STARTUP
FireFinder
FireFinder is the dense network of detectors, which detect fire at its very early stage. Our sensors constantly monitor air parameters in search of anomalies. If even trace amounts of smoke are detected, a single sensor sends information to a central server. Machine learning algorithms analyze the data received from the sensor network in real time to eliminate false alerts and determine the exact location of the fire and its development direction.
If the threat is confirmed, the appropriate local fire brigade services are alerted automatically.
We have prototype, which have proved their effectiveness in real conditions. We conducted controlled fires and our sensors detected the threat.
FireFinder's Story
As a machine learning practitioner, I wanted to use my knowledge a little differently than creating well-matched ads.
During my studies, I was looking for how AI or technology in general can help in the effort to counteract climate change. At one point we found wildfires to be still an unsolved problem and furthermore it is a growing problem. Currently used detection systems based on observation cameras are not a sufficient solution, so we decided to develop a system based on a dense network of sensors, that can smell smoke much earlier than the camera can see it.
We are also implementing a side project related to fire detection – SmokeFinder – AI analyzing the image from observation cameras and automatically detecting smoke. We are also working on a machine learning model that will estimate litter moisture on the basis of other meteorological parameters which is a very important parameter for foresters.
Also due to the fact that we have developed a certain fragment of universal technology, i.e. a box that can be powered from a solar panel or a battery, into which you can insert any sensor and can send measurements to the world, we are looking for smart farming solutions that we could start creating.