CN112001346B - Vital sign detection method and system based on multi-algorithm fusion collaboration - Google Patents
Vital sign detection method and system based on multi-algorithm fusion collaboration Download PDFInfo
- Publication number
- CN112001346B CN112001346B CN202010894404.5A CN202010894404A CN112001346B CN 112001346 B CN112001346 B CN 112001346B CN 202010894404 A CN202010894404 A CN 202010894404A CN 112001346 B CN112001346 B CN 112001346B
- Authority
- CN
- China
- Prior art keywords
- bunk
- vital sign
- human body
- detected
- sign data
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 77
- 230000004927 fusion Effects 0.000 title claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 102
- 230000009471 action Effects 0.000 claims abstract description 69
- 230000033001 locomotion Effects 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 34
- 230000002159 abnormal effect Effects 0.000 claims abstract description 16
- 210000003128 head Anatomy 0.000 claims description 37
- 230000008859 change Effects 0.000 claims description 20
- 210000001061 forehead Anatomy 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 238000011897 real-time detection Methods 0.000 claims description 6
- 230000001960 triggered effect Effects 0.000 claims description 4
- 201000010099 disease Diseases 0.000 abstract description 18
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 18
- 206010010144 Completed suicide Diseases 0.000 abstract description 10
- 230000006399 behavior Effects 0.000 abstract description 6
- 230000029058 respiratory gaseous exchange Effects 0.000 abstract description 4
- 206010063746 Accidental death Diseases 0.000 abstract description 3
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 abstract description 3
- 239000010931 gold Substances 0.000 abstract description 3
- 229910052737 gold Inorganic materials 0.000 abstract description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000009434 installation Methods 0.000 description 8
- 241000273930 Brevoortia tyrannus Species 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 206010010071 Coma Diseases 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000035939 shock Effects 0.000 description 3
- 208000024891 symptom Diseases 0.000 description 3
- 206010002383 Angina Pectoris Diseases 0.000 description 2
- 208000019025 Hypokalemia Diseases 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 208000029078 coronary artery disease Diseases 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 208000024896 potassium deficiency disease Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
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
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
The invention discloses a vital sign detection method and a vital sign detection system based on multi-algorithm fusion coordination, and belongs to the field of artificial intelligence. Through the mode of detecting limb actions in real time and detecting heart rate at regular time, some inconspicuous suicide and self-disabled behaviors can be detected, for example, the condition of suicide in a mode of selecting a Mongolian head and a wet tissue to block breathing at night when a prison is easy to slacken, and then, for example, self-disabled is selected to be carried out by selecting a back-facing monitoring camera, and sudden diseases occur; in the method, in 529 days of test operation of a certain monitor, a head-covering sleep sense message 4366 is detected, and a real message 3101 is confirmed, wherein 1 is the condition of head-covering suicide; 8 pieces of abnormal information of limb movements and heart rate vital sign data are detected, wherein 3 pieces of abnormal information are serious sudden diseases. The method can accurately alarm in time and correct the head covering condition in time; for the emergency, the gold rescue time is striven for, and the accidental death of the people in the presence is avoided.
Description
Technical Field
The invention relates to a vital sign detection method and a vital sign detection system based on multi-algorithm fusion coordination, and belongs to the field of artificial intelligence.
Background
The supervision is a special place, and the personnel under the control are limited in freedom, so that the personnel under the control are easy to change greatly in physiology and psychology, so that suicide, self-disabled, sudden diseases or aggravated conditions of the original diseases are caused, and meanwhile, factors such as fighting by the personnel under the control are also easy to cause or accelerate death. The manager in the monitoring department needs to pay close attention to the occurrence of the events, timely discover and control the behaviors such as suicide, self-disabled and the like, and timely discover and arrange rescue measures for sudden diseases or the situation of aggravated original diseases, so that abnormal death of the people in the presence is avoided as much as possible.
The existing prison management monitoring system monitors the behavior activities of on-press personnel in real time in a video monitoring and prison police patrol mode, but aims at the situation that some barely detected suicide and self-disabled behaviors are used, such as the situation that the prison police are prone to slackening at night, the mode of masking heads and wet tissues to block breathing is selected, the situation that some self-disabled is carried out by selecting the mode of back facing to the monitoring camera, and in addition, sudden diseases with unobvious symptoms occur, and the existing prison monitoring system cannot be effectively prevented and controlled in advance in time.
Disclosure of Invention
In order to solve the problem that the current supervision and management monitoring system cannot effectively monitor some conditions, the invention provides a vital sign detection method and a vital sign detection system based on multi-algorithm fusion coordination.
The method is applied to the monitoring room to detect vital signs of the on-press personnel, and the method adopts the mode of detecting vital sign data of limb actions in real time and simultaneously assisting in detecting vital sign data of heart rate at fixed time to detect the vital signs of the on-press personnel;
according to the method, the full coverage video data of the monitoring room is obtained through a fixed-mounted dome camera, and the partial coverage video data of the monitoring room is obtained through a rotatable spherical camera with adjustable focal length.
Optionally, the real-time detecting vital sign data of the limb movement includes:
detecting whether human bodies exist on all the berths in the monitoring room according to the full coverage video data of the monitoring room acquired by the hemispherical camera;
for the detected body position, acquiring limb actions of the human body by adopting a human skeleton micro action detection algorithm to acquire vital sign data, comparing the vital sign data with the limb action vital sign data corresponding to the position acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the vital sign data of the limb action of the human body;
for the position of the human body which is not detected, whether the human body has actions is further detected.
Optionally, for the position of the undetected human body, further detecting whether there is an action includes:
if the bunk is detected to have actions, further determining that the bunk is on the head of a person to be pressed according to whether the action change amplitude value is in a normal range or not, and triggering head covering alarm if the action change amplitude value is in the normal range; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if the position is detected to be not moved, acquiring the position on-press personnel information from a monitoring personnel management system in real time, judging whether the position is always unmanned, and simultaneously acquiring the on-press personnel information on duty at night to determine whether the position personnel is on duty; if the bunk is unmanned or the bunk is on the press, preparing for the next detection; if the bunk is someone and the bunk is not proper, the spherical camera which is rotatable in a linkage mode and adjustable in focal length is used for rechecking the bunk.
Optionally, the rotatable spherical camera of linkage and focus adjustable rechecks this bunk, includes:
controlling a rotatable spherical camera with adjustable focal length to acquire the bunk video picture, and judging whether a human body exists or not according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length;
if the human body exists in the bunk according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length, acquiring the limb movement of the human body by adopting a human body skeleton micro-movement detection algorithm to acquire vital sign data, comparing the vital sign data with the limb movement vital sign data corresponding to the bunk acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the limb movement vital sign data of the human body;
if the situation that the human body does not exist in the bunk is judged according to the bunk video picture acquired by the rotatable spherical camera with the adjustable focal length, abnormal alarm is triggered.
Optionally, the timing detection of heart rate vital sign data includes:
controlling a rotatable spherical camera with adjustable focal length to acquire whether the nth berth has a human body or not, wherein N is more than or equal to 1 and less than or equal to N, and N is the number of berths in the monitoring room;
if the nth berth is detected to have a human body, positioning the nth berth to the forehead or the neck of the human body by adopting a target tracking algorithm, and acquiring heart rate data through the skin of the forehead or the neck; if the heart rate data is in the normal range, detecting the next bunk; if the heart rate is not in the normal range, triggering a rescue alarm;
if the nth bunk is detected to have no human body, further detecting whether the nth bunk has actions.
Optionally, if it is detected that the nth bunk does not have a human body, further detecting whether the nth bunk has an action, including:
if the nth bunk is detected to have the action, further determining that the bunk is on the head of the person to be pressed according to whether the action change amplitude value is in a normal range or not, and triggering head covering alarm if the action change amplitude value is in the normal range; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if no action of the nth bunk is detected, acquiring information of the bunk on-press personnel from a monitoring personnel management system in real time, judging whether the bunk is not all the time, and simultaneously acquiring information of the on-press personnel on duty at the night to determine whether the bunk personnel are on duty; if the bunk is unmanned or the bunk is on the press, detecting the next bunk; if the berth is occupied and the berth is not worth the person, triggering abnormal alarm.
Optionally, in the process of detecting vital sign data of limb movement in real time, if the bunk is detected to have movement, further according to whether the movement variation amplitude value is in a normal range, the method includes:
and judging the difference of the video pictures corresponding to the bunk during the front and rear detection by an image detection algorithm, and determining whether the motion change amplitude value is in a normal range according to the difference of the video pictures corresponding to the bunk during the front and rear detection.
Optionally, in the process of detecting heart rate vital sign data at regular time, if the nth bunk is detected to have an action, further according to whether the action change amplitude value is in a normal range, the method includes:
and judging the difference of the video picture corresponding to the nth bunk detected at the last time through an image detection algorithm, and determining whether the motion variation amplitude value is in a normal range according to the difference of the video picture corresponding to the nth bunk detected at the last time.
The application also provides a vital sign monitoring system based on multi-algorithm fusion coordination, which is applied to monitoring rooms for vital sign detection of on-site personnel, and comprises a hemisphere camera which is fixedly installed to acquire monitoring room full coverage video data and a rotatable spherical camera with adjustable focal length to acquire monitoring room partial coverage video data;
the monitoring system adopts the vital sign detection method based on multi-algorithm fusion and synergy to detect the vital signs of the on-press personnel in real time.
Optionally, the monitoring system informs corresponding management personnel in different modes for alarms of different conditions; the monitoring system records the limb actions and heart rate vital sign data of all monitored on-press personnel in real time.
The invention has the beneficial effects that:
through the mode of detecting limb actions in real time and detecting heart rate at fixed time, the method can detect some inconspicuous suicide and self-disabled behaviors, such as the condition of suicidal action by selecting a mode of covering the head and blocking the breath by wet tissues at night when the prison is easy to lose, and the condition of self-disabled by selecting a mode of backing the monitoring camera, and the occurrence of sudden diseases with unobvious symptoms can be detected in time, the method can detect the condition of covering the head and sleeping sense information 4366 in 529 days of certain monitoring run, and confirms the condition of real information 3101, wherein 1 is the condition of covering the head and suicide; 8 pieces of abnormal information of limb movements and heart rate vital sign data are detected, wherein 3 pieces of abnormal information belong to serious sudden diseases (the sudden diseases are diagnosed as coronary heart disease, angina pectoris, hypokalemia and other diseases after urgent treatment by a medical doctor in a monitoring department). The method can accurately alarm the above conditions in time and correct the head covering condition in time; for the emergency, the gold rescue time is striven for, and the accidental death of the people in the presence is avoided.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of real-time detection in a vital sign detection method based on multi-algorithm fusion synergy in an embodiment of the invention.
Fig. 2 is a timing detection flow chart in a vital sign detection method based on multi-algorithm fusion synergy in an embodiment of the invention.
Fig. 3 is a schematic diagram of hardware device installation in a practical application scenario in an embodiment of the present invention.
Fig. 4 is a schematic diagram of the shop-through of a multi-person monitoring room in a practical application scenario according to an embodiment of the present invention.
Fig. 5 is a schematic view of a camera installed at a practical application scene according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings.
Embodiment one:
the embodiment provides a vital sign detection method based on multi-algorithm fusion coordination, which is applied to vital sign detection of on-press personnel in a monitoring room, wherein the method adopts a mode of detecting vital sign data of limb movements in real time and simultaneously assisting in detecting vital sign data of heart rate at fixed time to detect the vital sign of the on-press personnel;
according to the method, the full coverage video data of the monitoring room is obtained through a fixed-mounted dome camera, and the partial coverage video data of the monitoring room is obtained through a rotatable spherical camera with adjustable focal length.
The real-time detection of limb movement vital sign data comprises:
detecting whether human bodies exist on all the berths in the monitoring room according to the full coverage video data of the monitoring room acquired by the hemispherical camera;
for the detected body position, acquiring limb actions of the human body by adopting a human skeleton micro action detection algorithm to acquire vital sign data, comparing the vital sign data with the limb action vital sign data corresponding to the position acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the vital sign data of the limb action of the human body;
for the berth where the human body cannot be detected, further detecting whether the human body has actions; specifically, if the bunk is detected to act, whether the amplitude value of the action change is in a normal range is further determined, if the amplitude value of the action change is in the normal range, the bunk is determined to be on the head of a person to be pressed, and head covering alarm is triggered; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if the position is detected to be not moved, acquiring the position on-press personnel information from a monitoring personnel management system in real time, judging whether the position is always unmanned, and simultaneously acquiring the on-press personnel information on duty at night to determine whether the position personnel is on duty; if the bunk is unmanned or the bunk is on the press, preparing for the next detection; if the bunk is someone and the bunk is not proper, the spherical camera which is rotatable in a linkage mode and adjustable in focal length is used for rechecking the bunk.
Linkage rotatable and focus adjustable spherical camera rechecks this bunk, includes:
controlling a rotatable spherical camera with adjustable focal length to acquire the bunk video picture, and judging whether a human body exists or not according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length;
if the human body exists in the bunk according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length, acquiring the limb movement of the human body by adopting a human body skeleton micro-movement detection algorithm to acquire vital sign data, comparing the vital sign data with the limb movement vital sign data corresponding to the bunk acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the limb movement vital sign data of the human body;
if the situation that the human body does not exist in the bunk is judged according to the bunk video picture acquired by the rotatable spherical camera with the adjustable focal length, abnormal alarm is triggered.
The timing detection of heart rate vital sign data comprises:
controlling a rotatable spherical camera with adjustable focal length to acquire whether the nth berth has a human body or not, wherein N is more than or equal to 1 and less than or equal to N, and N is the number of berths in the monitoring room;
if the nth berth is detected to have a human body, positioning the nth berth to the forehead or the neck of the human body by adopting a target tracking algorithm, and acquiring heart rate data through the skin of the forehead or the neck; if the heart rate data is in the normal range, detecting the next bunk; if the heart rate is not in the normal range, triggering a rescue alarm;
if it is detected that the nth bunk does not have a human body, further detecting whether the nth bunk has an action, specifically including:
if the nth bunk is detected to have the action, further determining that the bunk is on the head of the person to be pressed according to whether the action change amplitude value is in a normal range or not, and triggering head covering alarm if the action change amplitude value is in the normal range; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if no action of the nth bunk is detected, acquiring information of the bunk on-press personnel from a monitoring personnel management system in real time, judging whether the bunk is not all the time, and simultaneously acquiring information of the on-press personnel on duty at the night to determine whether the bunk personnel are on duty; if the bunk is unmanned or the bunk is on the press, detecting the next bunk; if the berth is occupied and the berth is not worth the person, triggering abnormal alarm.
Embodiment two:
the embodiment provides a vital sign detection method based on multi-algorithm fusion collaboration, as shown in fig. 1 and 2, wherein the method detects vital sign data of limb actions in real time, meanwhile, vital sign detection is carried out on the on-press personnel in a mode of timing detection of heart rate vital sign data, so that the problem that an existing monitoring system for supervision is unable to effectively monitor some conditions is solved.
Specifically, this embodiment will be described by taking an example of application to the multi-person monitoring room shown in fig. 3. As shown in fig. 3, 4 and 5, the length and width dimensions of the multi-person monitoring room are generally 15m and 4.5m, and the height is 7 m; usually, a dome camera A and a dome camera B are respectively installed at opposite angles at the position 3 meters higher above the monitoring room, and a dome camera C with a tripod head is installed at the middle position of the long side above the monitoring room. Wherein, dome cameras A and B are fixed in installation angle, and dome camera C is rotatable in shooting angle, and zooming can be realized at the same time.
The dome cameras A and B acquire monitoring room pictures in real time after adjusting related shooting parameters, the dome camera C with the cradle head carries out timing inspection on each bunk, and specific inspection time intervals can be set according to actual conditions; meanwhile, when videos acquired by the dome cameras A and B in real time are analyzed, if abnormal conditions are detected, the dome camera C can be linked to review corresponding bunkers.
Assuming that there are 15 beds in the multi-person monitoring room, as shown in fig. 4, the number N of beds is 15, and a sign is usually attached to the corresponding position, for example, the numbers 1,2, …, N, …, N are attached to each of the beds, and the cameras A, B and C acquire the monitoring video frames and transmit the monitoring video frames to the background server, and the server can determine the corresponding area of the corresponding position in the frames by using OCR recognition technology.
The vital sign detection method based on multi-algorithm fusion synergy comprises two parts of real-time detection of vital signs of limb movements and timing detection of vital signs of heart rate, and the following parts are respectively introduced:
a first part: detecting vital signs of limb movements in real time;
fig. 1 is a flowchart of real-time detection in a vital sign detection method based on multi-algorithm fusion and collaboration, for real-time detecting vital sign data of limb movements to realize monitoring on-press personnel, which comprises:
s11, detecting whether human bodies exist on all the berths according to video pictures acquired by hemispheres A and B;
the hemispheres A and B are used as the standard for covering the whole monitoring room during installation, because the distance of shooting distance can influence the definition degree of the picture content, and the diagonal installation of the hemispheres A and B can mutually compensate for the places with unclear shooting; and combining video pictures obtained respectively according to the hemispheres A and B to obtain a clear picture of the whole monitoring room.
Detecting whether human bodies exist in all the berths by adopting a human body detection algorithm; the human body detection algorithm can judge whether a human body exists according to the exposed skin area of the human body on the corresponding berth or whether the relevant part of the human body exists.
S12, detecting the body position, further acquiring the limb movement of the human body by adopting a human skeleton micro-movement detection algorithm to acquire vital sign data, comparing the vital sign data with the limb movement vital sign data corresponding to the position acquired by the last detection, and sending out an alarm if the difference value of the vital sign data exceeds the normal value of the vital sign data of the limb movement of the human body.
S13, if the human body is not detected, detecting whether the human body is in motion or not.
S131, when detecting whether the bunk moves, the difference of the video pictures during the two detection can be judged through the image detection technology by comparing the bunk picture obtained by the last detection, to determine whether the bunk belongs to the sleeping situation of the person under the control of the head (i.e. the person may have unintentional actions when sleeping while covering the head).
When the bunk is determined to act, whether the amplitude value of the action change is in a normal range or not can be further determined, if the amplitude value of the action change is in the normal range, the bunk is likely to be carelessly covered by a person on the press, and after the person is warned, a prison wakes up the bunk to expose the head; if the action variation amplitude value is not in the normal range, for example, if the action variation amplitude value of the bunk person is detected twice before and after, the action variation amplitude value of the bunk person is extremely small or extremely large, sudden diseases or the situation that the breathing is blocked after the head is covered to cause coma shock can occur, and corresponding rescue measures can be needed after the alarm.
And S132, if the position is detected to be not moved, acquiring the information of the position on-press personnel from a monitoring personnel management system in real time, judging whether the position is always unmanned, and simultaneously acquiring the information of the on-press personnel on duty at night so as to determine whether the position personnel is on duty.
In general, at night, a multi-person monitoring room can arrange two on-press personnel to patrol, or possibly arrange different time intervals to patrol, so that when detecting that the bed does not act, the information of the on-press personnel on the bed and the information of the on-press personnel on duty at night need to be acquired from a monitoring personnel management system.
If no action is detected, the bunk is not arranged on the person under the control of the person under the control of the system, or the bunk is at the same value at the moment of detection, the detection is completed and the next detection is prepared.
And S14, if the on-site personnel are arranged on the berth without detecting the action, and the on-site personnel are not in proper value at the detection time, the spherical machine C with the cradle head is linked for rechecking, and whether the human body exists on the berth is judged according to the video picture acquired by the spherical machine C.
Because hemispheres a and B can cover the entire monitoring room in some cases, the human body may not be detected due to the shooting angle, while the dome C may be rotatable for shooting angle while zooming may be achieved, so shooting may be performed from angles other than hemispheres a and B to determine whether the bunk is not detected due to the shooting angle alone.
If the human body is still not detected according to the picture shot by the ball machine C, the alarm is directly given;
if a human body is detected according to the image shot by the dome camera C, the limb motion of the lying human body is correspondingly detected to obtain vital sign data, i.e. step S12 is executed.
At the time of real-time detection of limb action vital sign data, the present application is also assisted in detecting heart rate vital sign data at regular time and is carried out vital sign detection to the on-press personnel, namely:
a second part: detecting heart rate vital signs at regular time;
fig. 2 is a timing detection flowchart in the vital sign detection method based on multi-algorithm fusion coordination, for timing detection of heart rate vital sign data to realize monitoring of on-press personnel, including:
s21, detecting whether a human body exists in an nth place according to a video picture acquired by a ball machine C with a cradle head (N is more than or equal to 1 and less than or equal to N);
the detection method can adopt a human body detection algorithm to detect whether the nth bunk has human bodies or not; the human body detection algorithm can judge whether a human body exists according to the exposed skin area of the human body on the corresponding berth or whether the relevant part of the human body exists.
S22, if the nth berth is detected to have a human body, acquiring heart rate data through the skin of the forehead or the neck, and judging whether the heart rate data is in a normal range; if the range is in the normal range, detecting the next bunk; if the position is not in the normal range, the person on the road can have sudden diseases or other life-threatening conditions and alarm.
Specifically, when detecting that the nth bunk exists in the human body, a target tracking algorithm can be adopted to position the forehead or the neck of the human body of the bunk, and heart rate data can be calculated.
To accurately calculate heart rate data, an image enhancement algorithm may be employed to enhance the acquired video frames.
S23, if the human body exists in the nth berth is not detected, detecting whether the nth berth acts or not;
s231 can determine whether the bunk belongs to the sleeping situation of the person under the condition of sleeping while the person is sleeping while the head is covered by comparing the bunk with the bunk picture obtained by the last detection, and can determine the difference of the video pictures during the two previous and subsequent detections by the image detection technology.
When the nth bunk is determined to act, whether the value of the variation amplitude of the action is in a normal range or not can be further determined, if the value of the variation amplitude of the action is in the normal range, the bunk is likely to be carelessly covered by a person on the press, and after the person on the press alarms, a prison wakes up the bunk to expose the head; if the action variation amplitude value is not in the normal range, for example, if the action variation amplitude value of the bunk person is detected twice before and after, the action variation amplitude value of the bunk person is extremely small or extremely large, sudden diseases or the situation that the breathing is blocked after the head is covered to cause coma shock can occur, and corresponding rescue measures can be needed after the alarm.
S232, if the nth bunk is detected to be not moving, acquiring the bunk on-press personnel information from a monitoring personnel management system in real time, judging whether the bunk is not all the time, and simultaneously acquiring the on-press personnel information on duty at the night to determine whether the bunk personnel is on duty.
In general, at night, a multi-person monitoring room can arrange two on-press personnel to patrol, or possibly arrange different time intervals to patrol, so that when detecting that the bed does not act, the information of the on-press personnel on the bed and the information of the on-press personnel on duty at night need to be acquired from a monitoring personnel management system.
If it is detected that the nth bunk is not arranged on the extruder or the nth bunk is at the same value at the time of detection, the next bunk is detected.
S24, if the n-th berth arranges the on-press personnel, and the on-press personnel of the berth have improper values at the detection time, the on-press personnel directly give an alarm.
In actual situations, there may be situations where the nth person gets up and uses the toilet, and this situation may be misreported, but the situation is within the fault tolerance range. Because the activities of the people in the actual monitoring room are required to meet the corresponding time regulations.
In the process of the real-time and timing detection, in order to reduce detection accuracy caused by the reason of the video picture, after the video picture is acquired, color image enhancement based on HSV space is performed on the video picture, and the method is specifically as follows:
in order to ensure color undistorted, the RGB image is converted into HSV for processing, the brightness S in the RGB image is operated, histogram equalization is performed first, and then sharpening processing of a Gaussian-Laplace filter is performed to improve the brightness.
The sharpening process may be accomplished by spatial differentiation, where the response intensity of the differentiation operator is related to the degree of abrupt change in the image at that point, which enhances edges and other abrupt changes (e.g., noise) while weakening the areas of slow gray scale variation.
In the method for detecting the limb movements and detecting the heart rate at fixed time in real time, various alarming conditions exist, in practical application, different alarming sounds or alarming lamps with different colors can be set according to the practical conditions under various alarming conditions, different alarms can be notified to corresponding personnel, for example, when a human body is detected, but the movements are detected and the movement amplitude value is not in a normal range, the situation that the respiratory obstruction is caused by sudden illness or head covering of the personnel to lead to coma shock is likely, at the moment, corresponding rescue measures are needed after alarming, and then the medical staff can be notified of the alarming under the conditions, so that the medical staff can timely carry out the rescue measures. Other alarm conditions can be set correspondingly according to actual conditions, and detailed description is omitted here.
In addition, in practical application, the time interval for timing detection can be set according to the practical situation and the use of the computing power resource of the server. When abnormal conditions are monitored according to real-time limb action vital signs of videos acquired by hemispheres A and B in real time and corresponding berths are required to be rechecked by a linked dome camera C, if the dome camera C is in the process of timing heart rate vital sign data monitoring, priority can be set according to actual conditions, for example, timing heart rate vital sign data monitoring can be set to be paused first, recheck is carried out on the corresponding berths first, and timing heart rate vital sign data monitoring is continued after recheck is completed. Of course, whether to suspend timing heart rate sign data monitoring can also be determined according to the number of the to-be-timing monitored bunkers, for example, if only one bunkers need to monitor the timing heart rate sign data, the round of timing monitoring can be completed, and then the corresponding bunkers are checked again.
Through the mode of detecting limb actions in real time and detecting heart rate at fixed time, the method can detect some inconspicuous suicide and self-disabled behaviors, such as the condition of suicidal action by selecting a mode of covering the head and blocking the breath by wet tissues at night when the prison is easy to lose, and the condition of self-disabled by selecting a mode of backing the monitoring camera, and the occurrence of sudden diseases with unobvious symptoms can be detected in time, the method can detect the condition of covering the head and sleeping sense information 4366 in 529 days of certain monitoring run, and confirms the condition of real information 3101, wherein 1 is the condition of covering the head and suicide; 8 pieces of abnormal information of limb movements and heart rate vital sign data are detected, wherein 3 pieces of abnormal information belong to serious sudden diseases (the sudden diseases are diagnosed as coronary heart disease, angina pectoris, hypokalemia and other diseases after urgent treatment by a medical doctor in a monitoring department). The method can accurately alarm the above conditions in time and correct the head covering condition in time; for the emergency, the gold rescue time is striven for, and the accidental death of the people in the presence is avoided.
Embodiment III:
the embodiment provides a vital sign monitoring system based on multi-algorithm fusion coordination, which is applied to a monitoring station, wherein the hardware parts required by the system comprise cameras, a background server, a monitoring display screen and corresponding alarm devices which are arranged in each monitoring room in the monitoring station, and the system can be realized on the basis of the existing monitoring equipment of the monitoring room.
The present embodiment is also described by way of example in the application to the multi-person monitoring room shown in fig. 3.
As shown in fig. 3, 4 and 5, the length and width dimensions of the multi-person monitoring room are generally 15m and 4.5m, and the height is 7 m; usually, a dome camera A and a dome camera B are respectively installed at opposite angles at the position 3 meters higher above the monitoring room, and a dome camera C with a tripod head is installed at the middle position of the long side above the monitoring room.
In practical application, the camera is suitable to be arranged in a place which is not easily damaged by the outside and is near the monitoring target, the installation position does not influence the operation of field equipment and the normal activities of personnel, strong light is prevented from direct irradiation, and the target surface of the camera tube is not damaged. An object which shields the monitoring target must not exist in the view field of the lens, the monitoring target should be aligned from the direction of the light source, and the backlight installation should be avoided; when backlight installation is required, the contrast of the monitored area should be reduced.
Because the monitoring room sizes of different monitoring centers are different, different lens focal lengths and monitoring widths also determine different monitoring distances and camera erection requirements, and the conversion relation between the monitoring distances and the camera installation positions is as follows:
U≈f*W/a
h＝U*tan(13*3.1415926/180)+1.7
senser target size a (mm)
Monitoring width W (Rice)
Monitoring distance U (meter)
The camera mounting heights and lens lookup table are shown in table 1 below:
table 1: camera mounting height and lens lookup table
After the hardware equipment is installed, monitoring a monitoring room in real time by using the dome cameras A and B and the dome camera C with the cradle head, wherein the dome cameras A and B are always started after the related shooting parameters are adjusted, the dome camera C with the cradle head carries out timing inspection on each bed, and the specific inspection time interval can be set according to actual conditions; meanwhile, when videos acquired by the dome cameras A and B in real time are analyzed, if abnormal conditions are detected, the dome camera C with the cradle head can be linked to recheck the corresponding positions.
Specific detection methods can be described in embodiment two. And will not be described in detail herein.
Some steps in the embodiments of the present invention may be implemented by using software, and the corresponding software program may be stored in a readable storage medium, such as an optical disc or a hard disk.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (8)
1. The vital sign detection method based on multi-algorithm fusion coordination is characterized in that the method is applied to vital sign detection of the on-press personnel in a monitoring room, and the method adopts a mode of detecting vital sign data of limb movements in real time and simultaneously assisting in detecting vital sign data of heart rate at fixed time to detect the vital sign of the on-press personnel;
the method comprises the steps of obtaining full coverage video data of a monitoring room through a fixed-mounted dome camera, and obtaining partial coverage video data of the monitoring room through a rotatable spherical camera with adjustable focal length;
the real-time detection of limb movement vital sign data comprises:
detecting whether human bodies exist on all the berths in the monitoring room according to the full coverage video data of the monitoring room acquired by the hemispherical camera;
for the detected body position, acquiring limb actions of the human body by adopting a human skeleton micro action detection algorithm to acquire vital sign data, comparing the vital sign data with the limb action vital sign data corresponding to the position acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the vital sign data of the limb action of the human body;
for the berth where the human body cannot be detected, further detecting whether the human body has actions;
for the bunk that can't detect the human body, further detect whether there is action, include:
if the bunk is detected to have actions, further determining that the bunk is on the head of a person to be pressed according to whether the action change amplitude value is in a normal range or not, and triggering head covering alarm if the action change amplitude value is in the normal range; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if the position is detected to be not moved, acquiring the position on-press personnel information from a monitoring personnel management system in real time, judging whether the position is always unmanned, and simultaneously acquiring the on-press personnel information on duty at night to determine whether the position personnel is on duty; if the bunk is unmanned or the bunk is on the press, preparing for the next detection; if the bunk is someone and the bunk is not proper, the spherical camera which is rotatable in a linkage mode and adjustable in focal length is used for rechecking the bunk.
2. The method of claim 1, wherein the ganged rotatable and focus adjustable spherical camera rechecks the bunk, comprising:
controlling a rotatable spherical camera with adjustable focal length to acquire the bunk video picture, and judging whether a human body exists or not according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length;
if the human body exists in the bunk according to the bunk video picture acquired by the rotatable spherical camera with adjustable focal length, acquiring the limb movement of the human body by adopting a human body skeleton micro-movement detection algorithm to acquire vital sign data, comparing the vital sign data with the limb movement vital sign data corresponding to the bunk acquired by the last detection, and triggering a rescue alarm if the difference value of the vital sign data exceeds the normal value of the limb movement vital sign data of the human body;
if the situation that the human body does not exist in the bunk is judged according to the bunk video picture acquired by the rotatable spherical camera with the adjustable focal length, abnormal alarm is triggered.
3. The method of claim 1, wherein the timing detection of heart rate vital sign data comprises:
controlling a rotatable spherical camera with adjustable focal length to acquire whether the nth berth has a human body or not, wherein N is more than or equal to 1 and less than or equal to N, and N is the number of berths in the monitoring room;
if the nth berth is detected to have a human body, positioning the nth berth to the forehead or the neck of the human body by adopting a target tracking algorithm, and acquiring heart rate data through the skin of the forehead or the neck; if the heart rate data is in the normal range, detecting the next bunk; if the heart rate is not in the normal range, triggering a rescue alarm;
if the nth bunk is detected to have no human body, further detecting whether the nth bunk has actions.
4. A method according to claim 3, wherein if it is detected that the nth bunk is free of human body, further detecting whether the nth bunk is in motion comprises:
if the nth bunk is detected to have the action, further determining that the bunk is on the head of the person to be pressed according to whether the action change amplitude value is in a normal range or not, and triggering head covering alarm if the action change amplitude value is in the normal range; if it is not within the normal range of the device, determining that the bunk is dangerous to the people on the press, and triggering rescue alarm;
if no action of the nth bunk is detected, acquiring information of the bunk on-press personnel from a monitoring personnel management system in real time, judging whether the bunk is not all the time, and simultaneously acquiring information of the on-press personnel on duty at the night to determine whether the bunk personnel are on duty; if the bunk is unmanned or the bunk is on the press, detecting the next bunk; if the berth is occupied and the berth is not worth the person, triggering abnormal alarm.
5. The method according to claim 4, wherein the real-time detecting the vital sign data of the limb movement, if the bunk is detected to be moving, further comprises:
and judging the difference of the video pictures corresponding to the bunk during the front and rear detection by an image detection algorithm, and determining whether the motion change amplitude value is in a normal range according to the difference of the video pictures corresponding to the bunk during the front and rear detection.
6. The method according to claim 5, wherein in the process of detecting heart rate vital sign data at regular time, if the nth bunk is detected to have an action, further according to whether the action change amplitude value is in a normal range, the method comprises:
and judging the difference of the video picture corresponding to the nth bunk detected at the last time through an image detection algorithm, and determining whether the motion variation amplitude value is in a normal range according to the difference of the video picture corresponding to the nth bunk detected at the last time.
7. The vital sign monitoring system based on multi-algorithm fusion coordination is characterized by being applied to monitoring rooms for vital sign detection of on-premise personnel, comprising a semi-spherical camera fixedly installed to acquire monitoring room full coverage video data and a rotatable spherical camera with adjustable focal length to acquire monitoring room partial coverage video data;
the monitoring system adopts the vital sign detection method based on multi-algorithm fusion synergy as claimed in any one of claims 1-6 to detect the vital signs of the on-press personnel in real time.
8. The monitoring system of claim 7, wherein the monitoring system notifies corresponding administrators in different ways for alarms of different situations; the monitoring system records the limb actions and heart rate vital sign data of all monitored on-press personnel in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010894404.5A CN112001346B (en) | 2020-08-31 | 2020-08-31 | Vital sign detection method and system based on multi-algorithm fusion collaboration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010894404.5A CN112001346B (en) | 2020-08-31 | 2020-08-31 | Vital sign detection method and system based on multi-algorithm fusion collaboration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112001346A CN112001346A (en) | 2020-11-27 |
| CN112001346B true CN112001346B (en) | 2023-12-29 |
Family
ID=73464703
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010894404.5A Active CN112001346B (en) | 2020-08-31 | 2020-08-31 | Vital sign detection method and system based on multi-algorithm fusion collaboration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112001346B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113117306A (en) * | 2021-03-19 | 2021-07-16 | 江门职业技术学院 | Yoga auxiliary learning method, system and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106028915A (en) * | 2013-12-19 | 2016-10-12 | 皇家飞利浦有限公司 | A baby monitoring device |
| CN106725383A (en) * | 2016-12-28 | 2017-05-31 | 天津众阳科技有限公司 | Sleep state judgement system and method based on action and heart rate |
| CN110236506A (en) * | 2019-07-12 | 2019-09-17 | 金同磊 | Bad dream interference method and system |
| CN110706826A (en) * | 2019-10-29 | 2020-01-17 | 江苏大学 | Non-contact real-time multi-person heart rate and blood pressure measuring method based on video images |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201601143D0 (en) * | 2016-01-21 | 2016-03-09 | Oxehealth Ltd | Method and apparatus for health and safety monitoring of a subject in a room |
| GB201803508D0 (en) * | 2018-03-05 | 2018-04-18 | Oxehealth Ltd | Method and apparatus for monitoring of a human or animal subject |
-
2020
- 2020-08-31 CN CN202010894404.5A patent/CN112001346B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106028915A (en) * | 2013-12-19 | 2016-10-12 | 皇家飞利浦有限公司 | A baby monitoring device |
| CN106725383A (en) * | 2016-12-28 | 2017-05-31 | 天津众阳科技有限公司 | Sleep state judgement system and method based on action and heart rate |
| CN110236506A (en) * | 2019-07-12 | 2019-09-17 | 金同磊 | Bad dream interference method and system |
| CN110706826A (en) * | 2019-10-29 | 2020-01-17 | 江苏大学 | Non-contact real-time multi-person heart rate and blood pressure measuring method based on video images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112001346A (en) | 2020-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11065958B2 (en) | Control system and method | |
| WO2021253961A1 (en) | Intelligent visual perception system | |
| US9041810B2 (en) | System and method for predicting patient falls | |
| CN109493555A (en) | A kind of campus dormitory building safety defense monitoring system based on intelligent monitoring technology | |
| US20150199892A1 (en) | Noise Correcting Patient Fall Risk State System and Method for Predicting Patient Falls | |
| CN103929592A (en) | All-dimensional intelligent monitoring equipment and method | |
| CN110619735A (en) | System for monitoring and alarming falling object | |
| CN107067612B (en) | Portable infrared alarm and wireless video monitoring system and monitoring method | |
| CN108234926A (en) | A kind of human behavior discriminance analysis system | |
| CN112001346B (en) | Vital sign detection method and system based on multi-algorithm fusion collaboration | |
| US20110037850A1 (en) | Surveillance system for monitoring security personnel | |
| CN112489371A (en) | Swimming pool drowning prevention early warning system based on computer vision | |
| US11403754B2 (en) | Method and apparatus for monitoring of a human or animal subject | |
| WO2017012266A1 (en) | Beating-prevention automatic alarm and monitoring system and method for use on prison inmate | |
| CN108012075A (en) | The integrated control method and system of a kind of monitoring device and lighting apparatus | |
| WO2019206175A1 (en) | Method and device for intelligent monitoring and caring for alzheimer's patients | |
| CN111526280A (en) | Control method and device of camera device, electronic equipment and storage medium | |
| JP2000184359A (en) | Monitoring device and system therefor | |
| US20180011982A1 (en) | Infrared detectors and thermal tags for real-time activity monitoring | |
| CN205961324U (en) | Formula infrared ray video monitor system is stored up in cabin | |
| JP2008020981A (en) | Monitoring system | |
| US10909678B2 (en) | Method and apparatus for monitoring of a human or animal subject | |
| CN111210602A (en) | Night risk prevention and control method, system and storage medium | |
| CN114220244B (en) | Home-based care risk detection system and method with front study and judgment | |
| CN116016857A (en) | Abnormal behavior early warning system based on big data |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230803 Address after: Room 120-1, Block C, Innovation and Creative Industry Park, No. 5 Xinhua Road, Wuxi City, Jiangsu Province, 214000 Applicant after: Jiangsu zhengdehou Internet of things Technology Development Co.,Ltd. Address before: Room 120-1, Block C, Innovation and Creative Industry Park, No. 5 Xinhua Road, Wuxi City, Jiangsu Province, 214000 Applicant before: Jiangsu zhengdehou Internet of things Technology Development Co.,Ltd. Applicant before: Zhou Jincheng Applicant before: Cheng Zhi |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |