Swimming pool drowning monitoring method based on binocular vision

Abstract

The invention discloses a swimming pool drowning monitoring method based on binocular vision, and the method comprises the following steps: installing a plurality of binocular camera devices above a swimming pool at certain intervals, carrying out the shooting from top to bottom, and transmitting an image to a computer in real time; detecting a head image on the water surface by using a target detection algorithm according to the real-time image; calculating the three-dimensional coordinates of all point positions of the target head image, marking the highest point of the head, and judging whether the breathing part of the head is underwater or not; tracking the target by using a target tracking program algorithm; storing and analyzing the three-dimensional coordinates of the tracking target in real time, calculating the underwater duration of the human head breathing part, performing danger assessment according to the underwater single duration of the human head breathing part and the underwater time sum in unit time, and if a danger threshold value is exceeded, giving out a drowning alarm. Whether a swimmer is in a drowning state or not can be effectively detected in real time so as to prompt safety personnel to rescue.

Description

technical field [0001] The invention relates to the technical field of drowning monitoring, in particular to a method for monitoring drowning in a swimming pool based on binocular vision. Background technique [0002] According to statistics from the Ministry of Health, about 57,000 people drown in my country every year, which is equivalent to more than 150 people drowning every day. Many of them happen in public swimming pools with professional lifeguards. Due to the limitation of people's own physiological conditions, it is difficult for lifeguards to maintain a high degree of concentration for a long time, and it is difficult to efficiently ensure the safety of people in the swimming pool. In addition, there will be light reflections on the water surface of the swimming pool, which also makes it difficult for lifeguards to distinguish normal swimmers from drowning people in time when they are tired. For lifeguards, drowning is a low-probability event, and it is easy to p...

AI Technical Summary

Problems solved by technology

There are also many shortcomings in this type of system: 1) Sonar identification of a drowning person basically depends on how much air is contained in his body. If the lungs are full of water and the depth of the drowning, the signal obtained cannot be sure, and even no judgment is made.

3) If the body is very close to the wall, because the sonar signal contains echoes, the swimming pool wall is not conducive to eliminating noise signals, and it is difficult to extract the necessary signals for monitoring drowning

4) When there are too many people in the swimming pool, it is impossible to track and identify everyone

But there are the following disadvantages, 1, thermal imager and visible light combination equipment is expensive

2. When the camera is shooting from top to bottom, due to the interference of the swimming cap or hair, the thermal imager cannot detect the skin temperature of the human head, resulting in the failure of the whole method

3. Due to the second reason above, it is better to shoot from the side of the device, and the best is to shoot horizontally, which will inevitably cause the target to be blocked by swimmers and cause the entire set of equipment to fail

4. After a swimmer swims for a long time, the temperature of the skin on the face drops or a thin water layer is formed, and the temperature detection method of the thermal imager will also fail

Therefore, this method is not suitable for large-scale application and promotion.

Images