The operating principle

We don't have any weather controlling superpowers so we're not trying to stop these hazards. Instead we're trying to save as many lives as possible, by informing people about coming hazards and giving them valuable advice on how to act in these situations. So TeOmega has an soft which helps providing the best data when needed.

Preparations

We begin with the:

• Neural network (at the beginning completely stupid, because it needs training)
• Initial data: a database with labeled pictures of hazards and their opposite (for example a forest burning and a safe one) from NASA reports, satellite pictures, images shared on social networks, for training the machine

Processing information

• All the gathered information is filtered through the TeOmega double check algorithm and the images are labeled - hazards and not hazards
• The labeled images go into the machine learning (our AI) which improves all the time. (The larger the number of images, the more accurate the AI will be).
• If our TeOmega AI detects a hazard, using satellite images, people in the surrounding area will also be notified and asked if they notice any signs of that coming hazard.

Decision and Alert

• When a hazard is detected by our AI, or reported by someone, if it turns out to be a real threat, local authorities and researchers will be alerted and will also receive valuable information for decision-making.

TeOmega double check algorithm

This algorithm leads to a more precise labeling of all the images gathered for the AI. It consists of two major parts - the TeOmega Game and a team of specialists. Every time a hazard is reported, we collect pictures taken of it and satellite images and we direct them to both our game's users, and the team of specialists. Both need to check if the pictures contain actual hazards and then label them. If both parts label the pictures identically, they are further sent to our AI. Otherwise, if an image is labeled differently by the two groups, it will be sent again to both sides for relabeling, until a unanimous decision is taken. If a user makes a large number of consecutive mistakes, they will be considered lies and the user will be banned.

The Game

TeOmega is a platform that has only one purpose: saving lives. Through this game, our neural network (or AI) improves whenever a user decides to play a game and label some hazard images. With every game played a life may be saved!

Every user will be asked a series of simple questions in order to label our images. For example: Is this a flood or not? With all the gathered information we are able to help more and more in making crucial decisions on time when a hazard appears.

Our platform will also display an on-going hazard and relevant information about it as well as recommended actions.

• TeOmega game - http://teomega.infinityfreeapp.com/

Conclusions

Every year millions of people are affected by countless hazards. TeOmega's purpose is to identify coming disasters, help people which may be affected by such hazards and offer relevant information about them and professional advice.

We believe our algorithm unites people when hazards occur and that it is more than capable to accomplish all the previously mentioned tasks and help people in need. Even if we started with H2O hazards, the platform can expand and eventually become a well-known, reliable tool for all kinds of disasters.

Even if we didn't manage to develop everything we wanted in this short period of time, we are proud of our algorithm and idea of helping people, and in the end, TeOmega's platform and AI can always be improved. At last we are more than happy that we can help and also proud that TeOmega can make a difference.