, through the voice of Lieutenant-Colonel David Hantzo of the SDIS de la Meuse, evokes a “partnership” and ensure that there is “no economic interest” and “This approach is in the collective interest.”

Explanations of the different parameters

During these exercises, “fires on depolluted cars are lit in different configurations so that new firefighters can train”, summarizes David Hantzo. For XXII, this experiment should both enable it to validate its algorithms, ensure that they detect fire starts in real conditions, and teach its neural networks about other fire start circumstances. “The goal is to film a fire at multiple distances and different heights so that our systems are agnostic of distance and viewpoint angles”explains Damien Mulhem.

XXII wishes to better understand the outbreaks of fire on vehicles and above all to teach its neural networks the different types of fire and how to analyze them. Here, experts who are firefighters can explain live to the XXII teams the parameters to be taken into account (colors of flame, height, smoke, etc.) and their influence on the evolution of the fire, both to better understand the evolution of the fire and transmit the right information. Then, the software will be able to automatically detect these fires, analyze them and warn the firefighters by telling them what stage the fire has reached and assessing its severity.

The many uses of AI

The software developed by XXII can be used by anyone, whether it is a community or a private company and ultimately individuals. Damien Mulhem evokes a “agnostic platform that can be used in different ways”. Beyond fire detection, XXII would also like to develop a drowning detection solution. But the data on this subject is scarce, so it must be simulated.

Another subject that interests the start-up is that of the detection of special vehicles (firefighter, ambulance, police, SAMU, vehicle transporting organs, etc.). Once these vehicles have been detected, and the XXII software has been interconnected with smart cities, it is possible to control traffic lights to facilitate the circulation of these vehicles.

For firefighters, artificial intelligence can have many uses, but the cost of this technology currently limits its use. David Hantzo evokes the use of artificial intelligence coupled with sensors on nozzle holders to very quickly alert firefighters when the fire is about to degenerate when several elements are combined (temperature, light, smoke, heat). There is also personal protective equipment with probes that recreate an operational situation (heat, smoke, etc.) using artificial intelligence. Uses that could multiply in the years to come.

We would like to give thanks to the writer of this post for this incredible material

The XXII start-up relies on firefighters to train its algorithms to detect fires

Check out our social media profiles as well as other pages related to them