CN113516008A - Human body movement abnormity monitoring system based on human body skeleton key points - Google Patents
Human body movement abnormity monitoring system based on human body skeleton key points Download PDFInfo
- Publication number
- CN113516008A CN113516008A CN202110369221.6A CN202110369221A CN113516008A CN 113516008 A CN113516008 A CN 113516008A CN 202110369221 A CN202110369221 A CN 202110369221A CN 113516008 A CN113516008 A CN 113516008A
- Authority
- CN
- China
- Prior art keywords
- human body
- time
- posture
- body target
- real
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Alarm Systems (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The invention discloses a human body movement abnormity monitoring system based on human body skeleton key points, which detects a human body target in a monitoring image in real time, detects each skeleton key point of the human body target, determines the real-time movement posture of the human body target according to the real-time relative position relation of each skeleton key point of the human body target, generates habitual movement posture data of the human body target according to the movement posture data of the human body target in the past period of time, and automatically compares the movement posture data of the human body target in the near period of time with the habitual movement posture data to judge whether the movement posture data of the human body target is abnormal or not, thereby judging whether the human body target is abnormal or not in the near period of time. The human body activity abnormity monitoring system is applied to daily monitoring of the elderly, whether the daily activity of the elderly is abnormal or not can be automatically judged, the workload of the guardian is reduced, and the efficiency and the quality of health monitoring of the elderly are improved.
Description
Technical Field
The invention relates to a machine vision monitoring technology, in particular to a human body movement abnormity monitoring system based on human body skeleton key points.
Background
China has gradually entered into an aging society, and the aging of the population makes the health monitoring problem of the elderly increasingly prominent. In the aspect of health monitoring of the old, the prior art mainly utilizes video monitoring equipment to carry out video monitoring on the daily life of the old, and then the monitor knows the daily life condition of the old through monitoring the video. This kind of mode needs guardians to judge whether the daily life of old person is unusual through watching the artifical judgement of surveillance video, can not carry out automatic judgement to whether the daily life of old person is unusual or not, has increased guardianship people's guardianship work load, has influenced guardianship efficiency and quality.
Disclosure of Invention
The invention mainly aims to provide a human body activity abnormity monitoring system to solve the problem that whether the daily life of the old people is abnormal or not cannot be automatically judged in the conventional mode of monitoring the daily life of the old people by using a monitoring video.
The invention is realized by the following technical scheme:
a human body movement abnormity monitoring system based on human body skeleton key points comprises:
the human body target detection module is used for detecting a human body target in the monitoring image in real time;
the skeleton key point detection module is used for detecting each skeleton key point of the human body target in real time;
the motion posture determining module is used for determining the real-time motion posture of the human body target according to the real-time relative position relation of all the bone key points of the human body target;
and the abnormal behavior monitoring module is used for comparing activity posture data of the human body target from the time t1 to the time t0 with habitual activity posture data of the human body target to judge whether the human body target is abnormal, wherein the habitual activity posture data are generated according to the activity posture data of the human body target from the time t2 to the time t1, the time t0 is the current time, the time t1 is a first preset time before the time t0, and the time t2 is a second preset time before the time t 1.
Further, the habitual movement posture data comprises single average duration data of various movement postures of the human target from the time t2 to the time t 1; the abnormal behavior monitoring module is specifically configured to determine that the human body target is abnormal when it is detected that the single duration of any one activity posture of the human body target from the time t1 to the time t0 exceeds the single average duration of the activity posture of the human body target from the time t2 to the time t1 by a preset proportion.
Further, the time t1 is a day before the time t0, and the time t2 is a week or month before the time t 1.
Furthermore, the movable postures comprise sitting postures, standing postures, sleeping postures, walking postures, squatting postures and toileting.
Further, the system for monitoring human body activity abnormity based on human body skeleton key points further comprises:
and the alarm module is used for sending abnormal information to the mobile terminal of the guardian of the human body target when the human body target is abnormal.
Further, a world coordinate system is preset in the human body movement abnormity monitoring system based on the human body skeleton key points, a plurality of cameras are arranged in a monitoring area corresponding to the monitoring image, the ground of the monitoring area and the positions of the cameras in the world coordinate system are calibrated, and the human body movement abnormity monitoring system further comprises:
the position calculation module is used for calculating the real-time position of the human body target in the world coordinate system according to the real-time distance between each camera and the human body target;
the monitoring area is provided with furniture or sanitary wares, the positions of the furniture or sanitary wares in the world coordinate system are calibrated, and the moving posture determining module determines the real-time moving posture of the human body target according to the real-time relative position relation of the skeletal key points of the human body target, the real-time position of the human body target in the world coordinate system and the position range of the furniture or sanitary wares in the world coordinate system.
Further, the furniture or the sanitary appliance comprises a bed, and the activity posture determination module determines that the real-time activity posture of the human body target is the sleeping posture only when the real-time position of the human body target in the world coordinate system is within the position range of the bed in the world coordinate system and the real-time relative position relation of each bone key point of the human body target accords with the sleeping posture characteristic.
Further, the furniture or sanitary appliance comprises furniture for sitting, and the activity posture determination module determines that the real-time activity posture of the human body target is a sitting posture only when the real-time position of the human body target in the world coordinate system is within the position range of the furniture for sitting in the world coordinate system and the real-time relative position relation of each bone key point of the human body target conforms to the sitting posture characteristic.
Further, the furniture or the sanitary appliance comprises a toilet stool, and the activity posture determination module determines that the real-time activity posture of the human body target is toilet-like only when the real-time position of the human body target in the world coordinate system is located in the position range of the toilet stool in the world coordinate system and the real-time relative position relation of all skeletal key points of the human body target conforms to toilet-like characteristics.
Compared with the prior art, the human body movement abnormity monitoring system based on the human body skeleton key points provided by the invention detects the human body target in the monitoring image in real time, detects each skeleton key point of the human body target, determines the real-time movement posture of the human body target according to the real-time relative position relation of each skeleton key point of the human body target, generates habitual movement posture data of the human body target according to the movement posture data of the human body target in the past period of time, and automatically compares the movement posture data of the human body target in the near period of time with the habitual movement posture data to judge whether the movement posture data of the human body target is abnormal or not, thereby judging whether the human body target is abnormal or not in the near period of time. The human body activity abnormity monitoring system is applied to daily monitoring of the elderly, whether the daily activity of the elderly is abnormal or not can be automatically judged, the workload of the guardian is reduced, and the efficiency and the quality of health monitoring of the elderly are improved.
Drawings
FIG. 1 is a schematic diagram of the connection of a server, an artificial intelligence computing platform, a switch, and a camera;
FIG. 2 is a schematic diagram of a configuration of a human body movement abnormality monitoring system based on human body skeleton key points;
fig. 3 is a schematic diagram of the principle of the sleep posture determination by combining the real-time position of the human target in the world coordinate system and the position range of the bed in the world coordinate system.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the following embodiments and the accompanying drawings.
The human body movement abnormity monitoring system based on the human body skeleton key points provided by the embodiment of the invention can be used for monitoring the movement abnormity conditions of the human body targets in the monitoring area 17. The abnormal activities in the present invention do not mean major abnormalities such as falls and sudden diseases, but mean that the habits of daily living in a short period of time are different from the past habits, and are called abnormal activities. For example, the usual toilet-using time is generally 5-10 minutes, the toilet-using time in the near period is obviously longer than 10 minutes, for example, the usual sleeping time is generally 7-8 hours, the last sleeping time is obviously longer than 8 hours, and the like, which can be called as abnormal activity. The monitoring area 17 may be the following: hospitals, nursing homes, indoor home environments, nursing homes, and the like. Generally, one location may be a single monitoring area 17, and for a relatively large location, the location may be divided into a plurality of monitoring areas 17.
The human body activity abnormity monitoring system mainly comprises a human body target detection module 31, a bone key point detection module 32, an activity posture determination module 41, an abnormal behavior monitoring module 42, an alarm module 43 and the like. The human body target detection module 31 and the bone key point detection module 32 are deployed on the artificial intelligence computing platform 3, and are mainly used for performing feature detection, feature recognition and the like on a human body target in a monitored image. The activity posture determining module 41, the abnormal behavior monitoring module 42, the alarm module 43 and the like are deployed on the server 4, and are mainly used for performing calculation analysis on relevant features of the human body target image detected and recognized by the artificial intelligent computing platform 3 to determine the activity posture of the human body target. The server 4 can be deployed at the cloud end and is connected and communicated with the artificial intelligence computing platform 3 in a mobile internet or wired internet mode and the like, and the camera 1 can be connected and communicated with the artificial intelligence computing platform 3 through the switch 2.
The camera 1 obtains a monitoring image by taking an image of the monitoring area 17. The human target detection module 31 is used for detecting a human target in the monitored image in real time. The camera 1 can adopt a camera supporting an infrared night vision function, so that shooting at night when illumination is poor is facilitated. Besides the human body target image, other images exist in the monitoring image, and the images except the human body target are called background images. The human body target detection is a process of extracting a human body target image from a monitored image, and a human body target detection algorithm is a mature artificial intelligence algorithm at present and is not described herein again. In order to be able to detect human targets located at the corners of the monitored area 17, the cameras 1 may be respectively arranged at the corners of the monitored area 17, and each human target in the monitored area 17 is ensured to be monitored by shooting the monitored area 17 from different angles.
The bone key point detection module 32 is used for detecting each bone key point of the human body target in real time. Human skeleton key point detection is the basis of visual tasks such as analyzing activity postures and behaviors. The human skeleton key points mainly comprise main joint parts such as the vertex, the five sense organs, the neck, the trunk, the four limbs and the like, and the detection and the identification of the human skeleton key points can be realized by detecting and analyzing the two-dimensional monitoring image through a human skeleton key point detection algorithm.
The activity posture determination module 41 is configured to determine a real-time activity posture of the human body target according to a real-time relative position relationship of each bone key point of the human body target. The activity posture mainly represents the posture or action of the human body target, and does not represent the state that the human body target must be in activity currently. The posture of the human target in the inactive posture is also an active posture, for example the posture in sleep is also an active posture. In the present invention, the activity posture includes, but is not limited to, sitting posture, standing posture, sleeping posture, walking posture, squatting posture, toileting posture, etc. The real-time relative position relationship of each bone key point of the human body target is different under different activity postures, so that the real-time relative position relationship (for example, the relative position height of each bone key point) of each bone key point of the human body target can be calculated according to the real-time position of each bone key point of the human body target detected by the bone key point detection module 32 to determine the real-time activity posture of the human body target. The corresponding relation data between various moving postures of the human body and the relative position relation of the bone key points can be prestored, then the real-time relative position relation of the bone key points of the human body target is detected, and the real-time moving posture of the human body target is determined according to the prestored corresponding relation. The moving posture of the human body target can be recorded in real time according to a time axis, and the moving posture of the human body target at each moment is recorded in real time from the moment of recording the moving posture of the human body target, so that the moving posture of the human body target at the moment can be obtained by selecting any moment, and the moving posture data of the human body target in the past period can be obtained.
The abnormal behavior monitoring module 42 is configured to compare the activity posture data of the human body target from the time t1 to the time t0 with the habitual activity posture data of the human body target, so as to determine whether the human body target is abnormal. The habitual activity posture data is generated according to activity posture data of the human target from time t2 to time t1, wherein time t0 is the current time, time t1 is a time which is a first preset time before time t0, and time t2 is a time which is a second preset time before time t 1. The invention analyzes the abnormal movement condition of the human body target in the near period of time, and the abnormal movement condition of the human body target in the near period of time is obtained by comparing the movement posture data of the human body target in the near period of time with the habitual movement posture data of the human body target. The habitual activity posture data of the human body target are generated according to the activity posture data of the human body target in a longer period of time before the near period of time, and the habitual activity habit of the human body target in the longer period of time before the near period of time is reflected.
For example, the following steps are carried out: time t0 is time 10 at day zero 12/month in 2020, time t1 is time one day before time t0, that is, time 9 at day zero 12/month in 2020, and time t2 is time 1 week or 1 month before time t1, that is, time 2 at day zero 12/month in 2020 or time 9 at day zero 11/month in 2020. First, the abnormal behavior monitoring module 42 obtains habitual movement posture data for comparison with movement posture data of the human body target between 12/month/9 zero of 2020 and 12/month/9 zero of 2020 or between 11/month/9 zero of 2020 and 12/month/9 zero of 2020, from the movement posture data of the human body target between 12/month/9 zero of 2020 and 10/month/10 of 2020. Then, the abnormal behavior monitoring module 42 compares the habitual activity posture data with the activity posture data of the human target between zero point of 12/9/2020 and zero point of 12/10/2020, to determine whether the activity of the human target between zero point of 12/9/2020 and zero point of 12/10/2020 is abnormal. It can be seen that the habitual activity posture data is continuously updated as the current time t0 advances.
In terms of what the habitual activity posture data includes, the habitual activity posture data may include single average duration data of various activity postures of the human target from time t2 to time t 1. In this case, the abnormal behavior monitoring module is specifically configured to determine that the human target is abnormal when it is detected that the single duration of any one of the activity postures of the human target from the time t1 to the time t0 exceeds the single average duration of the activity postures of the human target from the time t2 to the time t1 by a preset ratio. Taking the time in the above example as a basis, taking the toilet duration of the old person as an example, when the duration of any one toilet of the old person between zero point of 12/9 th day in 2020 to zero point of 10 th day in 2020 exceeds the average duration of a single toilet of the old person between zero point of 2/2 th day in 2020 to zero point of 9/9 th day in 2020 or between zero point of 9/9 th day in 11/11 th in 2020 to zero point of 9 th day in 2020 to 12 th day in 2020, the toilet duration of the old person is abnormal.
Under some special conditions, if the moving posture of the human body target is simply judged according to the real-time relative position relation of all the bone key points, the misjudgment can be generated. For example, when the human target is not actually seated, only the buttocks are suspended to perform an action corresponding to the sitting posture. For example, when the human target is doing a campstool step, the real-time relative position relationship of the key points of each skeleton will also present a feature corresponding to a sitting posture. To avoid similar misjudgment problems, the activity posture of the human target can be judged in combination with the positions of the furniture or the sanitary wares in the monitoring area 17. Specifically, a world coordinate system may be preset in the human body movement abnormality monitoring system, and then a plurality of cameras 1 may be set to photograph the monitored area 17, where the ground of the monitored area 17 and the positions of the cameras 1 in the world coordinate system are calibrated. On the basis, the human body activity abnormity monitoring system further comprises a position calculation module 9. The position calculating module 9 is configured to calculate a real-time position of the human body target in the world coordinate system according to the real-time distance between each of the cameras 1 and the human body target. Meanwhile, the positions of the furniture or the sanitary wares in the world coordinate system are calibrated in advance, and then the moving posture determining module 41 determines the real-time moving posture of the human body target according to the real-time relative position relationship of the skeletal key points of the human body target, the real-time position of the human body target in the world coordinate system and the position range of the furniture or the sanitary wares in the world coordinate system. If the position of the human body target in the world coordinate system is not located in the position range of the furniture or the sanitary appliance in the world coordinate system, the current real-time moving posture of the human body target cannot be judged to be the human body moving posture corresponding to the furniture or the sanitary appliance. In the present invention, the furniture or sanitary equipment includes furniture and beds 16 such as seats and sofas for sitting, and toilets such as toilet bowls and squatting pans. In the present invention, the human body movement posture corresponding to the bed 16 is a sleeping posture, the human body movement posture corresponding to furniture for sitting (a seat, a sofa, etc.) is a sitting posture, and the human body movement posture corresponding to a toilet stool (a toilet bowl or a squatting pan) is toileting. The position range of the furniture or the sanitary appliance refers to the space position range occupied by the furniture or the sanitary appliance in the world coordinate system. For example, the furniture or the sanitary appliance can be regarded as a two-dimensional figure, the minimum bounding rectangle of the two-dimensional figure is regarded as a two-dimensional space (O-X-Y plane) occupied by the furniture or the sanitary appliance, coordinates of four corners of the minimum bounding rectangle in a world coordinate system are calibrated in advance (as shown in the figure, the four corners of the minimum bounding rectangle of the bed 16 are A, B, C, D four points respectively), and then a rectangular area defined by the coordinates of the four corners mapped to the O-X-Y plane is regarded as a position range of the furniture or the sanitary appliance in the world coordinate system. And the position of the human body target in the world coordinate system is determined by virtualizing the human body target into a point and detecting the position of the point in the world coordinate system. The joint point connecting the trunk and the lower limb in each skeleton key point of the human body target can be used as the virtual point of the human body target, such as the point P in the figure.
Specifically, for the sleeping posture, the active posture determining module 41 determines the real-time active posture of the human target as the sleeping posture only when the real-time position of the human target in the world coordinate system is within the position range of the bed 16 in the world coordinate system and the real-time relative position relationship of the skeletal key points of the human target conforms to the sleeping posture characteristics. Taking the illustration as an example, when the coordinate of the real-time position mapped to the O-X-Y plane of the point P is located within the rectangular area defined by the coordinates mapped to the O-X-Y plane of the four points A, B, C, D, it indicates that the real-time position of the human target in the world coordinate system is located within the position range of the bed 16 in the world coordinate system, and on the premise, if the real-time relative position relationship of the bone key points of the human target also conforms to the sleeping posture feature, the activity posture determination module 41 determines the real-time activity posture of the human target to be the sleeping posture.
For the sitting posture, the activity posture determination module 41 determines that the real-time activity posture of the human body target is the sitting posture only when the real-time position of the human body target in the world coordinate system is within the position range of the furniture for sitting in the world coordinate system and the real-time relative position relationship of the bone key points of the human body target conforms to the sitting posture characteristics.
For toilet, the activity posture determination module 41 determines that the real-time activity posture of the human body target is toilet only when the real-time position of the human body target in the world coordinate system is within the position range of the toilet stool in the world coordinate system and the real-time relative position relation of the skeletal key points of the human body target conforms to the toilet characteristics.
The judgment of sitting posture and toilet posture by combining the real-time position of the human body target in the world coordinate system and the position range of furniture and sanitary wares used for sitting, such as seats, sofas and the like in the world coordinate system is the same as the judgment of the sleeping posture. On the premise of analyzing the real-time relative position relation of each skeleton key point of the human body target, the real-time moving posture of the human body target can be more accurately judged by combining the real-time position of the human body target in the world coordinate system and the position range of each furniture or sanitary appliance in the world coordinate system.
The human body activity abnormity monitoring system based on the human body skeleton key points can also comprise an alarm module 43, wherein the alarm module 43 is used for sending abnormal information to the mobile terminal 5 of the guardian of the human body target when the human body target is abnormal. Therefore, the guardian of the human body target can conveniently know the abnormal condition of the human body target in time. For example, when the fact that the latest toilet duration of the old people is abnormal is detected, corresponding abnormal information can be sent to the mobile phone of the guardian of the old people in time, and the guardian can conveniently know the relevant conditions in time.
The above-described embodiments are merely preferred embodiments, which are not intended to limit the scope of the present invention, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (9)
1. A human body movement abnormity monitoring system based on human body skeleton key points is characterized by comprising:
the human body target detection module is used for detecting a human body target in the monitoring image in real time;
the skeleton key point detection module is used for detecting each skeleton key point of the human body target in real time;
the motion posture determining module is used for determining the real-time motion posture of the human body target according to the real-time relative position relation of all the bone key points of the human body target;
and the abnormal behavior monitoring module is used for comparing activity posture data of the human body target from the time t1 to the time t0 with habitual activity posture data of the human body target to judge whether the human body target is abnormal, wherein the habitual activity posture data are generated according to the activity posture data of the human body target from the time t2 to the time t1, the time t0 is the current time, the time t1 is a first preset time before the time t0, and the time t2 is a second preset time before the time t 1.
2. The system for monitoring abnormality in human activity based on key points of human bones as set forth in claim 1, wherein the habitual activity posture data includes a single average duration data of various activity postures of the human target from time t2 to time t 1; the abnormal behavior monitoring module is specifically configured to determine that the human body target is abnormal when it is detected that the single duration of any one activity posture of the human body target from the time t1 to the time t0 exceeds the single average duration of the activity posture of the human body target from the time t2 to the time t1 by a preset proportion.
3. The system for monitoring abnormality of human body activity based on key points of human bones as claimed in claim 1, wherein the time t1 is a time one day before the time t0, and the time t2 is a time 1 week or 1 month before the time t 1.
4. The system for monitoring abnormal human activities based on key points of human bones as claimed in claim 1, wherein the activities postures comprise sitting posture, standing posture, sleeping posture, walking posture, squatting posture and toileting posture.
5. The system for monitoring abnormality in human body activity based on key points of human bones as set forth in any one of claims 1 to 4, further comprising:
and the alarm module is used for sending abnormal information to the mobile terminal of the guardian of the human body target when the human body target is abnormal.
6. The system for monitoring abnormal human body activities based on key points of human bones as claimed in claim 1, wherein a world coordinate system is preset in the system for monitoring abnormal human body activities based on key points of human bones, a plurality of cameras are arranged in a monitored area corresponding to the monitored image, the ground of the monitored area and the positions of the cameras in the world coordinate system are calibrated, and the system for monitoring abnormal human body activities further comprises:
the position calculation module is used for calculating the real-time position of the human body target in the world coordinate system according to the real-time distance between each camera and the human body target;
the monitoring area is provided with furniture or sanitary wares, the positions of the furniture or sanitary wares in the world coordinate system are calibrated, and the moving posture determining module determines the real-time moving posture of the human body target according to the real-time relative position relation of the skeletal key points of the human body target, the real-time position of the human body target in the world coordinate system and the position range of the furniture or sanitary wares in the world coordinate system.
7. The human body movement anomaly monitoring system based on human body skeletal key points according to claim 6, wherein the furniture or sanitary appliance comprises a bed, and the movement posture determination module determines the real-time movement posture of the human body target as a sleeping posture only when the real-time position of the human body target in the world coordinate system is within the position range of the bed in the world coordinate system and the real-time relative position relationship of the skeletal key points of the human body target conforms to the sleeping posture characteristics.
8. The human activity anomaly monitoring system based on human skeletal key points of claim 6, wherein the furniture or sanitary appliances comprise furniture for sitting, and the activity posture determination module determines the real-time activity posture of the human target as a sitting posture only when the real-time position of the human target in the world coordinate system is within the range of positions of the furniture for sitting in the world coordinate system and the real-time relative positional relationship of the skeletal key points of the human target conforms to a sitting posture characteristic.
9. The human body movement anomaly monitoring system based on human body skeletal key points according to claim 6, wherein the furniture or sanitary appliance comprises a toilet stool, and the movement posture determination module determines the real-time movement posture of the human body target as toilet-like only when the real-time position of the human body target in the world coordinate system is within the position range of the toilet stool in the world coordinate system and the real-time relative position relationship of the skeletal key points of the human body target conforms to toilet-like characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110369221.6A CN113516008A (en) | 2021-04-06 | 2021-04-06 | Human body movement abnormity monitoring system based on human body skeleton key points |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110369221.6A CN113516008A (en) | 2021-04-06 | 2021-04-06 | Human body movement abnormity monitoring system based on human body skeleton key points |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113516008A true CN113516008A (en) | 2021-10-19 |
Family
ID=78062252
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110369221.6A Withdrawn CN113516008A (en) | 2021-04-06 | 2021-04-06 | Human body movement abnormity monitoring system based on human body skeleton key points |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113516008A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187561A (en) * | 2021-11-30 | 2022-03-15 | 广西世纪创新显示电子有限公司 | Abnormal behavior identification method and device, terminal equipment and storage medium |
| CN117789406A (en) * | 2024-02-27 | 2024-03-29 | 南方电网调峰调频发电有限公司 | Tunnel safety management method based on artificial intelligence |
-
2021
- 2021-04-06 CN CN202110369221.6A patent/CN113516008A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187561A (en) * | 2021-11-30 | 2022-03-15 | 广西世纪创新显示电子有限公司 | Abnormal behavior identification method and device, terminal equipment and storage medium |
| CN117789406A (en) * | 2024-02-27 | 2024-03-29 | 南方电网调峰调频发电有限公司 | Tunnel safety management method based on artificial intelligence |
| CN117789406B (en) * | 2024-02-27 | 2024-05-03 | 南方电网调峰调频发电有限公司 | Tunnel safety management method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10904492B2 (en) | Video monitoring system | |
| WO2020253162A1 (en) | Robot and control method therefor, and intelligent home control system | |
| US10095930B2 (en) | System and method for home health care monitoring | |
| CN109887238B (en) | Tumble detection system and detection alarm method based on vision and artificial intelligence | |
| CN113516008A (en) | Human body movement abnormity monitoring system based on human body skeleton key points | |
| CN109558865A (en) | A kind of abnormal state detection method to the special caregiver of need based on human body key point | |
| JP6520831B2 (en) | Display control apparatus, display control system, display control method, display control program, recording medium | |
| US20020044682A1 (en) | Method and apparatus for subject physical position and security determination | |
| US20150302310A1 (en) | Methods for data collection and analysis for event detection | |
| CN111491426A (en) | Intelligent light control system and control method | |
| CN102387345A (en) | Safety monitoring system based on omnidirectional vision for old people living alone | |
| Zouba et al. | A computer system to monitor older adults at home: Preliminary results | |
| JP6115692B1 (en) | Behavior detecting device, method and program, and monitored person monitoring device | |
| JP2019020993A (en) | Watching support system and method for controlling the same | |
| JP2011090408A (en) | Information processor, and action estimation method and program of the same | |
| WO2019021744A1 (en) | Alarm control system, detection unit, care support system, and alarm control method | |
| JP2024038108A (en) | Information processing device, watching system, and control program | |
| JP6142975B1 (en) | Monitored person monitoring apparatus and method, and monitored person monitoring system | |
| CN108846996B (en) | Tumble detection system and method | |
| GB2593931A (en) | Person monitoring system and method | |
| WO2020008726A1 (en) | Target object detection program and target object detection device | |
| CN113139440A (en) | Human body activity monitoring system | |
| JP7314939B2 (en) | Image recognition program, image recognition device, learning program, and learning device | |
| Andreoni et al. | Design and ergonomics of monitoring system for elderly | |
| Valentin | Multisensor fusion for monitoring elderly activities at home |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211019 |