CN115131940A

CN115131940A - Drowning prevention system

Info

Publication number: CN115131940A
Application number: CN202211053173.0A
Authority: CN
Inventors: 徐峰
Original assignee: Nanchang Cloud Guardian Technology Co ltd
Current assignee: Nanchang Cloud Guardian Technology Co ltd
Priority date: 2022-08-30
Filing date: 2022-08-30
Publication date: 2022-09-30

Abstract

The invention relates to a drowning prevention system, which acquires images and photos around a monitored water area through a camera to obtain a face image set, a face recognition module carries out recognition comparison on the face image set and a target face image set, judges the coincidence degree of facial features of the face image set and the target face image set, automatically sends out warning information, judges whether the face image set is continuously acquired or not, sends out artificial shouting warning information, continuously judges whether the face image set is acquired or not, sends out information pushing, continuously judges whether the face image set is acquired or not, sends out a time limit worksheet, a storage module stores the image and photo information and the target face image set, carries out timely rescue and rescue on drowning personnel through an emergency rescue device when a drowning accident occurs, and realizes early prevention and timely drowning occurrence, the timeliness of drowning early warning and rescue is improved.

Description

Prevent drowned system

Technical Field

The invention relates to the field of data processing, in particular to a drowning prevention system.

Background

Drowning accidents occur frequently in hot summer, and particularly, the drowning accident rate of immature students is high, so that the life safety of the immature students is seriously threatened, and therefore the drowning accidents need to be early warned in time and rescued in time after the drowning accidents occur.

The patent document with the publication number of CN110155274B discloses an underwater intelligent lifesaving system for a swimming pool, which comprises an underwater camera, an intelligent lifesaving system, a workstation, a handheld field display terminal and the like; an underwater camera is arranged in the swimming pool, the information such as images and data of all swimming people in the swimming pool is collected, drowning persons appearing underwater are identified and found through a computer artificial intelligence technology, a visual identification technology, a behavior identification technology and a deep learning technology, and meanwhile, alarm information is sent out to help life preservers to rescue the drowning persons in the shortest time.

The image of swimming crowd is caught through camera under water and through the drowned person of artificial intelligence technical identification to prior art, carries out timely rescue to the drowned person, and the distance that nevertheless camera gathered the image under water is limited, and just sends alarm signal after the drowned condition takes place, has the hysteresis quality of information acquisition processing, leads to there being the hysteresis quality in the early warning of drowning and the aspect of drowned rescue.

Disclosure of Invention

Therefore, the drowning prevention system provided by the invention can solve the problem of hysteresis of drowning early warning and drowning rescue.

In order to achieve the above object, the present invention provides a drowning prevention system, comprising:

the camera is connected with the server and used for monitoring the monitored water area and acquiring images and photos around the monitored water area to obtain image and photo information;

the alarm device is arranged at a position 3 meters away from the land junction of the monitoring water area, is connected with the server and is used for sending automatic alarm information and artificial shouting alarm information;

the server is connected with the camera, the alarm device, the emergency rescue device and the display screen and used for receiving, storing and analyzing the image and photo information;

the emergency rescue device is connected with the server and used for emergently rescuing drowning accidents in a monitored water area;

the display screen is connected with the server and used for displaying the real-time video dynamic information of the camera;

the server comprises a face recognition module, a voice recognition module, an action recognition module, a communication module, a work order issuing module, a data storage module, a data analysis module, an electronic patrol module and a management module, wherein the face recognition module is used for recognizing the image photo information to obtain a face image set and performing recognition comparison on the face image set and a target face image set to obtain recognition comparison information, the voice recognition module is used for recognizing human voice in a monitoring water area, the action recognition module is used for recognizing human limb actions in the monitoring water area, the communication module is used for communicating with a background manager and parents of students, the work order issuing module is used for issuing a time-limited work order to a gridder in charge of monitoring the water area in time, and the data storage module is used for storing the image photo information and the target face image set, the data analysis module is used for analyzing and monitoring the geographical position that the waters got into student's number of people, time quantum, age bracket and monitoring area, the electron module of patrolling is used for setting up the patrol cycle and showing the patrol suggestion, the management module is used for managing project, personnel, equipment, label, sign in, warning and account number.

Furthermore, the face recognition module is used for recognizing the image photo information to obtain a face image set, and performing recognition comparison on the face image set and a target face image set to obtain recognition comparison information, the recognition comparison information comprises first recognition comparison information and second recognition comparison information, the face image set only comprises a front face of a face, the calculation processing pressure of a drowning prevention system is reduced, the drowning prevention system only acquires the face image set of a complete face image in a mode that at least 3 cameras are arranged in the same monitoring water area, the recognition comparison information comprises eye characteristic information, mouth characteristic information, nose characteristic information and face proportion characteristic information,

if at least MxN eye feature information in the N eye feature information extracted from the face image set coincides with the eye feature information of the target face image set, the eye feature information of the face image set coincides with the eye feature information of the target face image set,

if at least MxN pieces of mouth feature information in the N pieces of mouth feature information extracted from the face image set coincide with the mouth feature information of the target face image set, the mouth feature information of the face image set coincides with the mouth feature information of the target face image set,

if at least MxN nose feature information in the N pieces of nose feature information extracted from the face image set is overlapped with the nose feature information of the target face image set, the nose feature information of the face image set is consistent with the nose feature information of the target face image set,

if the error between the facial proportion feature information extracted from the facial image set and the facial proportion feature information of the target facial image set is less than or equal to 10 percent, the facial proportion feature information of the facial image set is consistent with the facial proportion feature information of the target facial image set,

the first identification contrast information is that the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the face image set are not satisfied to be consistent with at least 3 characteristic information in the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the target face image set,

the second identification contrast information is that the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the face image set meet the requirement of being consistent with at least 3 characteristic information in the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the target face image set.

Furthermore, the alarm device comprises an audible and visual alarm and an artificial voice alarm device, the audible and visual alarm is connected with the server and the emergency aid device and used for sending out automatic alarm information, and the artificial voice alarm device is connected with the server and used for sending out artificial shouting alarm information to the monitored water area;

emergency rescue device includes a key alarm device and automatic rescue device, a key alarm device with the server is connected with the room of being on duty of dispatching for in time to sending alarm information and starting to the room of being on duty of dispatching audible-visual annunciator carries out audible-visual annunciator, automatic rescue device is connected with the server for throw life-saving appliance to drowned area.

Further, the audible and visual alarm sends out automatic warning information according to the identification contrast information, the server obtains first feedback information, the communication module carries out information pushing according to the first feedback information, the artificial voice warning device sends out artificial shouting warning information, the server obtains second feedback information, the communication module carries out information pushing according to the second feedback information and dials a video call for a target telephone, the server obtains third feedback information, the communication module carries out information pushing according to the third feedback information, the work order issuing module issues a time-limited work order and obtains fourth feedback information,

and the data storage module stores the fourth feedback information.

Further, the audible and visual alarm sends out automatic warning information according to the identification comparison information, the server obtains first feedback information, the automatic warning information comprises first automatic warning information and second automatic warning information,

if the identification contrast information is first identification contrast information, the audible and visual alarm sends out first automatic warning information,

if the identification contrast information is second identification contrast information, the audible and visual alarm sends out second automatic warning information,

the first automatic warning information is field information which is edited by a system in advance, and the second automatic warning information is student names and field information which is edited by the system in advance.

Further, the first feedback information includes first positive feedback information and first negative feedback information,

the first positive feedback information is that the face recognition module cannot recognize the face image set after 10 seconds after the audible and visual alarm sends the automatic warning information, and the first negative feedback information is that the face recognition module can recognize the face image set after 10 seconds after the audible and visual alarm sends the automatic warning information.

Furthermore, the communication module comprises a first pushing unit, a second pushing unit and a third pushing unit, the first pushing unit is used for pushing first pushing information to the background administrator according to the identification comparison information,

if the identification contrast information is first identification contrast information, the first push information comprises the positioning of a monitoring water area, a captured face picture and time,

if the identification contrast information is second identification contrast information, the first push information comprises the positioning of a monitoring water area, the captured face picture, the time, the school, the class and the name of a student,

the artificial voice warning device is used for sending out artificial shouting warning information according to the first pushing information, the server obtains second feedback information, the second feedback information comprises second positive feedback information and second negative feedback information, the second positive feedback information is that the human face recognition module cannot recognize the human face image set after the artificial voice warning device sends out the artificial shouting warning information for 10 seconds, the second negative feedback information is that the human face recognition module can still recognize the human face image set after the artificial voice warning device sends out the artificial shouting warning information for 10 seconds,

the second pushing unit is configured to push second pushing information to a target telephone number according to the second feedback information, the server obtains third feedback information, where the third feedback information includes third positive feedback information and third negative feedback information, the third positive feedback information is that the face recognition module cannot recognize the face image set after the video telephone is interrupted for 10 seconds, the third negative feedback information is that the face recognition module can still recognize the face image set after the video telephone is interrupted for 10 seconds,

the work order issuing module is used for issuing a time-limited work order according to the third feedback information, the server obtains fourth feedback information, the fourth feedback information is that the face recognition module cannot recognize the face image set after 3 minutes of time-limited work order issuing by the work order issuing module,

the third pushing unit is used for pushing the image photo information to a first target telephone number and a second target telephone number according to the fourth feedback information, the holder of the first target telephone number is a parent of a student corresponding to the face image set, and the holder of the second target telephone number is a school of the student corresponding to the face image set.

Furthermore, the data storage module is used for storing the image photo information and the target face image set, the storage module also stores thematic education information, the thematic education information is drowning prevention thematic education information of a school for students collecting face identification information in a monitoring water area, the target face image set is stored in a picture uploading mode,

the management module comprises an item management unit, a personnel management unit, an equipment management unit, a label management unit, a check-in management unit, a reminding management unit and an account management unit, wherein the item management unit is used for managing item information including adding, modifying and deleting the item information, the items are school names and area names, the personnel management unit is used for managing the target face image set and includes adding, deleting, retrieving and presenting identification records of the target face image set, the equipment management unit is used for adding, deleting and inquiring the sound equipment, the label management unit is used for marking and inquiring students corresponding to the face image set, the check-in management unit is used for checking-in management of attendance personnel, including adding, retrieving and checking, the reminding management unit is used for reminding the background manager when the equipment is abnormal, and the account management unit is used for carrying out addition, inquiry and editing management on the account information of the drowning prevention system.

Furthermore, the automatic rescue device judges and rescues drowning accidents in a monitored water area according to the voice recognition module and the action recognition module,

the voice recognition module adopts a long voice recognition technology and is used for optimizing a video scene, recognizing voice call-for-help content in a video and judging whether a person in a water area drowns or not, the optimization is to reduce the noise of the collected voice to obtain clear human language vocabulary information, the noise reduction is to collect at least one splash voice of the water surface in advance for the drowning prevention system, convert the splash voice into a splash character sequence, convert all voice information into a total character sequence after the drowning prevention system collects all voice information on site, deduct a character sequence with the coincidence ratio of the total character sequence and the splash character sequence more than or equal to 85 percent, and read the residual character sequence of the total character sequence after deducting,

and if the overlap ratio of the residual character sequence of the total character sequence after deduction and a preset first target character sequence is more than or equal to 80%, determining that drowning occurs to people in the water area.

Furthermore, the action recognition module is used for accurately recognizing the limb information of the person in the image, tracking, positioning and identifying the attribute of the person, judging whether the person in the water area drowns or not,

the positioning is to determine at least one propagation center of the water surface wave through the shape of the water surface wave and the propagation speed of the water surface wave by an AI algorithm, the attribute is identified to judge whether the person in the water area is swimming normally or drowning, the human body action character sequence of the person in the water area is judged by setting a first target human body action character sequence, if the coincidence degree of the human body action character sequence of the person in the water area and the first target human body action character sequence is more than or equal to 80 percent, the drowning of the person in the water area is judged,

the voice recognition module and the action recognition module simultaneously judge that a person in the water area drowns, the server sends an instruction to a throwing device on the automatic rescue device, the throwing position of the throwing device is within 1 meter range of the propagation center of at least one water surface wave, the throwing device throws at least 2 life buoys or at least 2 life ropes or at least 2 life poles, the selection of life saving equipment is judged according to the distance D between the throwing position and the throwing device, if D is more than 0 and less than 5 meters, the throwing device throws at least 2 life buoys to the throwing position, if D is more than or equal to 5 and less than 8 meters, the throwing device throws at least 2 life buoys to the throwing position, and if D is more than or equal to 8 meters, the throwing device throws at least 2 life buoys to the throwing position,

after the life-saving equipment is thrown to the throwing position by the throwing device, the action recognition module recognizes the limb action of drowning people, judging the collected human body action character sequence of the personnel in the water area by setting a second target human body action character sequence, if the contact ratio of the collected human body action character sequence of the personnel in the water area and the second target human body action character sequence is more than or equal to 80 percent, judging that the personnel in the water area take the lifesaving equipment, if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the second target human body action character sequence is less than 20 percent, and if the person in the water area is judged not to take the thrown lifesaving equipment, the server continuously sends an instruction to a throwing device on the automatic rescue device, and the throwing device continuously throws the lifesaving equipment to the throwing position.

Compared with the prior art, the invention has the advantages that the camera acquires images around a monitored water area to obtain a face image set, the face recognition module carries out recognition comparison on the face image set and a target face image set, judges the coincidence degree of facial features of the face image set and the target face image set, automatically sends out warning information, judges whether the face image set is continuously acquired or not, sends out artificial shouting warning information, continuously judges whether the face image set is acquired or not, sends out information pushing, continuously judges whether the face image set is acquired or not, sends out a time limit worksheet, the storage module stores the image information and the target face image set, carries out timely rescue on drowning personnel according to an emergency rescue device when a drowning accident occurs, and realizes early prevention and timely drowning, the timeliness of drowning early warning and rescue is improved.

Particularly, the embodiment of the invention carries out camera shooting and snapshot on the monitored water area through the camera arranged outdoors, carries out all-weather monitoring on the monitored water area, has less influence on the performance of the camera by the external environment, can normally work, has longer service life, can keep normal working state in long-time monitoring, ensures the requirement of real-time monitoring of the monitored water area, and improves the prevention effect on drowning conditions.

Particularly, when the specific embodiment of the invention identifies the face of the student entering the monitored water area, the student entering the monitored water area is identified, so that whether the personal information of the student entering the monitored water area is consistent with the target face image set or not can be identified, the entering monitored water area can be positioned, the time of the entering monitored water area can be recorded, the recording of multiple factors of personnel, geographic positions and time is realized, the original data is provided for a subsequent data analysis module, and the comprehensiveness of data recording is improved.

Particularly, when the audible and visual alarm performs automatic voice warning on the students entering the monitored water area, whether the students entering the monitored water area accord with the target face image set or not can be distinguished according to the face identification comparison information, a warning effect can be achieved on the students not according with the target face image set, accurate name information of the students can be accurately sent out during automatic voice warning on the students according with the target face image set, the students are psychologically frightened, and the warning effect on the students entering the monitored water area is improved.

Particularly, when the early warning telephone is sent to parents of students entering the monitored water area, the early warning telephone is sent to the parents of the students in time through the video telephone, the parents of the students can be reminded in time, and the students reminded in the mode are all in the condition of artificial voice warning without leaving, so that the students neglect the danger of the monitored water area, are in danger, and the parents dissuade or repel the students, so that the warning function of the students entering the monitored water area is improved.

Particularly, in the specific embodiment of the invention, the data analysis module is used for analyzing the students entering the monitored water area to obtain an analysis result, so that the number of the students entering the monitored water area can be counted, and the monitored water area can be conveniently monitored in a key point; the students are involved in the risk and the frequency, so that the important education of the students who frequently enter the monitored water area is facilitated; in the time period of the student concerning the danger, the patrol of personnel is increased in the key time period; the age bracket of the student concerning the risk is convenient for the education department to perform the special education; dangerous area statistics is convenient for establishing a dangerous area database; the higher department can immediately modify the drowning prevention work requirement of the area governed by the lower department through the data analysis of the platform, and the drowning prevention knowledge is enhanced and popularized.

Particularly, the embodiment of the invention forces a background manager to regularly patrol the monitored water area at regular time through the electronic patrol module, thereby reducing the occurrence rate of drowning accidents.

Particularly, the specific embodiment of the invention simplifies the system processing flow and improves the operation efficiency of the system by arranging the management module; through setting up a key alarm module, use when making peripheral personnel discover that someone is drowned prevent that drowned system carries out a key and reports to the police, can be directly to sending out the institute and send alarm information, send out the institute and can directly go out police or contact near policeman or in time contact the head of a student, shortened policeman or the head of a student head of a family or the time that the rescue personnel went to the monitoring waters scene, field device is from taking life buoy, survival pole and life rope, peripheral personnel can directly carry out very first time's rescue to drowned student guaranteeing under the circumstances of self safety, guaranteed student's life health.

Drawings

Fig. 1 is a schematic structural diagram of a drowning prevention system according to an embodiment of the present invention.

Detailed Description

In order that the objects and advantages of the invention will be more clearly understood, the invention is further described in conjunction with the following examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.

It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, which are only for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.

Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

Referring to fig. 1, an embodiment of the present invention provides a drowning prevention system, which includes:

the camera 100 is connected with the server and used for monitoring a monitored water area and acquiring images and photos around the monitored water area to obtain image and photo information;

the alarm device 200 is arranged at a position 3 meters away from the land junction of the monitored water area, is connected with the server and is used for sending automatic alarm information and artificial shouting alarm information;

the server 300 is connected with the camera, the alarm device, the emergency rescue device and the display screen and used for receiving, storing and analyzing the image and photo information;

the emergency rescue device 400 is connected with the server and is used for emergently rescuing the drowning accident in the monitored water area;

the display screen 500 is connected with the server and used for displaying real-time video dynamic information of the camera;

the server comprises a face recognition module, a voice recognition module, an action recognition module, a communication module, a work order issuing module, a data storage module, a data analysis module, an electronic patrol module and a management module, wherein the face recognition module is used for recognizing the image photo information to obtain a face image set and performing recognition comparison on the face image set and a target face image set to obtain recognition comparison information, the voice recognition module is used for recognizing human voice in a monitored water area, the action recognition module is used for recognizing human limb actions in the monitored water area, the communication module is used for communicating with a background manager and parents of students, the work order issuing module is used for issuing a time-limited work order to a gridder in charge of monitoring the water area in time, and the data storage module is used for storing the image photo information and the target face image set, the data analysis module is used for analyzing and monitoring the geographical position that the waters got into student's number of people, time quantum, age bracket and monitoring area, the electron module of patrolling is used for setting up the patrol cycle and showing the patrol suggestion, the management module is used for managing project, personnel, equipment, label, sign in, warning and account number.

Specifically, the alarm device comprises an audible and visual alarm and an artificial voice alarm device, the audible and visual alarm is connected with the server and the emergency aid device and used for sending out automatic alarm information, the artificial voice alarm device is connected with the server and used for sending out artificial shouting alarm information to a monitored water area,

emergent salvation device includes a key alarm device and automatic salvation device, a key alarm device with the server is connected with the room on duty of dispatching out for in time send alarm information and start to the room on duty of dispatching out audible-visual annunciator carries out audible-visual annunciator, automatic salvation device is connected with the server, is used for throwing life-saving equipment to drowned area.

Specifically, the distance around the monitoring water area is 3 to 5 meters from the boundary between the edge of the monitoring water area and the land in the embodiment of the invention.

Specifically, the audible and visual alarm sends out automatic warning information according to the identification comparison information, the server obtains first feedback information,

the communication module pushes information according to the first feedback information, the artificial voice warning device sends out artificial shouting warning information, the server obtains second feedback information,

the communication module pushes information according to the second feedback information and makes a video call to a target telephone, the server obtains third feedback information,

the communication module pushes information according to the third feedback information, the work order issuing module issues a time-limited work order to obtain fourth feedback information,

and the data storage module stores the fourth feedback information.

Specifically, in the embodiment of the invention, the camera acquires images around a monitored water area to obtain a face image set, the face recognition module performs recognition comparison on the face image set and a target face image set, judges the coincidence degree of facial features of the face image set and the target face image set, automatically sends out warning information, judges whether the face image set is continuously acquired, sends out artificial shouting warning information, continuously judges whether the face image set is acquired, sends out information pushing, continuously judges whether the face image set is acquired, sends out a time limit worksheet, the storage module stores the image information and the target face image set, carries out timely rescue on drowning personnel according to an emergency rescue device when a drowning accident occurs, and realizes early prevention and timely rescue on the drowning condition, the timeliness of drowning early warning and rescue is improved.

Specifically, the face image set only comprises the front face of the face, the calculation processing pressure of the drowning prevention system is reduced, and the drowning prevention system only collects the face image set of the complete face image in a mode that at least 3 cameras are arranged in the same monitoring water area.

Specifically, the identification contrast information comprises eye characteristic information, mouth characteristic information, nose characteristic information and face proportion characteristic information,

if at least 90% xN of the N eye feature information extracted from the face image set coincides with the eye feature information of the target face image set, the eye feature information of the face image set coincides with the eye feature information of the target face image set,

if at least 90% xn of the N mouth feature information extracted from the face image set coincides with the mouth feature information of the target face image set, the mouth feature information of the face image set coincides with the mouth feature information of the target face image set,

if at least 90% XN of the N pieces of nose feature information extracted from the face image set coincide with the nose feature information of the target face image set, the nose feature information of the face image set coincides with the nose feature information of the target face image set,

and if the error between the facial proportion feature information extracted from the facial image set and the facial proportion feature information of the target facial image set is less than or equal to 10%, the facial proportion feature information of the facial image set is consistent with the facial proportion feature information of the target facial image set.

Specifically, the first identification contrast information is that the eye feature information, the mouth feature information, the nose feature information and the face proportion feature information of the face image set do not meet the requirement of matching with at least 3 feature information of the eye feature information, the mouth feature information, the nose feature information and the face proportion feature information of the target face image set,

and the second identification contrast information is that the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the face image set meet the requirement of being consistent with at least 3 characteristic information in the eye characteristic information, the mouth characteristic information, the nose characteristic information and the face proportion characteristic information of the target face image set.

Specifically, the camera monitors a monitored water area and collects images and photos around the monitored water area to obtain image and photo information, wherein the photo information comprises the positioning of the monitored area, face image information and photo collection time.

Specifically, the embodiment of the invention carries out camera shooting and snapshot on the monitored water area through the camera arranged outdoors, carries out all-weather monitoring on the monitored water area, has less influence on the performance of the camera by the external environment, can normally work, has longer service life, can keep normal working state in long-time monitoring, ensures the requirement of real-time monitoring of the monitored water area, and improves the prevention effect on drowning conditions.

Specifically, the face recognition module is configured to recognize the image photo information to obtain a face image set, and perform recognition comparison between the face image set and a target face image set to obtain recognition comparison information, where the recognition comparison information includes first recognition comparison information and second recognition comparison information.

Specifically, when the specific embodiment of the present invention identifies the faces of the students entering the monitored water area, the students entering the monitored water area are identified, so that not only can whether the personal information of the students entering the monitored water area is consistent with the target face image set be identified, but also the students entering the monitored water area can be positioned and the time of the students entering the monitored water area can be recorded, thereby realizing the recording of multiple factors of personnel, geographic position and time, providing the original data for the subsequent data analysis module, and improving the comprehensiveness of data recording.

Specifically, the audible and visual alarm sends out automatic warning information according to the identification comparison information, the server obtains first feedback information, the automatic warning information comprises first automatic warning information and second automatic warning information,

the first automatic warning message is a field message edited by the system in advance,

the second automatic warning information is the name of the student and the field information edited by the system in advance.

Specifically, the first feedback information includes first positive feedback information and first negative feedback information,

the first positive feedback information is that the face recognition module can not recognize the face image set after 10 seconds after the audible and visual alarm sends out the automatic warning information,

the first negative feedback information is that the face recognition module can recognize the face image set after 10 seconds after the audible and visual alarm sends out the automatic warning information.

Specifically, when the audible and visual alarm carries out automatic voice warning on the students entering the monitored water area, whether the students entering the monitored water area accord with the target face image set or not can be distinguished according to the face identification contrast information, a warning effect can be achieved for the students not according with the target face image set, accurate name information of the students can be accurately sent out when the students accord with the target face image set are subjected to automatic voice warning, the students are psychologically frightened, and the warning effect of the students entering the monitored water area is improved.

Specifically, the communication module comprises a first pushing unit, a second pushing unit and a third pushing unit, the first pushing unit is used for pushing first pushing information to the background administrator according to the identification comparison information,

and if the identification contrast information is second identification contrast information, the first push information comprises the positioning of the monitored water area, the captured face picture, the time, the school of the student, the class and the name.

Specifically, the artificial voice warning device is used for sending out artificial shouting warning information according to the first push information, the server obtains second feedback information, the second feedback information comprises second positive feedback information and second negative feedback information,

the second positive feedback information is that the face recognition module can not recognize the face image set after the artificial voice warning device sends out artificial shouting warning information for 10 seconds,

the second negative feedback information is that the face recognition module can still recognize the face image set after the artificial voice warning device sends out artificial shouting warning information for 10 seconds.

Specifically, the second pushing unit is configured to push second pushing information to the target phone number according to the second feedback information, the server obtains third feedback information,

the second push information is a video phone of the monitored water area,

the third feedback information includes third positive feedback information and third negative feedback information,

the third positive feedback information is that the face recognition module cannot recognize the face image set after the video telephone is interrupted for 10 seconds,

the third negative feedback information is that the face recognition module can still recognize the face image set after the video telephone is interrupted for 10 seconds.

Particularly, the second propelling movement unit is right when the head of a family of student that gets into the monitoring waters sends the early warning phone, through video telephone to the head of a family of student in time send the early warning phone, can be timely remind the head of a family of student, and the student who reminds through this kind of mode is at artifical voice warning and the condition that does not leave, explains that the student has neglected the danger in monitoring waters, places oneself in among the danger, and the head of a family's dissuasion or the laugh have more the warning effect this moment, has improved the warning effect that gets into the monitoring waters to the student.

Specifically, the work order issuing module is configured to issue a time-limited work order according to the third feedback information, the server obtains fourth feedback information,

and the fourth feedback information is that the face recognition module cannot recognize the face image set after 3 minutes after the time-limited work order is issued by the work order issuing module.

Specifically, the third pushing unit is configured to push the image photo information to a first target phone number and a second target phone number according to the fourth feedback information,

the holder of the first destination telephone number is a parent of the set of face images corresponding to the student,

the holder of the second target telephone number is the school of the student corresponding to the face image set.

Particularly, the data storage module is used for right image photo information and target face image set are stored, the storage module still stores thematic education information, the thematic education information is drowning prevention thematic education information of a student who collects face recognition information in a monitoring water area for a school, and the target face image set is stored in a picture uploading mode.

Specifically, the data analysis module is used for analyzing the number of people, time period, age group and geographical position of a monitoring area of a monitored water area entering a student.

Specifically, the data analysis module analyzes the students entering the monitored water area to obtain an analysis result, so that the number of people entering the monitored water area can be counted, and the monitored water area can be conveniently monitored in a key point; the students are involved in the risk and the frequency, so that the important education of the students who frequently enter the monitored water area is facilitated; the students are involved in the time period of the risk, so that the patrol of the personnel is increased in the key time period; the age bracket of the student concerning the risk is convenient for the education department to perform the special education; dangerous area statistics is convenient for establishing a dangerous area database; the higher department can immediately modify the drowning prevention work requirement of the area governed by the lower department through the data analysis of the platform, and the drowning prevention knowledge is enhanced and popularized.

Specifically, the electronic patrol module is used for setting a patrol period and presenting a patrol prompt, and the period of the electronic patrol is set and the patrol prompt is given according to different seasons, so that a background manager can patrol a monitored water area according to the patrol period.

Specifically, according to the embodiment of the invention, a background manager is forced to regularly and regularly patrol the monitored water area through the electronic patrol module, so that the occurrence rate of drowning accidents is reduced.

Specifically, the management module is used for managing projects, personnel, equipment, tags, check-in, reminding and account numbers, and comprises a project management unit, a personnel management unit, an equipment management unit, a tag management unit, a check-in management unit, a reminding management unit and an account number management unit,

the project management unit is used for managing project information, including adding, modifying and deleting project information, the project is school name and area name, the personnel management unit is used for managing the target face image set, including newly-increased, deleted, retrieving and presenting the identification record of the target face image set, the equipment management unit is used for newly-increased, deleted and inquiring management of sound equipment, a camera and an alarm rod, the label management unit is used for marking and inquiring students corresponding to the face image set, the mark comprises 'dangerous persons', the attendance management unit is used for signing in management of attendance personnel, including newly-increased, retrieved and checked, the reminding management unit is used for reminding the background manager when the equipment is abnormal, the account management unit is used for newly-increased, modified and deleted account information of the drowning-preventing system, Query and edit management.

Specifically, the one-key alarm device is connected with the server and the dispatching office for sending alarm information to the dispatching office in time and starting the audible and visual alarm to give audible and visual alarm, and the one-key alarm device is provided with a life buoy, a life rod and a life rope.

Specifically, the automatic rescue device judges and rescues drowning accidents in a monitored water area according to the voice recognition module and the action recognition module,

the voice recognition module adopts a long voice recognition technology and is used for optimizing a video scene, recognizing voice help calling contents in a video, judging whether a person in a water area drowns or not,

the optimization is to reduce noise of the collected sound to obtain clear human language vocabulary information,

the noise reduction is that the drowning prevention system collects at least one kind of water splash sound on the water surface in advance, the water splash sound is converted into a water splash character sequence, after the drowning prevention system collects all the sound information on the site, all the sound information is converted into a total character sequence, the character sequence with the contact ratio with the water splash character sequence being more than or equal to 85 percent in the total character sequence is deducted, and the residual character sequence of the total character sequence after deducting is read, so as to realize the voice call-for-help content in the identification video,

Specifically, the motion recognition module is used for accurately recognizing the information of the limbs of the person in the image, tracking, positioning and identifying the attribute of the person, judging whether the person in the water area drowns or not,

the infrared sensor is arranged on the camera of the drowning prevention system for tracking, so that the human body and the environment can be accurately identified,

the positioning is to determine a propagation center of at least one surface wave by an AI algorithm for a shape of the surface wave and a propagation velocity of the surface wave,

the attribute recognition is to judge whether the person in the water area is in normal swimming or drowning according to the AI algorithm, judge the human body action character sequence of the person in the water area by setting a first target human body action character sequence,

if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the first target human body action character sequence is more than or equal to 80 percent, judging that the personnel in the water area drowns,

and if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the first target human body action character sequence is less than 20%, judging that the personnel in the water area is not drowned.

Specifically, if the voice recognition module and the action recognition module simultaneously judge that the person in the water area drowns, the server sends an instruction to a throwing device on the automatic rescue device,

the throwing azimuth place of the throwing device is within 1 meter of the propagation center of the at least one water surface wave, and the throwing device throws at least 2 life buoys or at least 2 life ropes or at least 2 life poles.

Specifically, the selection of the lifesaving equipment is judged according to the distance D between the throwing position and the throwing device,

if D is more than 0 and less than 5 meters, the throwing device throws at least 2 life buoys to the throwing azimuth place,

if D is more than or equal to 5 and less than 8 meters, the throwing device throws at least 2 lifesaving rods to the throwing azimuth place,

and if D is more than or equal to 8 meters, throwing at least 2 life ropes to the throwing azimuth place by the throwing device.

Specifically, after the life-saving equipment is thrown to the throwing position by the throwing device, the action recognition module recognizes the limb action of drowned people, and judges the human action character sequence of the collected people in the water area by setting a second target human action character sequence,

if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the second target human body action character sequence is more than or equal to 80 percent, the personnel in the water area is judged to take the lifesaving equipment,

if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the second target human body action character sequence is less than 20%, the personnel in the water area is judged not to take the thrown lifesaving equipment, the server continuously sends an instruction to a throwing device on the automatic rescue device, and the throwing device continuously throws the lifesaving equipment to the throwing position.

Specifically, after judging whether the personnel in the water area take the lifesaving equipment or not, the effect of using the lifesaving equipment by the personnel in the water area is judged by setting a third target human body action character sequence,

if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the third target human body action character sequence is more than or equal to 80 percent, the personnel in the water area is judged to effectively use the lifesaving equipment,

and if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the third target human body action character sequence is less than 20%, judging that the personnel in the water area does not effectively use the lifesaving equipment.

In particular, when it is determined that the life saving equipment is not effectively used by the person in the water area,

if D is more than 0 and less than 5 meters, the throwing device continuously throws at least 2 lifesaving rods or at least 2 lifesaving ropes to the throwing azimuth place, the lifesaving rods or the lifesaving ropes are matched with the lifebuoy for use, the rescue success rate of personnel in the water area is improved,

if D is more than or equal to 5 and less than 8 meters, the throwing device continues to throw at least 2 life ropes to the throwing azimuth place, the life ropes and the life rods are matched for use, the rescue success rate of personnel in the water area is improved,

if D is larger than or equal to 8 meters, the throwing device continues to throw at least 2 life ropes and at least 2 life poles to the throwing position, the life ropes and the life poles are used in a matched mode, firstly, people in the water area can smoothly reach the distance of the life poles through the life ropes, and the success rate of rescuing the people in the water area is improved.

Specifically, the specific embodiment of the invention simplifies the system processing flow and improves the operation efficiency of the system by arranging the management module; through setting up a key alarm module, use when making peripheral personnel discover that someone is drowned prevent that drowned system carries out a key and reports to the police, can be directly to sending out the institute and send alarm information, send out the institute and can directly go out police or contact near policeman or in time contact the head of a student, shortened policeman or the head of a student head of a family or the time that the rescue personnel went to the monitoring waters scene, field device is from taking life buoy, survival pole and life rope, peripheral personnel can directly carry out very first time's rescue to drowned student guaranteeing under the circumstances of self safety, guaranteed student's life health.

Specifically, the cooperative work flow of the face recognition module, the voice recognition module and the action recognition module is as follows,

when the face recognition module recognizes a face image within a range of 3 meters of a land junction of a monitored water area and does not recognize the face image in the water surface of the monitored water area, the drowning prevention system automatically alarms, and voice recognition and action recognition are not needed at the moment;

when the face recognition module recognizes face images in the range of 3 meters of the land junction of the monitored water area and in the water surface of the monitored water area, the drowning prevention system warns people in the range of 3 meters of the land junction of the monitored water area and performs voice recognition and action recognition on the people in the water surface of the monitored water area;

when the face recognition module does not recognize the face image in the monitoring water area within 3 meters distance range of the land junction and recognizes the face image in the monitoring water area water surface, the drowning prevention system performs voice recognition and action recognition on the personnel in the monitoring water area water surface according to the voice recognition module and the action recognition module.

So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is apparent to those skilled in the art that the scope of the present invention is not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention; various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A drowning prevention system, comprising:

the server comprises a face recognition module, a voice recognition module, an action recognition module, a communication module, a work order issuing module, a data storage module, a data analysis module, an electronic patrol module and a management module, wherein the face recognition module is used for recognizing the image photo information to obtain a face image set and performing recognition comparison on the face image set and a target face image set to obtain recognition comparison information, the voice recognition module is used for recognizing human voice in a monitoring water area, the action recognition module is used for recognizing human limb actions in the monitoring water area, the communication module is used for communicating with a background manager and parents of students, the work order issuing module is used for issuing a time-limited work order to a gridder in charge of monitoring the water area in time, and the data storage module is used for storing the image photo information and the target face image set, data analysis module is used for the geographical position that analysis monitoring waters got into student's times of the people, time quantum, age bracket and monitoring area, electron patrol module is used for setting up patrol cycle and presents the patrol suggestion, management module is used for managing project, personnel, equipment, label, sign in, warning and account number.

2. The drowning prevention system according to claim 1, wherein the face recognition module is configured to recognize the image photo information to obtain a face image set, and perform recognition comparison between the face image set and a target face image set to obtain recognition comparison information, the recognition comparison information includes first recognition comparison information and second recognition comparison information, the face image set includes only a front face of a face, so as to reduce the computational processing pressure of the drowning prevention system, the drowning prevention system collects only the face image set of a complete face image by installing at least 3 cameras in a same monitoring water area, the recognition comparison information includes eye feature information, mouth feature information, nose feature information and face proportion feature information,

if at least MxN eye feature information in the N eye feature information extracted from the face image set coincides with eye feature information of a target face image set, the eye feature information of the face image set coincides with the eye feature information of the target face image set, wherein M is the percentage of the eye feature information extracted from the face image set coinciding with the eye feature information of the target face image set in the eye feature information of the face image set in the target face image set,

if at least M x N pieces of nose feature information in the N pieces of nose feature information extracted from the face image set coincide with the nose feature information of the target face image set, the nose feature information of the face image set is consistent with the nose feature information of the target face image set,

3. The drowning prevention system according to claim 2, wherein the alarm device comprises an audible and visual alarm connected to the server and the emergency assistance device for issuing an automatic alarm message, and an artificial voice warning device connected to the server for issuing an artificial shout warning message to the monitored water area;

4. The drowning prevention system according to claim 3, wherein the audible and visual alarm sends out automatic warning information according to the identification contrast information, the server obtains first feedback information, the communication module carries out information push according to the first feedback information, the artificial voice warning device sends out artificial shouting warning information, the server obtains second feedback information, the communication module carries out information push according to the second feedback information and makes a video call to a target telephone, the server obtains third feedback information, the communication module carries out information push according to the third feedback information, the work order issuing module issues a time-limited work order to obtain fourth feedback information,

and the data storage module stores the fourth feedback information.

5. The drowning prevention system according to claim 4, wherein the audible and visual alarm issues an automatic warning message based on the identification comparison message, the server obtains a first feedback message, the automatic warning message includes a first automatic warning message and a second automatic warning message,

the first automatic warning information is field information edited by a system in advance, and the second automatic warning information is student names and the field information edited by the system in advance.

6. The drowning prevention system of claim 5, wherein the first feedback information comprises first positive feedback information and first negative feedback information,

7. The drowning prevention system of claim 6, wherein the communication module comprises a first pushing unit, a second pushing unit and a third pushing unit, the first pushing unit is used for pushing first pushing information to the background administrator according to the identification comparison information,

if the identification contrast information is second identification contrast information, the first push information comprises the positioning of a monitored water area, a captured face picture, time, school of a student, class and name,

the artificial voice warning device is used for sending out artificial shout warning information according to the first pushing information, the server obtains second feedback information, the second feedback information comprises second positive feedback information and second negative feedback information, the second positive feedback information is that the human face recognition module cannot recognize the human face image set after the artificial voice warning device sends out the artificial shout warning information for 10 seconds, the second negative feedback information is that the human face recognition module can still recognize the human face image set after the artificial voice warning device sends out the artificial shout warning information for 10 seconds,

8. The drowning prevention system according to claim 5, wherein the data storage module is configured to store the image photograph information and a target face image set, the storage module further stores topical education information, the topical education information is specific education information for preventing drowning performed by a school on students whose face identification information is collected in a monitored water area, the target face image set is stored by picture upload,

the management module comprises an item management unit, a personnel management unit, an equipment management unit, a label management unit, a check-in management unit, a reminding management unit and an account management unit, wherein the item management unit is used for managing item information including adding, modifying and deleting the item information, the items are school names and area names, the personnel management unit is used for managing the target face image set including adding, deleting, retrieving and presenting identification records of the target face image set, the equipment management unit is used for adding, deleting and inquiring the sound equipment, the equipment management unit is used for carrying out marking and inquiring on students corresponding to the face image set, the check-in management unit is used for carrying out check-in management on attendance personnel including adding, retrieving and checking, the reminding management unit is used for reminding the background manager when the equipment is abnormal, and the account management unit is used for performing addition, inquiry and editing management on the account information of the drowning prevention system.

9. The drowning prevention system according to claim 8, wherein the automatic rescue device judges and rescues drowning accidents in the monitored water area according to the voice recognition module and the motion recognition module,

10. The drowning prevention system of claim 9 wherein the action recognition module is used to accurately recognize the information of the human body in the image, track, locate and identify the attributes of the human body, determine whether the drowning of the human body occurs in the water area,

after the life-saving equipment is thrown to the throwing position by the throwing device, the action recognition module recognizes the limb action of drowning people, judging the human body action character sequence of the collected personnel in the water area by setting a second target human body action character sequence, if the coincidence degree of the human body action character sequence of the collected personnel in the water area and the second target human body action character sequence is more than or equal to 80 percent, judging that the personnel in the water area take the lifesaving equipment, if the coincidence degree of the collected human body action character sequence of the personnel in the water area and the second target human body action character sequence is less than 20 percent, if the situation that the people in the water area do not take the thrown lifesaving equipment is judged, the server continuously sends an instruction to a throwing device on the automatic rescue device, and the throwing device continuously throws the lifesaving equipment to the throwing position.