CN212752473U - Restless behavior early warning system based on deep learning and RASS scoring - Google Patents
Restless behavior early warning system based on deep learning and RASS scoring Download PDFInfo
- Publication number
- CN212752473U CN212752473U CN202021987274.1U CN202021987274U CN212752473U CN 212752473 U CN212752473 U CN 212752473U CN 202021987274 U CN202021987274 U CN 202021987274U CN 212752473 U CN212752473 U CN 212752473U
- Authority
- CN
- China
- Prior art keywords
- wall
- rass
- early warning
- warning system
- deep learning
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Emergency Alarm Devices (AREA)
Abstract
The utility model discloses a restless behavior early warning system based on deep learning and RASS scoring, which comprises a bottom plate and a mounting seat, wherein the mounting seat is arranged on the outer wall of the upper end of the bottom plate, the outer wall of the upper end of the mounting seat is provided with a stand column, the stand column is provided with a fixed ring block by the outer wall of the circumference of the upper side, when dust is cleaned, the rotating rod is repeatedly rotated back and forth through the rotating rod, and the wiping sleeve is driven to rotate together, so that the dust on the lens can be wiped by the wiping sleeve, when the rotating rod stops working, the rotating rod and the wiping sleeve are pulled aside by the pulling force of a spring, the lens is prevented from being blocked, meanwhile, the wiping sleeve can be taken down from the rotating rod, then the cleaning is carried out, the repeated use is convenient, when the wiping sleeve is taken out, only the mounting knob is required to be rotated, the labor amount of workers is reduced, the behaviors and the faces of the patients are effectively monitored, and alarm information can be provided in time.
Description
Technical Field
The utility model belongs to the technical field of video monitoring is relevant, concretely relates to restless action early warning system based on degree of depth learning and RASS score.
Background
The video monitoring is an important component of a safety precaution system, the traditional monitoring system comprises a front-end camera, a transmission cable and a video monitoring platform, the camera can be divided into a network digital camera and an analog camera and can be used for collecting front-end video image signals, the system is a comprehensive system with strong precaution capacity, the video monitoring is visual, accurate, timely and rich in information content and is widely applied to many occasions, in recent years, with the rapid development of computers, networks, image processing and transmission technologies, the video monitoring technology is greatly developed, the latest monitoring system can be used as a smart phone, and meanwhile, automatic identification, storage and automatic alarm are carried out on images. The video data are transmitted back to the control host (or the smart phone can act as) through the 3G/4G/WIFI, and the host can perform operations such as real-time watching, recording, playback, calling out and storage on the images, so that mobile internet video monitoring is realized.
The existing video monitoring technology has the following problems: at present, a camera in video monitoring needs to monitor a patient or a person with abnormal behavior for a long time, so that dust is easily contaminated on a lens, the condition of misjudgment is easily caused, and analysis and alarm cannot be timely performed.
SUMMERY OF THE UTILITY MODEL
An object of the utility model is to provide a restless action early warning system based on degree of depth study and RASS grade to the camera of proposing among the above-mentioned background art at present video monitoring easily makes the dust be infected with the problem of more easily taking place the erroneous judgement on the camera lens is solved.
In order to achieve the above object, the utility model provides a following technical scheme: an agitation behavior early warning system based on deep learning and RASS scoring comprises a bottom plate and a mounting seat, wherein the mounting seat is mounted on the outer wall of the upper end of the bottom plate, the outer wall of the upper end of the mounting seat is provided with a stand column, the stand column is provided with a fixed ring block by the outer wall of the circumference of the upper side, the outer wall of the circumference of the fixed ring block is provided with a connecting rod, the outer wall of the front end of the connecting rod is provided with a placing box, a monitor is arranged inside the outer wall of the front end of the monitor, a lens is arranged inside the outer wall of the front end of the placing box, the outer wall of the front end of the monitor is provided with a case, the outer wall of the front end of the monitor is provided with a wiping sleeve, the outer wall of the lower end of the wiping sleeve is provided with a mounting knob, the inside of the case, the inside installation screw that is provided with of rotary rod lower extreme outer wall, the inside of installation screw is provided with the installation screw rod, be provided with the fixed plate on the frame on spindle right side, fixed plate and rotary rod pass through spring coupling, be provided with the cross slot on placing the case under the rotary rod, driving motor and watch-dog are all through quick-witted incasement portion power electric connection.
Preferably, the outer wall of the lower end of the bottom plate is provided with universal wheels.
Preferably, a battery cover is arranged on the upper portion of the outer wall of the rear end of the case, and a battery is arranged inside the outer wall of the rear end of the case.
Preferably, the wiping sleeve is made of a cylindrical cloth material, and the wiping sleeve is sleeved on the circumferential outer wall of the rotating rod.
Preferably, the rotating rod is disposed inside the transverse slot.
Preferably, the installation knob is provided with anti-skidding line on the circumference outer wall, the diameter of installation knob is greater than the diameter of swiveling lever.
Compared with the prior art of video monitoring, the utility model provides a restless action early warning system based on degree of depth study and RASS score possesses following beneficial effect:
the utility model discloses video monitoring is planned to be based on 5G high-speed network and facial discernment, motion identification equipment, carry out the integrated analysis based on the characteristic information of limbs action, facial expression two dimensions, constitute early yellow early warning and red warning to patient's restlessness, delirium incident identification, utilize simultaneously and clean the camera lens on this watch-dog in the use, effectively clear up the dust, when clearing up the dust, through the rotary rod back and forth rotation, the rotary rod drives the cleaning cover on its outer wall and also rotates together, utilize the cleaning cover to clean the dust on the camera lens like this, and when the rotary rod stops working, utilize the pulling force of spring to pull rotary rod and cleaning cover aside, avoid blocking the camera lens, this cleaning cover also can be taken off from the rotary rod simultaneously, then will wash, be convenient for used repeatedly, and when taking out the cleaning cover, only need with the installation knob rotatory and take down can, the work of cleaning before this camera lens is independently accomplished, need not the manual work and cleans, has reduced staff's the amount of labour, and effectual action and the facial monitoring to patient can in time provide alarm information.
Drawings
The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention, and together with the description, do not constitute a limitation of the invention, in which:
fig. 1 is a schematic view of a front three-dimensional structure of an agitation behavior early warning system based on deep learning and RASS scoring;
fig. 2 is a schematic view of a video surveillance front plane structure provided by the present invention;
fig. 3 is a schematic view of a video monitoring front parsing structure provided by the present invention;
fig. 4 is a schematic view of a front partial parsing structure of the video monitor proposed by the present invention;
in the figure: 1. a base plate; 2. a mounting seat; 3. a column; 4. a fixed ring block; 5. a connecting rod; 6. a chassis; 7. a monitor; 8. a lens; 9. placing a box; 10. wiping the sleeve; 11. installing a knob; 12. rotating the rod; 13. a drive motor; 14. a fixing plate; 15. a spring; 16. a transverse groove; 17. a crankshaft; 18. a machine base; 19. installing a screw rod; 20. and installing screw holes.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments in the present invention, all other embodiments obtained by a person skilled in the art without creative work belong to the protection scope of the present invention.
Referring to fig. 1-4, the present invention provides a technical solution: an agitation behavior early warning system based on deep learning and RASS scoring comprises a bottom plate 1 and a mounting seat 2, wherein the mounting seat 2 is mounted on the outer wall of the upper end of the bottom plate 1, an upright post 3 is arranged on the outer wall of the upper end of the mounting seat 2, a fixed ring block 4 is arranged on the outer wall of the circumference of the upper side of the upright post 3, a connecting rod 5 is arranged on the outer wall of the circumference of the fixed ring block 4, a placing box 9 is arranged on the outer wall of the front end of the connecting rod 5, a monitor 7 is arranged in the outer wall of the front end of the placing box 9, a lens 8 is arranged in the outer wall of the front end of the monitor 7, a case 6 is arranged on the outer wall of the upper end of the placing box 9, a wiping sleeve 10 is arranged on the outer wall of the front end of the monitor 7, a mounting knob 11 is arranged on the outer wall of the lower end of the wiping sleeve 10, the inside installation screw 20 that is provided with of rotary rod 12 lower extreme outer wall, the inside of installation screw 20 is provided with installation screw 19, is provided with fixed plate 14 on the frame 18 on spindle 17 right side, and fixed plate 14 and rotary rod 12 pass through spring 15 to be connected, are provided with horizontal groove 16 on placing the case 9 under rotary rod 12, and driving motor 13 and watch-dog 7 all pass through quick-witted case 6 internal power electrical connection.
An agitation behavior early warning system based on deep learning and RASS scoring comprises a bottom plate 1 and a mounting seat 2, wherein universal wheels are arranged on the outer wall of the lower end of the bottom plate 1, so that a video monitoring device can be conveniently transferred by utilizing the universal wheels and can be conveniently carried, a battery cover is arranged on the upper portion of the outer wall of the rear end of a case 6, a battery is arranged in the outer wall of the rear end of the case 6, so that a driving motor 13 and a monitor 7 can be conveniently supplied with power through the battery, and when the video monitoring device is used or carried, the interference caused by electric wires or the use production place cannot be met, a wiping sleeve 10 is made of cylindrical cloth materials, the wiping sleeve 10 is sleeved on the circumferential outer wall of a rotating rod 12, so that the lens 8 can be cleaned through the wiping sleeve 10, the dust on the lens 8 can be cleaned conveniently, the monitoring and the alarming for a patient can be conveniently carried, make rotary rod 12 can rotate in horizontal slot 16 like this, conveniently utilize to clean cover 10 and clean camera lens 8, be provided with anti-skidding line on the installation knob 11 circumference outer wall, the diameter of installation knob 11 is greater than the diameter of rotary rod 12, like this when rotatory installation knob 11, avoids appearing the condition that the hand is smooth, avoids simultaneously cleaning cover 10 and drops.
The utility model discloses a theory of operation and use flow: the utility model discloses install the back, during the staff used this video monitoring who has restless action early warning system based on degree of depth study and RASS mark, can utilize the universal wheel on this bottom plate 1 to shift it to the ward, aim at patient with watch-dog 7, then open watch-dog 7 through the switch, utilize camera lens 8 on the watch-dog 7 to catch patient's action, and carry out the analysis through study, mark for patient's action and facial expression, and in time carry out alarm processing, can specifically divide into the limbs action and divide 6 minutes: 1 minute is no movement, 2 minutes of limbs move horizontally and slowly, 3 minutes of limbs wave slowly or move horizontally and rapidly, 4 minutes of limbs wave rapidly and frequently, 5 minutes of limbs try to pull body pipelines (stomach tubes, infusion pipelines, catheters and the like), and 6 minutes of limbs move vertically (try to sit up); facial expression 3 points: 1 is divided into no expression or quiet expression, 2 is divided into occasional frowns or mouth opening, and 3 is divided into frequent frowns; firstly, identifying limb activities, wherein 1-2 are identified expressions, and if 1-2 are identified, the monitoring is continued without alarming for the moment; if the score is 3, yellow early warning is carried out, no alarm is carried out, the score is 3-4, expression is identified, if the score is 1-2, the alarm is not carried out for the moment, and monitoring is continued; if 3 points are given, yellow early warning is carried out, no alarm is given, and 5-6 points are given with immediate red alarm;
during learning, the first step: collecting a sufficient number of video clip samples and daily behavior samples of the restless patient; step two: extracting 2D coordinates of each human skeleton key point of a plurality of frames from the restless video fragment sample and the daily behavior sample in the step one by using a skeleton key point extraction algorithm, constructing a restless skeleton sequence and a daily behavior skeleton sequence, and storing the restless skeleton sequence and the daily behavior skeleton sequence; step three: constructing a space-time diagram from the sequence file data in the step two, and performing deep learning; step four: step three, training an ending stroke agitation model;
when in work: shooting a real-time video by using a camera of the device, and decoding and converting the video; step two: storing the acquired real-time video, extracting 2D coordinate information of skeleton key points of each human body of multiple frames from the video data in the memory, constructing a sequence, and temporarily storing the sequence in the memory; step three: constructing a space-time diagram of the sequence obtained in the second step, inputting a restless model to obtain an output result, and simultaneously performing motion recognition on the video content obtained in the second step to judge whether restless behaviors occur or not; step four: continuously repeating the second step and the third step, detecting the agitation behavior of the real-time data, if the agitation behavior occurs, giving an alarm, and if the agitation behavior does not occur, entering the next round of detection;
when the monitor 7 is used for monitoring a patient for a long time, the lens 8 is prevented from being stained with dust, the driving motor 13 needs to be started to remove dust, the model of the driving motor 13 is Y60, when the driving motor 13 is started, the crankshaft 17 in the driving motor drives the rotary rod 12 to rotate back and forth, and meanwhile, the rotary rod 12 drives the wiping sleeve 10 on the outer wall of the rotary rod 12 to rotate together, so that the dust on the lens 8 can be wiped by the wiping sleeve 10, when the driving motor 13 stops, the rotary rod 12 and the wiping sleeve 10 are pulled aside by the spring 15 to avoid blocking the lens 8, and the wiping sleeve 10 can also be taken down from the rotary rod 12, and then cleaning is carried out, so that the cleaning device is convenient to reuse, and when the wiping sleeve 10 is taken out, only the mounting knob 11 needs to rotate and be taken down.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. Restless action early warning system based on degree of depth study and RASS score, including bottom plate (1) and mount pad (2), its characterized in that: the mounting seat (2) is mounted on the outer wall of the upper end of the bottom plate (1), the outer wall of the upper end of the mounting seat (2) is provided with a stand column (3), the stand column (3) is provided with a fixing ring block (4) by means of the outer wall of the circumference of the upper side, the outer wall of the circumference of the fixing ring block (4) is provided with a connecting rod (5), the outer wall of the front end of the connecting rod (5) is provided with a placing box (9), a monitor (7) is arranged inside the outer wall of the front end of the placing box (9), a lens (8) is arranged inside the outer wall of the front end of the monitor (7), the outer wall of the upper end of the placing box (9) is provided with a case (6), the outer wall of the front end of the monitor (7) is provided with a wiping sleeve (10), the outer wall of the lower end of the wiping sleeve (10) is provided with, driving motor (13) front end inside is provided with crankshaft (17) that stretches out forward, be provided with rotary rod (12) on crankshaft (17) circumference outer wall, rotary rod (12) lower extreme outer wall inside is provided with installation screw (20), the inside of installation screw (20) is provided with installation screw (19), be provided with fixed plate (14) on frame (18) on crankshaft (17) right side, fixed plate (14) and rotary rod (12) are connected through spring (15), be provided with cross slot (16) on placing case (9) under rotary rod (12), driving motor (13) and watch-dog (7) are all through quick-witted case (6) internal power source electric connection.
2. The deep learning and RASS scoring based agitation behavior early warning system according to claim 1, wherein: the outer wall of the lower end of the bottom plate (1) is provided with universal wheels.
3. The deep learning and RASS scoring based agitation behavior early warning system according to claim 1, wherein: a battery cover is arranged on the upper portion of the outer wall of the rear end of the case (6), and a battery is arranged inside the outer wall of the rear end of the case (6).
4. The deep learning and RASS scoring based agitation behavior early warning system according to claim 1, wherein: the wiping sleeve (10) is made of cylindrical cloth materials, and the wiping sleeve (10) is sleeved on the circumferential outer wall of the rotating rod (12).
5. The deep learning and RASS scoring based agitation behavior early warning system according to claim 1, wherein: the rotating rod (12) is arranged inside the transverse groove (16).
6. The deep learning and RASS scoring based agitation behavior early warning system according to claim 1, wherein: the anti-skidding thread is arranged on the circumferential outer wall of the mounting knob (11), and the diameter of the mounting knob (11) is larger than that of the rotating rod (12).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202021987274.1U CN212752473U (en) | 2020-09-12 | 2020-09-12 | Restless behavior early warning system based on deep learning and RASS scoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202021987274.1U CN212752473U (en) | 2020-09-12 | 2020-09-12 | Restless behavior early warning system based on deep learning and RASS scoring |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN212752473U true CN212752473U (en) | 2021-03-19 |
Family
ID=74991258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202021987274.1U Expired - Fee Related CN212752473U (en) | 2020-09-12 | 2020-09-12 | Restless behavior early warning system based on deep learning and RASS scoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN212752473U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115493052A (en) * | 2022-11-16 | 2022-12-20 | 山东省物化探勘查院 | Portable level |
-
2020
- 2020-09-12 CN CN202021987274.1U patent/CN212752473U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115493052A (en) * | 2022-11-16 | 2022-12-20 | 山东省物化探勘查院 | Portable level |
| CN115493052B (en) * | 2022-11-16 | 2023-02-03 | 山东省物化探勘查院 | Portable level |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN212752473U (en) | Restless behavior early warning system based on deep learning and RASS scoring | |
| CN107729859B (en) | Prevent intelligent fingerprint identification machine that fingerprint leaked based on thing networking | |
| CN103997622A (en) | Automatic cleaning method of unattended monitoring camera | |
| CN108647645B (en) | Multi-modal epilepsy diagnosis system and method based on video analysis | |
| CN114785918B (en) | Resident daily garbage throwing behavior monitoring device and processing system | |
| CN210093384U (en) | Intelligent video analysis monitoring equipment | |
| CN112036337A (en) | Intelligence evaluation workstation of washing hand | |
| CN112764231B (en) | CT medical image is with reading piece ware based on AI technique | |
| CN216962413U (en) | Laryngoscope convenient to clearance camera lens | |
| CN109740566A (en) | A kind of computer digital image processing equipment | |
| CN220087413U (en) | Underwater optical wireless sensor network safety management equipment | |
| CN213244207U (en) | Informationized VR (virtual reality) technology platform for aquaculture | |
| CN213279914U (en) | Auxiliary tool for security monitoring equipment | |
| CN210994922U (en) | Automatic moisturizing device for instruments | |
| CN216391161U (en) | Multi-angle image acquisition equipment | |
| CN111948411A (en) | Blood mixing device and blood mixing management system | |
| CN213522164U (en) | Security protection monitored control system that thermoinduction triggered | |
| CN206311678U (en) | A kind of transformer is not drained the oil device body imaging system | |
| CN118238159A (en) | Security patrol robot system applied to intelligent park | |
| CN200975981Y (en) | Cell morphology automatic recognition device | |
| CN213179990U (en) | Double-temporal remote sensing image change monitoring device | |
| CN215115364U (en) | Dust collection system for environmental monitoring | |
| CN216318659U (en) | Multi-functional shallow for nursing | |
| CN220965015U (en) | Visual supervisory equipment with automatic alarm function | |
| CN217060119U (en) | Online monitoring device for odor factors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210319 Termination date: 20210912 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |