CN111368743B - Swimming pool deepwater area early dangerous behavior detection method based on monitoring video - Google Patents

Swimming pool deepwater area early dangerous behavior detection method based on monitoring video Download PDF

Info

Publication number
CN111368743B
CN111368743B CN202010148596.5A CN202010148596A CN111368743B CN 111368743 B CN111368743 B CN 111368743B CN 202010148596 A CN202010148596 A CN 202010148596A CN 111368743 B CN111368743 B CN 111368743B
Authority
CN
China
Prior art keywords
swimmer
frame
head
detection
frames
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010148596.5A
Other languages
Chinese (zh)
Other versions
CN111368743A (en
Inventor
王璠
张卫冬
艾轶博
张英杰
王月
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Science and Technology Beijing USTB
Original Assignee
University of Science and Technology Beijing USTB
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Science and Technology Beijing USTB filed Critical University of Science and Technology Beijing USTB
Priority to CN202010148596.5A priority Critical patent/CN111368743B/en
Publication of CN111368743A publication Critical patent/CN111368743A/en
Application granted granted Critical
Publication of CN111368743B publication Critical patent/CN111368743B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/08—Alarms for ensuring the safety of persons responsive to the presence of persons in a body of water, e.g. a swimming pool; responsive to an abnormal condition of a body of water
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
    • Y02A40/81—Aquaculture, e.g. of fish

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Human Computer Interaction (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Psychiatry (AREA)
  • Social Psychology (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a swimming pool deepwater zone early dangerous behavior detection method based on a monitoring video, which can realize swimming pool deepwater zone early dangerous behavior detection and alarm by using the monitoring video. The method comprises the following steps: acquiring videos of a swimmer in a swimming pool in real time, and detecting the head of the swimmer in the videos to obtain a swimmer head detection frame; judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame; if the swimming state is not in the normal swimming state, the head-body ratio of the swimmer is used for judging whether the swimmer is in the standing state, and if the swimmer is in the standing state, an alarm is given. The present invention relates to the field of monitoring.

Description

Swimming pool deepwater area early dangerous behavior detection method based on monitoring video
Technical Field
The invention relates to the field of monitoring, in particular to a method for detecting early dangerous behaviors in a deep water area of a natatorium based on a monitoring video.
Background
With the enhancement of fitness consciousness of people, more and more people begin to select swimming, and the number of indoor swimming pools is increased. Although many countries have strict requirements for swimming pools, such as a certain number of life-saving workers, the gradient and depth of swimming pools have strict standards, the situation of drowning accidental death in indoor swimming pools often happens. Many drowners do not have obvious struggling and help seeking actions when drowned, so that the life-saving personnel do not find the drowned person and miss the best rescue time. In indoor swimming pools, swimming pools are generally divided into two types: deep water and shallow water. The deep water area has deeper water depth, and a swimmer cannot stand in water, and can only keep a state (treading water) that the legs are straightened, and only the head part is exposed out of the water surface (the shoulders are few) when the water is treaded on; the shallow water is shallow, the swimmer can directly stand in the swimming pool, the general number of the water playing people is more, the swimmer can not necessarily swim, the shoulder and back can be exposed out of the water surface when standing, and the data of the deep water area and the shallow water area have larger difference.
The main method for judging whether the swimmer has drowning or not is judged according to the rescue experience of the swimmer, and the main characteristics of the swimmer are that the swimmer has a loud help calling and struggling, but the swimmer can open eyes, stay in eyes, stand in water or be in a vertical ladder shape in the water, and the two arms extend to two sides or forward, as shown in film and television drama. Because most drowners cannot call for help and cannot wave hands, when the number of swimming people in the swimming pool is too large, the life-saving personnel may not find early drowners in time, and when the drowners sink, the life-saving personnel belong to the later stage of drowning, and most of the best opportunity for rescue is likely to be missed.
The lifeguard cannot pay attention to each swimmer at any time, but the monitoring video can do so. Along with the development of intelligent monitoring and video understanding, monitoring videos can be adopted to detect and track targets and judge complex behaviors, so that monitoring videos in natatorium are used for assisting a life fighter in early-stage drowning dangerous behavior judgment, the best rescue opportunity is grasped, and the method has important significance for reducing accidents.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method for detecting early dangerous behaviors in a swimming pool deepwater area based on a monitoring video, which aims to solve the problem of how to use the monitoring video to detect and alarm the early dangerous behaviors in the swimming pool deepwater area in the prior art.
In order to solve the technical problems, an embodiment of the present invention provides a method for detecting early dangerous behavior in a deep water area of a natatorium based on a surveillance video, including:
acquiring videos of a swimmer in a swimming pool in real time, and detecting the head of the swimmer in the videos to obtain a swimmer head detection frame;
judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame;
if the swimming state is not in the normal swimming state, the head-body ratio of the swimmer is used for judging whether the swimmer is in the standing state, and if the swimmer is in the standing state, an alarm is given.
Further, before capturing video of the swimmer, detecting the head of the swimmer according to the captured video to obtain a detection frame of the head of the swimmer, the method further includes:
acquiring pictures of a swimmer in a swimming pool to form a training set, wherein the head of the swimmer in the pictures is marked;
training the yolov3 deep learning network by using the acquired training set to obtain a swimmer head detection module, wherein the swimmer head detection module is used for detecting the swimmer head to obtain a swimmer head detection frame.
Further, the swimmer head detection of the swimmer in the video further includes:
and if the head detection frame is not detected within the preset time period, alarming.
Further, the step of judging whether the swimmer is in the normal swimming state according to the time sequence position change of the swimmer head detection frame comprises the following steps:
according to the swimming condition of the target swimmer in the first n frames of the video, determining an initial interval frame number t for judging the cross ratio of the areas of the detection frames;
in the images after n frames, judging whether the area cross ratio value of the detection frames in the 2 images of the interval t frames is in a preset normal range, if so, the detection frames are in a normal swimming state.
Further, determining the initial interval frame number t for judging the cross-over ratio of the detection frame areas according to the swimming condition of the target swimmer in the first n frames of the video comprises:
s1, acquiring the first n frames of images of a target swimmer video;
s2, determining a first frame image f and a 0 th frame image f according to the acquired previous n frame images of the target swimmer video 0 Image without overlapping detection frameThen t 0 Frame image->Corresponding interval frame number t=t 0 ；
S3, respectively calculating the t 0 +1 frame imageDetection frame and frame 0 image f 0 Frame 1 image f 1 Frame 2 image f 2 Area cross ratio of detection frame->Taking the frame corresponding to the detection frame with the minimum cross-over ratio value to update the t 0 +1 frame image +.>Corresponding interval frame number->Wherein i=t 0 +1-t-1，t 0 +1-t，t 0 +1-t+1；
S4, updating according to the operation of S3Next frame image +.>Corresponding interval frame numberWherein i=t 0 +2-t-1，t 0 +2-t，t 0 +2-t+1, and sequentially updated according to the operation of S3Corresponding to the number of interval frames, the image f n The corresponding interval frame number is used as an initial interval frame number t for judging the cross-over ratio of the areas of the detection frames;
wherein the image f n Corresponding interval frame numberWhere i=n-t-1, n-t, n-t+1.
Further, the determining whether the area cross ratio value of the detection frames in the 2 images of the interval t frame is within a preset normal range includes:
determining an a-th frame image f according to the determined initial interval frame number t a And the a-t frame image f a-t Whether the cross-ratio value of the detection frame areas is in a preset normal range, if so, the detection frame areas are in a normal swimming state and pass through the formulai=a-t-1, a-t, a-t+1 correct f a Frame-time interval frame number, where a>n。
Further, the step of judging whether the swimmer is in an upright state by using the head-to-body ratio of the swimmer, and if the swimmer is in the upright state, the step of alarming comprises:
judging whether the head-body ratio of the swimmer exceeds a preset alarm threshold value, if so, judging that the swimmer is in an upright state, and giving an alarm.
The technical scheme of the invention has the following beneficial effects:
in the scheme, the collected monitoring video of the swimmer in the swimming pool is used for detecting the head of the swimmer in the video to obtain a swimmer head detection frame; judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame; if the swimming device is not in the normal swimming state, the head-body ratio of the swimmer is used for judging whether the swimmer is in the dangerous standing state, and if the swimmer is in the standing state, the swimmer is alarmed. Therefore, the early dangerous behavior of the swimmer in the swimming pool is judged, so that the detection and alarm of the early dangerous behavior of the deep water area of the swimming pool are realized, and reliable rescue information is provided for the life-saving personnel.
Drawings
Fig. 1 is a schematic flow chart of a method for detecting early dangerous behavior in a deep water area of a natatorium based on a surveillance video according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a swimmer head detection result provided by an embodiment of the present invention;
fig. 3 is a schematic diagram of updating frame interval number of the first 100 frames according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a variation curve of a cross-over ratio of a detection frame corresponding to a corresponding frame interval number in the first 100 frames according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages to be solved more apparent, the following detailed description will be given with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, the method for detecting early dangerous behavior in a deep water area of a natatorium based on a surveillance video provided by the embodiment of the invention includes:
s101, acquiring videos of a swimmer in a swimming pool in real time, and detecting the head of the swimmer in the videos to obtain a swimmer head detection frame;
s102, judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame;
s103, if the swimming device is not in the normal swimming state, judging whether the swimming device is in the standing state or not by using the head-body ratio of the swimming device, and if the swimming device is in the standing state, giving an alarm.
According to the swimming pool deep water area early dangerous behavior detection method based on the monitoring video, the head of a swimmer in the video is detected through the collected monitoring video of the swimmer in the swimming pool, and a swimmer head detection frame is obtained; judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame; if the swimming device is not in the normal swimming state, the head-body ratio of the swimmer is used for judging whether the swimmer is in the dangerous standing state, and if the swimmer is in the standing state, the swimmer is alarmed. Therefore, the early dangerous behavior of the swimmer in the swimming pool is judged, so that the detection and alarm of the early dangerous behavior of the deep water area of the swimming pool are realized, and reliable rescue information is provided for the life-saving personnel.
In a specific embodiment of the foregoing method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video, further, before capturing video of a swimmer, detecting a head of the swimmer according to the captured video, and obtaining a detection frame of the head of the swimmer, the method further includes:
acquiring pictures of a swimmer in a swimming pool to form a training set, wherein the head of the swimmer in the pictures is marked;
training the yolov3 deep learning network by using the acquired training set to obtain a swimmer head detection module, wherein the swimmer head detection module is used for detecting the swimmer head to obtain a swimmer head detection frame.
In this embodiment, capturing a video of a swimmer in a swimming pool through a camera device, sampling the captured video according to Z (e.g., 25) frames per second, labeling the head of the swimmer in a picture obtained by sampling by adopting a rectangular frame, dividing the labeled image into a training set and a testing set according to a preset proportion, inputting the picture in the training set into a yolov3 deep learning network, and training the yolov3 deep learning network to obtain a swimmer head detection module, for example, a detection result is shown in fig. 2; and testing the obtained swimmer head detection module by using the image in the test set.
In a specific embodiment of the foregoing method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video, further, the detecting a swimmer head of a swimmer in the video further includes:
if the head detection frame is not detected within a preset period of time (e.g., 3 minutes), an alarm is given.
In a specific embodiment of the foregoing method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video, further, the determining whether the swimmer is in a normal swimming state according to a time sequence position change of a head detection frame of the swimmer includes:
according to the swimming condition of the target swimmer in the first n frames of the video, determining an initial interval frame number t for judging the cross ratio of the areas of the detection frames;
in the images after n frames, judging whether the area cross ratio value of the detection frames in the 2 images of the interval t frames is in a preset normal range, if so, the detection frames are in a normal swimming state.
In a specific embodiment of the foregoing method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video, further, determining, according to a swimming condition of a target swimmer in the previous n frames of video, an initial interval frame number t for determining an area intersection ratio of detection frames includes:
s1, acquiring the first n frames of images of a target swimmer video;
s2, determining a first frame image f and a 0 th frame image f according to the acquired previous n frame images of the target swimmer video 0 Image without overlapping detection frameThen t 0 Frame image->Corresponding interval frame number t=t 0 ；
S3, respectively calculating the t 0 +1 frame imageDetection frame and frame 0 image f 0 Frame 1 image f 1 Frame 2 image f 2 Area cross ratio of detection frame->Taking the cross-sum to be more than the frame corresponding to the detection frame with the minimum comparison valueNew t 0 +1 frame image +.>Corresponding interval frame number->Wherein i=t 0 +1-t-1，t 0 +1-t，t 0 +1-t+1，
S4, updating according to the operation of S3Next frame image +.>Corresponding interval frame numberWherein i=t 0 +2-t-1，t 0 +2-t，t 0 +2-t+1, and sequentially updated according to the operation of S3Corresponding to the number of interval frames, the image f n The corresponding interval frame number is used as an initial interval frame number t for judging the cross-over ratio of the areas of the detection frames;
wherein the image f n Corresponding interval frame numberWhere i=n-t-1, n-t, n-t+1.
In this embodiment, for example, n=100, in practical application, the value of n may be determined according to the practical application scenario, so that the initial interval frame number t of the area overlapping ratio of the detection frames to be determined later is determined according to the swimming condition of the previous 100 frames of the target swimmer, and the results of updating and determining the frame interval number of the previous 100 frames are shown in fig. 3. As can be seen from fig. 3, in the 11 th frame, a detection frame completely non-overlapping with the first frame is found, then the number of frame intervals is updated, and finally the initial number of frame intervals is determined to be 11, and the previous 100 frames intersect with the adjacent t frame detection frames according to the corresponding number of frame intervals, as shown in fig. 4, and the ratio is between-0.05 and 0.05.
In a specific embodiment of the foregoing method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video, further, the determining whether the area intersection ratio value of the detection frames in the 2 images of the interval t frames is within a preset normal range includes:
determining an a-th frame image f according to the determined initial interval frame number t a And the a-t frame image f a-t Whether the cross-ratio value of the detection frame areas is in a preset normal range, if so, the detection frame areas are in a normal swimming state and pass through the formulai=a-t-1, a-t, a-t+1 correct f a Frame-time interval frame number, where a>n; otherwise, a pre-alarm is performed, and S103 is performed.
In this embodiment, the interval frame number is corrected in the process of judging the normal running speed, so that erroneous judgment caused by a large difference between the current running speed and the running speed of the previous n frames can be avoided.
In this embodiment, the swimming state is divided into a water-treaded state and an upright state, wherein the upright state is a dangerous state. The head-body ratio of normal people is about 1:6.5-1:7.5, and the head-body ratio in water is slightly different due to the angle of the camera and the refraction of water. In the water treading state, the fully straightened state of the two legs is only instant, so that when water treading, the head-body ratio of a swimmer is obviously different from about 1:2-1:3 due to the leg curling state of the swimmer.
In this embodiment, a video of 20 seconds of the swimmer is collected, the criterion of normal swimming of the swimmer is performed by using the cross-over ratio of the area of the detection frame, if the cross-over ratio of the area of the detection frame is not within the preset normal range in 16 seconds, the abnormal change of the swimming speed is indicated, the pre-alarm is performed, and S103 is executed to determine whether the swimmer is in the upright state.
In this embodiment, if the average value of the head-body ratio is 1:2.4664 in 16-18 seconds, the state of treading on water is determined, and if the head-body ratio exceeds the set alarm threshold for 18-20 seconds, the state of standing is determined, and an alarm is given.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.

Claims (4)

1. The method for detecting the early dangerous behavior of the deep water area of the natatorium based on the monitoring video is characterized by comprising the following steps of:
acquiring videos of a swimmer in a swimming pool in real time, and detecting the head of the swimmer in the videos to obtain a swimmer head detection frame;
judging whether the swimmer is in a normal swimming state according to the time sequence position change of the swimmer head detection frame;
if the swimming state is not in the normal swimming state, judging whether the swimmer is in an upright state or not by utilizing the head-body ratio of the swimmer, and if the swimmer is in the upright state, giving an alarm;
wherein, the step of judging whether the swimmer is in a normal swimming state according to the time sequence position change of the head detection frame of the swimmer comprises the following steps:
according to the swimming condition of the target swimmer in the first n frames of the video, determining an initial interval frame number t for judging the cross ratio of the areas of the detection frames;
in the images after n frames, judging whether the area intersection ratio value of the detection frames in the 2 images of the interval t frames is in a preset normal range, if so, the detection frames are in a normal swimming state;
wherein, according to the swimming condition of the target swimmer in the previous n frames of the video, determining the initial interval frame number t for judging the cross ratio of the areas of the detection frames comprises:
s1, acquiring the first n frames of images of a target swimmer video;
s2, determining a first frame image f and a 0 th frame image f according to the acquired previous n frame images of the target swimmer video 0 Image without overlapping detection frameThen t 0 Frame image->Corresponding interval frame number t=t 0 ；
S3, respectively calculating the t 0 +1 frame imageDetection frame and frame 0 image f 0 Frame 1 image f 1 Frame 2 image f 2 Area cross ratio of detection frame->Taking the frame corresponding to the detection frame with the minimum cross-over ratio value to update the t 0 +1 frame image f t0+1 Corresponding interval frame number->Wherein i=t 0 +1-t-1，t 0 +1-t，t 0 +1-t+1；
S4, updating according to the operation of S3Next frame image +.>Corresponding interval frame number->Wherein i=t 0 +2-t-1，t 0 +2-t，t 0 +2-t+1And updates +.>Corresponding to the number of interval frames, the image f n The corresponding interval frame number is used as an initial interval frame number t for judging the cross-over ratio of the areas of the detection frames;
wherein the image f n Corresponding interval frame numberWherein i=n-t-1, n-t, n-t+1;
the determining whether the area cross ratio value of the detection frames in the 2 images of the interval t frames is within a preset normal range includes:
determining an a-th frame image f according to the determined initial interval frame number t a And the a-t frame image f a-t Whether the cross-ratio value of the detection frame areas is in a preset normal range, if so, the detection frame areas are in a normal swimming state and pass through the formulai=a-t-1, a-t, a-t+1 correct f a Frame-time interval frame number, where a>n。
2. The method for detecting early dangerous behavior in a natatorium deepwater zone based on surveillance video of claim 1, wherein before capturing video of a swimmer, detecting the swimmer's head according to the captured video, the method further comprises:
acquiring pictures of a swimmer in a swimming pool to form a training set, wherein the head of the swimmer in the pictures is marked;
training the yolov3 deep learning network by using the acquired training set to obtain a swimmer head detection module, wherein the swimmer head detection module is used for detecting the swimmer head to obtain a swimmer head detection frame.
3. The surveillance video-based natatorium deepwater zone early dangerous behavior detection method of claim 1, further comprising the step of detecting a swimmer's head of a swimmer in a video:
and if the head detection frame is not detected within the preset time period, alarming.
4. The method for detecting early dangerous behavior in a deep water area of a natatorium based on surveillance video according to claim 1, wherein the step of using the head-to-body ratio of the swimmer to determine whether the swimmer is in an upright position, and if so, the step of alarming comprises:
judging whether the head-body ratio of the swimmer exceeds a preset alarm threshold value, if so, judging that the swimmer is in an upright state, and giving an alarm.
CN202010148596.5A 2020-03-05 2020-03-05 Swimming pool deepwater area early dangerous behavior detection method based on monitoring video Active CN111368743B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010148596.5A CN111368743B (en) 2020-03-05 2020-03-05 Swimming pool deepwater area early dangerous behavior detection method based on monitoring video

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010148596.5A CN111368743B (en) 2020-03-05 2020-03-05 Swimming pool deepwater area early dangerous behavior detection method based on monitoring video

Publications (2)

Publication Number Publication Date
CN111368743A CN111368743A (en) 2020-07-03
CN111368743B true CN111368743B (en) 2023-08-04

Family

ID=71207017

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010148596.5A Active CN111368743B (en) 2020-03-05 2020-03-05 Swimming pool deepwater area early dangerous behavior detection method based on monitoring video

Country Status (1)

Country Link
CN (1) CN111368743B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112165600B (en) * 2020-08-26 2023-02-03 深圳市云网万店科技有限公司 Drowning identification method and device, camera and computer system
CN112309077A (en) * 2020-09-23 2021-02-02 苏宁云计算有限公司 Drowning alarm method and drowning alarm device for swimming pool

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019237536A1 (en) * 2018-06-11 2019-12-19 平安科技（深圳）有限公司 Target real-time tracking method and apparatus, and computer device and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10803724B2 (en) * 2011-04-19 2020-10-13 Innovation By Imagination LLC System, device, and method of detecting dangerous situations
CN106022230B (en) * 2016-05-11 2019-05-14 太原理工大学 A kind of swimming pool drowning event detecting method based on video
CN110084223A (en) * 2019-05-10 2019-08-02 南京硅基智能科技有限公司 A kind of monitoring method and system for swimming pool
CN110569772B (en) * 2019-08-30 2022-03-08 北京科技大学 Method for detecting state of personnel in swimming pool

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019237536A1 (en) * 2018-06-11 2019-12-19 平安科技（深圳）有限公司 Target real-time tracking method and apparatus, and computer device and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于回归深度卷积网络的船舶图像与视频检测;黄志坚;张成;王慰慈;;上海海事大学学报(第04期);全文 *

Also Published As

Publication number Publication date
CN111368743A (en) 2020-07-03

Similar Documents

Publication Publication Date Title
CN112489371B (en) Swimming pool drowning prevention early warning system based on computer vision
CN111368743B (en) Swimming pool deepwater area early dangerous behavior detection method based on monitoring video
CN110569772B (en) Method for detecting state of personnel in swimming pool
Tao et al. Fall incidents detection for intelligent video surveillance
CN107566797A (en) A kind of drowned monitor and detection device of swimming pool
CN111310556A (en) Drowning prevention safety supervision system based on primary and middle school student area and monitoring method thereof
CN110446015A (en) A kind of abnormal behaviour monitoring method based on computer vision and system
CN108647575A (en) Drowned method for early warning based on optical visual analysis
CN107622505A (en) A kind of drowned monitor and detection method of swimming pool
CN110298837B (en) Method for detecting fire-fighting road occupation abnormal object based on interframe difference method
Alshbatat et al. Automated vision-based surveillance system to detect drowning incidents in swimming pools
CN106165391A (en) The image capturing strengthened
CN104347022B (en) Method and system for detecting failure of spliced screen
CN112990057A (en) Human body posture recognition method and device and electronic equipment
US20200053320A1 (en) Drowning Detection Enhanced by Swimmer Analytics
US20210174073A1 (en) A Method and Apparatus for Swimmer Tracking
CN108737785B (en) Indoor automatic detection system that tumbles based on TOF 3D camera
CN108492517A (en) A kind of fire monitoring system for building
CN113158962A (en) Swimming pool drowning detection method based on YOLOv4
CN113408479A (en) Flame detection method and device, computer equipment and storage medium
CN113438443A (en) Emergency place monitoring and early warning system with image feature recognition function
Zhang et al. A novel camera-based drowning detection algorithm
TWI493510B (en) Falling down detection method
US11769387B2 (en) Method and apparatus for detecting drowning
Wang et al. Detection of early dangerous state in deep water of indoor swimming pool based on surveillance video

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant