CN110008865A - A kind of kindergarten's morning detecting method and its device - Google Patents
A kind of kindergarten's morning detecting method and its device Download PDFInfo
- Publication number
- CN110008865A CN110008865A CN201910225265.4A CN201910225265A CN110008865A CN 110008865 A CN110008865 A CN 110008865A CN 201910225265 A CN201910225265 A CN 201910225265A CN 110008865 A CN110008865 A CN 110008865A
- Authority
- CN
- China
- Prior art keywords
- kindergarten
- morning
- neuron models
- neuron
- mouth
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- G16H30/00—ICT specially adapted for the handling or processing of medical images
-
- 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
- G16H70/00—ICT specially adapted for the handling or processing of medical references
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Neurology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Epidemiology (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of kindergarten's morning detecting methods, first creation neuron models, are then trained by a large amount of blebs, brothers' mouth, angina, blood-shoot-eye illness picture to neuron models；The image information in personnel to be tested's hand and oral cavity is acquired by image acquisition units again；Then the image information input neuron models of image acquisition units acquisition are detected, finally exports result；Then the mouth, hand and face that this method shoots child by camera carry out examination calculating using the neuron after training, entire detection process is fast, and computer system, will not be tired, as long as early period, neuron trained, that is, can guarantee accuracy；In addition, the invention also discloses a kind of kindergarten's morning checking device, including image acquisition units, image processing module, neuron models module and output module.
Description
Technical field
The present invention relates to the kindergarten field Chen Jian, in particular to a kind of the method and device of the inspection of kindergarten's morning.
Background technique
The inspection of kindergarten's morning refers to that the health worker of kindergarten's every morning carries out hygienic medical inspection to the child in every garden Wei Ru,
Main inspection item has bleb, brothers' mouth, angina, blood-shoot-eye illness etc., and existing test mode is not professional enough, is not each
Health worker has the professional skill of doctor, in addition, shown according to investigation, average 260 people of kindergarten, every institute, everyone survey body temperature,
Check hand, mouth, eyes, at least 5-8 seconds, then a garden checks and needs 40 minutes.The time is long per capita, heavy workload, time collection
In, health doctor is easy fatigue, it is possible to have phenomena such as missing inspection, false retrieval.It is more that there are also students, has no idea to accomplish every diary
Long-term analysis can not be made to student health without data record by recording data.
Summary of the invention
The case where time-consuming in order to solve existing kindergarten's morning inspection, is easy to happen missing inspection, false retrieval, the present invention provides one kind
High efficiency, kindergarten's morning detecting method of low serious forgiveness.
To achieve the goals above, present invention provide the technical scheme that a kind of kindergarten's morning detecting method, first creation mind
Through meta-model, learning framework is calculated by pytorch and establishes neuron models；
Then neuron models are trained by a large amount of blebs, brothers' mouth, angina, blood-shoot-eye illness picture；
The image information in personnel to be tested's hand and oral cavity is acquired by image acquisition units again；
Then the image information input neuron models of image acquisition units acquisition are detected, calculates characteristic point point
Cloth situation and characteristic point probability value；
Warning finally is issued if characteristic point probability value is greater than the set value, otherwise personnel to be detected pass through detection.
In the above-mentioned technical solutions, the neuron models are using SSD algorithm to target detection and feature extraction.
Based on the above technical solution, the neuron training includes to bleb, brothers' mouth, angina, blood-shoot-eye illness
Illness point on picture is labeled, then the picture of mark is carried out repetition training to neuron models using SSD algorithm.
As a kind of preferred technical solution, image acquisition units use high-definition camera.
Further, neuron models are learnt again after detecting bleb, brothers' mouth, angina, blood-shoot-eye illness every time.
It is described to learn to be to utilize the picture for detecting bleb, brothers' mouth, angina, blood-shoot-eye illness again to neuron models again
It is trained.
In addition, the present invention also provides a kind of kindergarten's morning checking devices, comprising: image acquisition units, it is to be detected for acquiring
The hand of personnel, face, mouth images information；Image processing module carries out the image information of image acquisition units acquisition special
Sign is extracted；Neuron models module, for the creation of neuron models and the training of neuron models；Output module is used for
Export image comparison result.
The device further includes non-visual LED light supplement lamp, and the non-visual LED light supplement lamp is for image acquisition units acquisition figure
Light filling is carried out when picture.
In addition, the device further includes remote server, the remote server is for storing data.
The output module in the device is acoustic-optic alarm as a preferred technical solution,.
The beneficial effect of the present invention compared with the existing technology is: this method by camera shoot the mouth of child, hand and
Then face carries out examination calculating using the neuron after training, entire detection process is fast, obtains through overtesting, average every
A child only needs that entire detection process, and computer system can be completed within 3-5 seconds, will not be tired, as long as neuron early period is instructed
Perfect, that is, can guarantee accuracy, in addition, data can also save in the server, the later period can be checked, can also trace with
Track.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention；
Fig. 2 is the device of the invention structure chart.
Specific embodiment
Specific embodiments of the present invention will be further explained with reference to the accompanying drawing.It should be noted that for
The explanation of these embodiments is used to help understand the present invention, but and does not constitute a limitation of the invention.In addition, disclosed below
The each embodiment of the present invention involved in technical characteristic can be combined with each other as long as they do not conflict with each other.
As shown in Figure 1, the present invention provides a kind of kindergarten's morning detecting method, comprising the following steps:
S1: creation neuron models calculate learning framework by pytorch and establish neuron models, using in pytorch
Torch.utils.data.Dataset data-interface, realize training dataset, test data set；
S2: being trained neuron models, helps the side such as acquisition, purchase with community school district by medical industry
Formula collects that a large amount of bleb, brothers mouthful, picture that the picture of angina is collected are more, and to the model of later period training, precision is higher,
In the present embodiment, 1000 are collected respectively using diseases such as bleb, brothers' mouth, anginas, then pass through software LabelImg
Or BBox-Label-Tool is labeled the illness point on above-mentioned picture, the picture marked is inputted network model, so
Python language call pytorch is used afterwards, and established neuron models are trained using SSD algorithm.
S3: shooting the oral cavity of child to be detected, the picture of hand by camera, then by picture transfer to neuron mould
In type；
S4: calling trained neuron models, and the picture for carrying out camera shooting is detected, and calculates characteristic point point
Cloth situation and characteristic point probability value；
S5: when to detecting that characteristic point probability value is greater than the set value, (in the present embodiment, using 0.6) passing through computer
Audible and visual alarm, prompt teacher carry out intervention detection.
In addition, needing to be learned again after detecting bleb, brothers' mouth, angina, blood-shoot-eye illness every time in neuron models
Practise, then learn be using detect bleb, brothers' mouth, angina, blood-shoot-eye illness picture further neuron models are instructed
Practice, training method is identical as training method before.The destination of study is to guarantee the accuracy of the detection of neuron again,
Above-mentioned detection mode is the detection method for hand bleb and herpes of mouth, is described in detail below for red
The detection method of eye disease,
Facial image is shot by camera first；
The facial image of shooting is input in opencv, the classifier then carried by opencv
(haarcascade_eye_tree_eyeglasses.xml), pupil position is positioned；
Then the CascadeClassifier classifier carried in the image shot by opencv, positions the wheel of eyes
Wide position.Then the picture in eye contour calculates red similar value, show that red is occupied in entire eyes picture
Very, more than 20%, then audible and visual alarm reminds teacher to think to screen intervention.
Then the mouth, hand and face that this method shoots child by camera are carried out using the neuron after training
It screens and calculates, entire detection process is fast, obtains through overtesting, and average each child, which only needs to can be completed for 3-5 seconds, entirely to be detected
Journey, and computer system, will not be tired, as long as early period, neuron trained, that is, can guarantee accuracy, in addition, data may be used also
To save in the server, the later period can be checked, can also trace tracking.
As shown in Fig. 2, in addition, the present invention also provides a kind of kindergarten's morning checking devices, comprising: image acquisition units are used for
Acquire the hand, face, mouth images information of personnel to be detected；In the present embodiment, image acquisition units use multiple high definitions
Camera is respectively used to shooting hand, face and the camera in oral cavity, can be improved efficiency in this way, in addition, each camera
All it is equipped with non-visual LED light supplement lamp, when dark, openable LED light supplement lamp carries out light filling, so that the photo of shooting is not
Post-processing can be impacted.
In addition, the device further includes image processing module, the image information for acquiring image acquisition units carries out special
Sign is extracted；The illness point on above-mentioned picture is labeled by software LabelImg or BBox-Label-Tool.
It further, further include neuron models module, for the creation of neuron models and the instruction of neuron models
Practice；The module includes that two parts one are to calculate learning framework by pytorch to establish neuron models, using pytorch
In torch.utils.data.Dataset data-interface, realize training dataset, test data set；Another part is
Neuron trains part, and the picture that image processing module is handled well is imported into neuron models, then uses python language
Tone pytorch is trained established neuron models using SSD algorithm.
It further include output module, for exporting image comparison as a result, in the present embodiment, output module is sound-light alarm dress
It sets, when testing result is abnormal, sounds the alarm, kindergarener is reminded to carry out artificial screening.
In addition, the device further includes remote server, for storing data, the later period can check remote server, can also
To trace tracking.
In conjunction with attached drawing, the embodiments of the present invention are described in detail above, but the present invention is not limited to described implementations
Mode.For a person skilled in the art, in the case where not departing from the principle of the invention and spirit, to these embodiments
A variety of change, modification, replacement and modification are carried out, are still fallen in protection scope of the present invention.
Claims (10)
1. a kind of kindergarten's morning detecting method, it is characterised in that:
Neuron models are created, learning framework is calculated by pytorch and establishes neuron models；
Neuron models are trained by a large amount of blebs, brothers' mouth, angina, blood-shoot-eye illness picture；
The image information in personnel to be tested's hand and oral cavity is acquired by image acquisition units；
By image acquisition units acquisition image information input neuron models detect, calculate characteristic point distribution situation with
And characteristic point probability value；
Warning is issued if characteristic point probability value is greater than the set value, otherwise personnel to be detected pass through detection.
2. kindergarten's morning detecting method according to claim 1, it is characterised in that: the neuron models use SSD algorithm
To target detection and feature extraction.
3. kindergarten's morning detecting method according to claim 2, it is characterised in that: the neuron training include to bleb,
Brothers' mouth, angina, the illness point on blood-shoot-eye illness picture are labeled, then by the picture of mark using SSD algorithm to neuron
Model carries out repetition training.
4. kindergarten's morning detecting method according to claim 1, it is characterised in that: described image acquisition unit is taken the photograph using high definition
As head.
5. kindergarten's morning detecting method according to claim 1, it is characterised in that: the neuron models are detecting every time
Learnt again after bleb, brothers' mouth, angina, blood-shoot-eye illness.
6. kindergarten's morning detecting method according to claim 5, it is characterised in that: described to learn to be to utilize to detect blister again
Rash, brothers' mouth, angina, blood-shoot-eye illness picture neuron models are trained again.
7. a kind of kindergarten's morning checking device characterized by comprising
Image acquisition units, for acquiring the hand, face, mouth images information of personnel to be detected；
The image information of image acquisition units acquisition is carried out feature extraction by image processing module；
Neuron models module, for the creation of neuron models and the training of neuron models；
Output module, for exporting image comparison result.
8. kindergarten's morning checking device according to claim 7, it is characterised in that: it further include non-visual LED light supplement lamp, it is described
Non- visual LED light supplement lamp is for carrying out light filling when image acquisition units acquisition image.
9. kindergarten's morning checking device according to claim 7, it is characterised in that: it further include remote server, it is described long-range
Server is for storing data.
10. kindergarten's morning checking device according to claim 7, it is characterised in that: the output module is sound-light alarm dress
It sets.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910225265.4A CN110008865A (en) | 2019-03-25 | 2019-03-25 | A kind of kindergarten's morning detecting method and its device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910225265.4A CN110008865A (en) | 2019-03-25 | 2019-03-25 | A kind of kindergarten's morning detecting method and its device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110008865A true CN110008865A (en) | 2019-07-12 |
Family
ID=67167836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910225265.4A Pending CN110008865A (en) | 2019-03-25 | 2019-03-25 | A kind of kindergarten's morning detecting method and its device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110008865A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111161872A (en) * | 2019-12-03 | 2020-05-15 | 王洁 | Intelligent management system for child health |
| CN111281361A (en) * | 2020-03-09 | 2020-06-16 | 康瑞健康管理(杭州)有限公司 | Student health monitoring system based on big data |
| CN111358445A (en) * | 2020-03-23 | 2020-07-03 | 苏州昊瑞智能装备科技有限公司 | Morning check machine |
| CN111598865A (en) * | 2020-05-14 | 2020-08-28 | 上海锘科智能科技有限公司 | Hand-foot-and-mouth disease detection method, device and system based on thermal infrared and RGB (red, green and blue) double-taking |
-
2019
- 2019-03-25 CN CN201910225265.4A patent/CN110008865A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111161872A (en) * | 2019-12-03 | 2020-05-15 | 王洁 | Intelligent management system for child health |
| CN111161872B (en) * | 2019-12-03 | 2022-08-23 | 王洁 | Intelligent management system for child health |
| CN111281361A (en) * | 2020-03-09 | 2020-06-16 | 康瑞健康管理(杭州)有限公司 | Student health monitoring system based on big data |
| CN111358445A (en) * | 2020-03-23 | 2020-07-03 | 苏州昊瑞智能装备科技有限公司 | Morning check machine |
| CN111598865A (en) * | 2020-05-14 | 2020-08-28 | 上海锘科智能科技有限公司 | Hand-foot-and-mouth disease detection method, device and system based on thermal infrared and RGB (red, green and blue) double-taking |
| CN111598865B (en) * | 2020-05-14 | 2023-05-16 | 上海锘科智能科技有限公司 | Hand-foot-mouth disease detection method, device and system based on thermal infrared and RGB double-shooting |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110008865A (en) | A kind of kindergarten's morning detecting method and its device | |
| WO2019028592A1 (en) | Teaching assistance method and teaching assistance system using said method | |
| CN111528859B (en) | Child ADHD screening and evaluating system based on multi-modal deep learning technology | |
| Scott et al. | Healthcare applications of single camera markerless motion capture: a scoping review | |
| CN107886503A (en) | A kind of alimentary canal anatomical position recognition methods and device | |
| CN106023516A (en) | Examination monitoring method and system and examination room monitor | |
| CN109858540A (en) | A kind of medical image recognition system and method based on multi-modal fusion | |
| CN108030498A (en) | A kind of Psychological Intervention System based on eye movement data | |
| CN113856186B (en) | Pull-up action judging and counting method, system and device | |
| CN110660454A (en) | Cancer pain real-time assessment instrument and assessment method thereof | |
| CN109620266A (en) | The detection method and system of individual anxiety level | |
| CN108922617A (en) | A kind of self-closing disease aided diagnosis method neural network based | |
| CN116098621B (en) | Emotion face and physiological response recognition method based on attention mechanism | |
| CN113143274A (en) | Emotion early warning method based on camera | |
| CN106175657B (en) | A kind of eyesight automatic checkout system | |
| CN113343853B (en) | Intelligent screening method and device for dental caries of children | |
| Wang et al. | [Retracted] Student Physical Fitness Test System and Test Data Analysis System Based on Computer Vision | |
| US20210390687A1 (en) | Method for enriching a learning base | |
| CN115798040B (en) | Automatic segmentation system of cardiopulmonary resuscitation AI | |
| CN110600127A (en) | Video acquisition and analysis system and method for realizing cognitive disorder screening function by video excitation of facial expressions | |
| CN111428577B (en) | Face living body judgment method based on deep learning and video amplification technology | |
| CN107832698A (en) | Learning interest testing method and device based on array lens | |
| CN113673856A (en) | 3D-based hospital sensory control simulation training and examination system | |
| Cherin et al. | A hygiene monitoring system | |
| CN117671774B (en) | Face emotion intelligent recognition analysis equipment |
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 |