Water falling detection and rescue control system based on deep learning

Abstract

The invention discloses a water falling detection and rescue control system based on deep learning, belonging to the field of computer vision, and the method comprises the steps: building an underwater personnel image data set; using a LabelImg tool to mark the position of a person in water and the category of the person in water in the original image, and storing the marking information generatedby each image in an xml file format for network training; using a YOLOv2 deep learning target detection framework i for training a data set, making clustering analysis on the established data set before network training, and dividing the data set into a training set, a verification set and a test set according to the proportion of 6: 2: 2. The rescue control system comprises a monitoring video input unit, a drowning person detection unit and an alarm rescue unit. The response efficiency of water surface rescue is effectively improved, and the survival probability of people falling into wateris improved.

Description

technical field [0001] This patent relates to the field of computer vision, in particular to a deep-learning-based falling-water detection and rescue control system in a monitoring environment. Background technique [0002] The large number of park artificial lakes, reservoirs and lakes have weak patrols, safety measures are not in place, and professional water emergency rescue equipment is lacking, especially in summer, resulting in frequent drowning accidents. In order to avoid tragedies, some scenes have realized the all-weather monitoring of the state of the water area, but someone still needs to watch the monitoring screen, which is not only time-consuming and laborious, but also cannot guarantee real-time rescue of the drowning personnel. Real-time automatic monitoring and improving emergency response capabilities have become particularly important. [0003] At present, the main reference for the detection of people falling into the water is the human body detection m...

AI Technical Summary

Problems solved by technology

[0002] The large number of park artificial lakes, reservoirs and lakes have weak patrols, safety measures are not in place, and there is a lack of professional water emergency rescue equipment, especially in summer, resulting in frequent drowning accidents

In order to avoid tragedies, some scenes have achieved round-the-clock monitoring of the state of the water area, but someone still needs to monitor the monitoring screen. This is not only time-consuming and laborious, but also cannot guarantee real-time rescue of the drowning personnel. Real-time automatic monitoring and improving emergency response capabilities become particularly important

[0003] At present, the main reference for the detection of people falling into the water is the ground human body detection method. The traditional image processing-based detection of human body falling into the water faces problems such as complex water background, easy to be affected by light, difficult to distinguish the reflection from the real situation of falling into the water, and differences in the situation of falling into the water.

Images