Proposal of a swimming pool drowning detection system using cameras and raspberry Pi based on machine learning

Abstract

Drowning deaths represent the third leading cause of accidental deaths worldwide. This is because traditional techniques for the supervision and care of people, especially children, in large pools are inefficient or, in some cases, non-existent. Nowadays, this problem has become a topic of interest for several researchers who seek to propose different methods of drowning detection. This research work seeks to propose the process to be followed to develop a drowning detection system in swimming pools using cameras and Raspberry Pi based on Machine Learning. To achieve the objective, the use of the Triple Diamond design methodology was proposed. In the development of the first diamond, the information was organized in a Lotus Blossom Diagram, then the problematic situation and the main objective were described. In the development of the second diamond, a bibliometric analysis was performed, searching for information with search equations and then sorting and filtering it, and finally including it in morphological matrices. As a result, an electrical diagram of the system and a flow diagram of the algorithm based on a Support Vector Machine were proposed.