Drowning identification method and device, camera and computer system

An identification method and camera technology, applied in computer parts, computing, closed-circuit television systems, etc., can solve the problems of single determination method, easy to cause delay, affecting equipment maintenance costs, etc., and achieve the effect of ensuring the accuracy of identification.

Active Publication Date: 2021-01-01
深圳市云网万店科技有限公司
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The judgment method of this kind of drowning behavior is too simple to fully capture the drowning behavior in the swimming pool;
[0008]Second, most of the image processing units are installed in the back-end server, and after the image is obtained by the camera, it is transmitted to the back-end server for unified processing through the wireless module. The intermediate link of one...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Drowning identification method and device, camera and computer system
  • Drowning identification method and device, camera and computer system
  • Drowning identification method and device, camera and computer system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] In order to use embedded devices to identify drowning, it is necessary to perform model training and build a reasoning framework in advance. The process includes:

[0062] S1. Obtain a training data set;

[0063] Swimming image data of swimmers in various water environment can be collected in advance, and a training data set can be generated according to the image data.

[0064] S2. Using the training data set to train the preset model to obtain the trained preset model;

[0065] The preset models can include detection models such as tiny-yolov3, centernet, mobilenetv2-ssd, inceptionv2-ssd, and Efficientdet-D2. Preferably, the weight value of the model can be updated by using the backpropagation method during training, and the loss function can use the binorm obtained from the recognition result of the model and the real result of the image. The input of the model during the training process is the image in the training data set, and the recognition result includes th...

Embodiment 2

[0092] Corresponding to the above-mentioned embodiments, this application proposes a drowning identification method, such as figure 2 As shown, the method includes:

[0093] 210. Acquire a real-time image captured by a camera, where the real-time image includes at least a first image frame and a preset number of adjacent frames of the first image frame;

[0094] 220. Identify the swimmer included in the real-time image and the swimmer's posture, where the posture includes an upright posture and a non-upright posture;

[0095] 230. When it is recognized that the swimmer is in an upright posture, determine the degree of shape change of the image of the swimmer in the first image frame relative to the image of the swimmer in the adjacent frame;

[0096] Preferably, the identifying the swimmer contained in the real-time image and the swimmer's posture includes:

[0097] 231. Using the preset frame of the camera to identify the swimmer and the swimmer's posture included in the r...

Embodiment 3

[0112] Corresponding to the above method, the present application provides a drowning identification device, such as image 3 As shown, the device includes:

[0113] An acquisition module 310, configured to acquire a real-time image taken by the camera, the real-time image at least including a first image frame and a preset number of adjacent frames of the first image frame;

[0114] An identification module 320, configured to identify the swimmer contained in the real-time image and the swimmer's posture, the posture including an upright posture and a non-upright posture;

[0115] A processing module 330, configured to determine the degree of shape change of the image of the swimmer in the first image frame relative to the image of the swimmer in the adjacent frame when the swimmer is identified as being in an upright posture ;

[0116] The warning module 340 is configured to determine that the swimmer is drowning and issue a warning when the degree of shape change exceeds ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a drowning identification method and device, a camera and a computer system, and the method comprises the steps: obtaining a real-time image photographed by the camera, whereinthe real-time image at least comprises a first image frame and a preset number of adjacent frames of the first image frame; identifying a swimmer and the posture of the swimmer in the real-time image, wherein the posture comprises an upright posture and a non-upright posture; when it is recognized that the swimmer is in the upright posture, determining the shape change degree of an image of the swimmer in the first image frame relative to an image of the swimmer in the adjacent frames; and when the shape change degree exceeds a first preset threshold value, determining that the swimmer is drowned and giving an alarm. The method saves operation required by multi-camera information superposition and hardware cost required for deploying multiple cameras, determines whether a user is drownedor not according to the shape change degree of the user in an upright posture, and guarantees the accuracy of identification.

Description

technical field [0001] The invention relates to the fields of video monitoring and computer vision, in particular to a drowning identification method, device, camera and computer system. Background technique [0002] To keep swimmers safe, swimmers can be monitored to avoid drowning incidents. In the prior art, there are mainly three monitoring and alarm systems for drowning events: [0003] System 1: Swimmers wear detection equipment. The system mainly monitors and calculates the swimmer's heartbeat frequency and residence time under the water surface through devices such as wristbands or headbands worn by the swimmer to determine whether the swimmer is in a drowning state. However, because the equipment is easily damaged, the loss of this system during use is relatively large, and the accuracy rate is not high, which brings a lot of inconvenience to the swimmers themselves, so the utilization rate is not high. [0004] System 2: Use sonar to detect whether swimmers are ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): H04N7/18G08B21/08G06K9/00
CPCH04N7/18G08B21/08G06V40/20
Inventor 荆伟贺稳定孙皓代利坚许立群
Owner 深圳市云网万店科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap