Swimming pool anti-drowning multi-camera target tracking method and device, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a swimming pool anti-drowning multi-camera target tracking method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring images shot by a plurality of cameras located above the water surface of the swimming pool to obtain original images; inputting the original image into a target detection model to perform swimming pool human body target detection so as to obtain a swimming pool human body target detection frame; performing target tracking on the swimming pool human body target detection frame by adopting a DeepSORT algorithm to obtain a swimming pool human body target swimming track; outputting the target swimming trajectory of the human body in the swimming pool; wherein the target detection model is obtained by training a convolutional neural network by taking a plurality of images with position labels of the human body target as a sample set. By implementing the method provided by the embodiment of the invention, the position of the human body target in the swimming pool can be timely and accurately detected, and target tracking is performed, so that the behavior of the human body target in the swimming pool can be accurately judged according to the tracking result, and the accuracy of anti-drowning supervision is improved.

Description

technical field [0001] The invention relates to a target tracking method, more specifically to a swimming pool anti-drowning multi-camera target tracking method, device, computer equipment and storage medium. Background technique [0002] With the development of sports, people's enthusiasm for participating in sports is increasing, but swimming, one of the popular sports, has become the sport with the highest incidence of safety accidents. According to incomplete statistics released by the Ministry of Health, about 57,000 people die in drowning accidents in my country every year, of which 56.04% are teenagers, making them the number one cause of death among teenagers, and China's drowning death rate is the highest in the world. [0003] At present, the operation mode of most swimming pools is based on the traditional manual mode of operation, including the supervision of drowning prevention, which is also in the form of lifeguards overlooking the entire swimming pool from a ...

AI Technical Summary

Problems solved by technology

[0003] At present, the operation mode of most swimming pools is based on the traditional manual mode of operation, including the supervision of drowning prevention, which is also in the form of lifeguards overlooking the entire swimming pool from a high position. However, since the main background of the target detection problem in the swimming pool is water, When swimmers are exercising, the water in the swimming pool will fluctuate. Sunlight and lights will produce reflections on the water surface. This reflection is difficult to remove through pretreatment technology. In addition, the position of the reflection will continue to change with the fluctuation of the water surface; the age of the swimmers The span is very wide, and there will be various actions and behaviors on the shore of the pool and people in the pool, and the part of the swimmer below the water surface is affected by the refraction, turbidity and fluctuation of the water, making it difficult to observe; A variety of facilities and sundries, such as stands, life-saving equipment, training equipment, and personal belongings of swimmers, etc. There will also be various floating objects in the swimming pool, such as lane lines and training equipment, and other influencing factors, resulting in the accuracy of manual supervision. The problem is not high, the existing hardware monitoring such as sensor bracelets and external radio frequency devices, etc., the accuracy of this method is not high

Images