3 Emirati students create a system that reveals exposure to sinking artificial intelligence

In one of the laboratories of the University of Zayed, and between the computer screens and the complicated codes, three students have worked on a project that might save endless spirits, as innovation was not only an academic idea, but a humanitarian message that carried it on the responsibility of a student team that includes Hamad Al -Zaabi, Saif Al -Zaabi, and Majid Al -Hammadi, under the supervision of Dr. Sarah Kahil, to find a solution to the problem of drowning that kills more than 300,000 people annually in the world, most of them children. The students told «Emirates Today»: “We have pushed a shocking statistic about the rates of drowning children in the swimming pools to think about a solution to confront this problem, as the owners of ages from one to 14 years are the most vulnerable to drowning, despite the presence of rescuers and surveillance cameras.”

They added: «From here the idea was born to focus on creating an artificial intelligence system to detect drowning in a timely manner, so that it is able to distinguish between natural swimming movements and any abnormal behavior indicating drowning, while sending an immediate warning to save those who are subjected to drowning before it is too late.

Intensive tests

They continued: «After extensive tests, we chose the artificial intelligence model (Yolov8) thanks to its high accuracy and speed, then we developed an integrated web platform that provides a live broadcast from the swimming pool cameras, and immediate alerts when monitoring any emergency situation, and multiple support for distance supervision, and the protection of data with encryption technologies High ».

The students pointed to the test of the innovative system model on various data to simulate different scenarios, as it included a person who swims naturally, another begins to resist and drown, and thirdly loses consciousness and floats without movement, and a group of people one of them is in danger, and the rest swim naturally.

The most prominent results

The students said: «The results showed the success of the system in identifying various cases by a precision of 99.7%, as it was able to distinguish between natural swimming and drowning situations, as well as the possibility of improving the model through additional training on wider data with no less than the accuracy of predicting in terms of cases Drinking, the system’s immediate response and the sending of an immediate alert to the concerned authority to quickly intervene to save the person, stressing that these experiences have proven the effectiveness of innovation and its ability to provide a practical solution that contributes to reducing the accidents of drowning ». They added, “We invented a group of visual data for underwater actions, which contained dozens of scenarios for drowning and swimming, and explained more than 69 thousand video frameworks, which is an unprecedented scientific treasure that will be published soon to serve researchers globally.”

Project impact

They explained: «The impact of the project did not stop at the university’s borders, but rather extended to become a real societal contribution, as innovation is not just a technical tool, but rather a solution that can save the lives of children and others who are subjected to drowning, and the data collected will help scientists and researchers develop more One of the technologies in the fields of public safety, smart homes, and smart cities.

They said: “We faced various challenges such as fast water movement, environmental noise, and the extent of the system’s ability to expand, but in the end we succeeded in developing a reliable system that suits various environments.”

“What my students presented is not just an artificial intelligence system, but rather a clear indication that technology is able to serve human if it is used properly, this project proves that the real innovation lies in its ability to make the difference in the life the people”.

Innovative students:

. We developed an integrated web platform that provides a live broadcast of swimming pool cameras, and sends immediate alerts when monitoring any emergency.

. Innovation test was done on various data to simulate 3 scenarios, and he was able to identify various cases with a accuracy of up to 99.7%

. We created visual data for underwater actions, which included dozens of scenarios for drowning and swimming, and explained more than 69,000 video frameworks.