Every day, every single day, devices in fire suppression systems need to be checked! When these are enormous warehouses spread across vast territories, a security guard has to constantly walk around and inspect all of them — just for this one task. The pressure gauge has become a symbol of punishment in fire safety authorities.

Millions of analog sensors are also used in gas and water distribution systems, as well as in many other industries.

We constantly have to deal with fires in so-called “expensive places”, where almost billions have been spent on fire protection systems. Yes, they have the most advanced automatic fire suppression equipment, but it still needs to be monitored. Speslab has a division that installs fire protection systems, and we constantly encounter cases where newly purchased fire suppression cylinders have already been emptied.

We will not go into who controls this and how — let’s call it the human factor. It is the human factor that controls it.

Looking at modern pressure gauges has become a kind of torture: they are getting smaller and smaller. No one designs them with a retired security guard in mind.

Devices with digital monitoring do exist, but for now they remain something of a rarity and are significantly more expensive. Therefore, they are not competitive, and as a result, they are rarely installed. Even if this statement raises some skepticism and the digital format soon starts gaining popularity (which causes even greater skepticism), the old systems — already installed across the country and around the world — will remain in operation for many, many decades.

That is why Speslab has developed an “analog-to-digital converter” for such devices — an pointer neural network. A camera is placed in front of an analog gauge; it can even be an existing video surveillance camera, as long as it has a clear view of the instrument.

Video streams from cameras are processed through Speslab’s pointer neural network, which analyzes the readings of analog instruments and converts them into digital values. Apart from the original idea, everything is relatively simple and can be solved by training neural networks using the free Gambit software, so I will not go into technical details here (but feel free to contact me — I can always explain more).

The system can operate either on a PC with the GOALcity video surveillance system installed or on a videoblazer device.

Speslab has already been called a “killer.” We are constantly killing the human factor. Video cameras are digitizing the world through the power of neural networks.

Now all analog instruments can be monitored 24/7 without replacing expensive equipment and without being affected by the human factor.

Daniil Grishanin
=Speslab=

Arrow neural network: Fire extinguishing equipment no longer needs to be checked