How technology is helping us fight wildfires

“Video surveillance” MTS.

In this article, we will talk about how modern technologies help fight forest fires. Within the team, we came up with a video analytics solution that significantly improves the fire monitoring process.

A few introductory

Forests cover almost two-thirds of the territory of Russia – more than 1 billion hectares. Occurs annually from 9,000 to 35,000 forest fires, covering from 500,000 to several million hectares. We have to allocate significant resources to extinguish fires and eliminate the consequences of natural disasters: according to the estimates of Rosleskhoz, the average amount of damage from forest fires per year is about 20 billion rublesof which from 3 billion to 7 billion rubles – loss of wood.

There are two causes of fires: natural factor and human. The second can be combated by various measures: prohibiting visits to forests during fire seasons, stopping grass burning, and fines for making fires. But nature itself cannot be penalized: lightning strikes, for example, cannot be prevented, and this is one of the most common causes of fires.

It is obvious that 100% elimination of both factors is problematic, as well as preventing fires completely. What then to do?

If it is impossible to prevent the causes, it is necessary to notice the source of fire in time and take measures to localize it. Moreover, it is very important to react quickly, since the speed of deployment of response and fire fighting services depends on it: it is easier to put out a small fire at an early stage, you can get by with fewer resources.

What detection tools can be used?

  • satellites;

  • planes and helicopters;

  • unmanned aerial vehicles (UAVs);

  • video monitoring tools.

Each monitoring method has its pros and cons.

  • satellites fly high, see a vast area, but become useless when cloudy;

  • planes and helicopters are convenient for monitoring due to the ability to cover a large area, as well as immediately localize fire sources from them. But there are nuances: the fire season in Russia is from May to September, and in some regions even from April to October. Monitoring should be constant, and if there are no fires, then the equipment flies, in fact, empty, while one flight can cost several hundred thousand rubles.

  • Drones, seemingly solve the problem of budgets, but the communication channel and battery capacity limit the range and flight time. A UAV cannot be in the air around the clock, with strong winds it becomes difficult to use, an operator is needed to control it, who sometimes needs to rest – in general, this story is also not about 24/7/365.

We saw an exit video monitoring. Simply put, the masts/towers are equipped with video cameras that, in the event of smoke or fire, determine the source of ignition using video analytics systems, and then send an alarm to the operator and other interested parties. Better to use PTZ cameraswhich can rotate 360 ​​degrees and zoom in well, for example, these:

An example of a PTZ camera from one of the MTS facilities
An example of a PTZ camera from one of the MTS facilities

Benefits of using video analytics

Firstly, if there is a conditional operator (or you can do without it), the monitoring itself is not handled by a person, but by software. This frees a person from routine work: the operator is not obliged to admire a static picture of the same type day and night.

A medical fact is that a person can watch monitors without losing attention for no more than 15 minutes, after which he needs rest or replacement.

Video analytics unloads the employee and reduces the risk of human error, and when the detector is triggered by smoke or fire, the system itself issues a light and sound alert.

Secondly, the camera, or rather the software and hardware complex (hereinafter referred to as PAC), can work those same 24 hours a day, 7 days a week, 365 days a year.

Thirdly, PAK does not require refueling with expensive aviation kerosene =)

The disadvantage of the technique and how we eliminated it

Is there any downside to this solution? Yes, and it lies in the lack of high-quality recognition in the dark. At night, the forest can also catch fire and burn out over a huge area. Is it possible to do something about this shortcoming? Yes!

Our team came up with a solution: to use not only the optical channel (the camera itself), but also thermal. To do this, you need to use a device that combines a camera and a thermal imager in one housing, for example:

An example of a bispectral rotating system camera + thermal imager
An example of a bispectral rotating system camera + thermal imager

And what about the thermal imager?

The thermal imager in our PACK provides the system with qualitatively new capabilities, unlike a conventional camera:

  • the ability to work in complete darkness;

  • smoke, haze, smog, light, shadows, reflections, rain, snow, fog – no longer a hindrance;

  • algorithm operation at long distances;

  • a general increase in the accuracy of determining the location of a fire.

Improve detection accuracy

It is not enough to see smoke or a fire and transmit an alarm, you also need to provide the dispatcher with specific coordinates of the source of fire, where the fire brigade will go.

Our 3D positioning technology allows you to accurately determine the location of the fire (up to 33 km) and the path to it for response teams. Each platform of the complex is equipped with a built-in electronic compass, as well as a system for setting coordinates and installation height. This information is automatically synchronized with the platform.

In addition, the platform remembers up to 20 (at least 9 recommended) calibration points to improve the accuracy of determining the coordinates of a fire.

To display a map of the area and fire points, the complex can use both offline vector maps (based on Open source sources) of any detail, and online GIS maps (acquisition of an API access key and Internet access is required), as well as freely distributed offline DEM -height maps (in the current version, type 30M).

Interface example
Interface example

The system supports 3 scanning modes:

• preset scan,

• linear scanning,

• pattern scanning.

For example, you can set up multiple presets to cover the entire area for continuous patrols in all weathers, with higher priority areas (detailed or more frequent scans) or quick scan areas.

At the request of the customer, the archive can be stored both on the local registrar / server, and in cloud storage. The operator’s workplace can be organized both locally, directly at the installation site, and remotely – anywhere in the world, an Internet channel of 4-5 Mbps is enough. In one system, the number of terminal devices themselves in the form of bispectral cameras is limited to 2048 pieces; in practice, it is difficult to exceed this limit. Such characteristics make it possible to build geographically distributed systems of practically unlimited scale.

Photo from a real MTS project
Photo from a real MTS project

More familiar security devices, such as other cameras, wearable recorders, access terminals, security sensors, and so on, can seamlessly connect to the system. Accordingly, the described HSS can be an element of the general security system of the customer. This will allow you to receive the necessary notifications and access to the archive based on a single platform.

Our system has proved its effectiveness in practice — last summer we successfully implemented a project in the Chelyabinsk region. There, our software and hardware complex monitors an area of ​​30 km, monitors fires and smoke around the clock, using the optical and thermal imaging spectrum with reference to coordinates on the ground and maps. The system works normally, has proven itself perfectly.

Thank you for your time on this article! If you have any questions, we will be happy to answer them in the comments.

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *