Poseidon drowning detection system

The Poseidon drowning detection system is a computer aided drowning detection system. The system is commonly installed in bathing areas, such as municipal swimming baths, water-parks, therapeutic baths, and other places where there is standing water in which there is a risk of drowning. It is named after the Greek god of the seas, Poseidon.

Principle of operation

The patented technology consists of a network of (overhead and/or underwater) cameras that are connected to a computer vision system that uses multiple overlapping views (epipolar geometry, fundamental matrix, essential matrix, etc.) to detect motionless bathers below the surface of the water in swimming pools.

An incident is detected within 10 to 15 seconds and the algorithms in the system allow to provide the lifeguards with a precise 3 Dimensional position of the person in difficulty. This is designed to heighten safety in swimming pools and contribute to faster rescues by lifeguards.

TV program on the principles of Poseidon

Competing companies

Poseidon has been installed in more than 200 pools world wide and has contributed to save more than 26 lives in Australia, Belgium, France, Germany, Japan, Nederlands, UK... Examples of Poseidon detection and lifeguard rescue 1 2

Poseidon is the sole company having actual rescues and holding Computer Aided Drowning Detection patents describing the technology used for both Overhead and Underwater technologies.

Some companies use standard security cameras with motion detection or similar technology or high definition digital colour CCTV type cameras (Poolview Ltd, Swimguard, and The Swimming Pool Safety Company Ltd) with flat screens to provide the lifeguards with multiple and simultaneous views of the total below water area of a swimming pool but provide no detection.

In all cases the purpose is to allow lifeguards to detect a swimmer in difficulty before they drown and sink to the bottom of the pool.

See also

This article is issued from Wikipedia - version of the 11/30/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.