CN208969808U - Baby monitor based on Face datection and sound detection - Google Patents
Baby monitor based on Face datection and sound detection Download PDFInfo
- Publication number
- CN208969808U CN208969808U CN201821336191.9U CN201821336191U CN208969808U CN 208969808 U CN208969808 U CN 208969808U CN 201821336191 U CN201821336191 U CN 201821336191U CN 208969808 U CN208969808 U CN 208969808U
- Authority
- CN
- China
- Prior art keywords
- baby
- module
- sound
- sent
- detection module
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The utility model discloses a kind of baby monitor based on Face datection and sound detection, monitor integrated camera and sound collection equipment acquire the image data and voice data of baby in real time, give Face datection algorithm, humanoid detection algorithm and speech detection algorithm, pass through the data result of return, parse baby data characteristics come judge baby whether waking state, the equipment of warning information at one's side to guardian is sent if awake not deal with if normal condition.
Description
Technical field
The utility model relates to baby monitor technical fields, it particularly relates to which a kind of examined based on Face datection and sound
The baby monitor of survey.
Background technique
With the development of society, various intelligence band baby's equipment are come into being.Particularly with oneself parent with baby, they are right
The craving degree of these smart machines is bigger.Wherein, parent be desirable in can be trusted when baby's sleep let go and do it
Its thing.Currently, existing baby monitor equipment multi-pass is crossed the sound for monitoring baby and is moved by camera head monitor baby
Make to guard infantile state, but is also easy whether erroneous judgement baby is waking state by sound decibel and camera shooting movement.
Utility model content
The purpose of this utility model is that a kind of baby monitor based on Face datection and sound detection is proposed, to overcome
The above deficiencies in the existing technologies.
To realize the above-mentioned technical purpose, the technical solution of the utility model is achieved in that
Baby monitor based on Face datection and sound detection, the baby monitor include:
Image capture module is sent to Face datection mould for acquiring the image information of baby, and by described image information
Block and humanoid detection module;
The result of judgement for judging whether baby blinks and open one's mouth, and is sent to infantile state by face detection module
Analysis module;
Humanoid detection module is sent to infantile state for judging whether baby has limb action, and by the result of judgement
Analysis module;
Sound acquisition module is sent to sound inspection for acquiring the acoustic information around baby, and by the acoustic information
Survey module;
Sound detection module is sent to infantile state analysis mould for judging whether baby cries and shout, and by the result of judgement
Block;
Infantile state analysis module, for judging whether baby regains consciousness, when baby blinks, opens one's mouth, there is limb action and cries
When crying out, determine that baby is awake, and issue alerting signal to remote intelligent terminal, when baby blinks, opens one's mouth, there is limb action and cries
When thering is any state not occur in crying out, determine baby also in sleep;
Power module, for image capture module, face detection module, humanoid detection module, sound acquisition module, sound
Sound detection module and infantile state analysis module provide power supply.
Further, described image acquisition module is camera.
Further, the sound acquisition module is microphone.
The utility model has the beneficial effects that the utility model is by existing human face detection tech, humanoid detection technique harmony
Whether whether sound detection technique is integrated on baby monitor, by opening one's mouth, blinking to baby, whether have limb action and be
Simple logical operation is done in no sobbing, therefore, it is determined that whether the baby in sleep regains consciousness, and then can when baby regains consciousness
It notifies parent at the first time, reduces guardian's accompanying and attending to the time under sleeping condition of baby.
Detailed description of the invention
Fig. 1 is the functional module structure schematic diagram of monitor described in the utility model.
Specific embodiment
Below with reference to the attached drawing in the utility model embodiment, the technical scheme in the embodiment of the utility model is carried out clear
Chu is fully described by.
As shown in Figure 1, being supervised described in embodiment according to the present utility model based on the baby of Face datection and sound detection
Device is protected, the baby monitor includes power module 6 and the image capture module 1, the face that are separately connected with the power module 6
Detection module 3, humanoid detection module 2, sound acquisition module 4, sound detection module 5 and infantile state analysis module 7.
Image capture module 1 is sent to Face datection mould for acquiring the image information of baby, and by described image information
Block 3 and humanoid detection module 2.Specifically, described image acquisition module 1 uses high-definition camera.
The result of judgement for judging whether baby blinks and open one's mouth, and is sent to infantile state by face detection module 3
Analysis module 7.Specifically, face detection module 3 is realized based on technology contents disclosed in attachment 1 in the present embodiment, packet
Include following testing process: 1, the method (letter of utilization orientation histogram of gradients (Histogram of Oriented Gradients)
Claim HOG), the data comparison that picture is converted into HOG form and training has been detected whether into face;2, using facial facial feature estimation
The algorithm of (face landmark estimation) extracts characteristic and the training data comparison of face's eyes and mouth,
Judge whether baby opens eyes and open one's mouth.
Humanoid detection module 2 is sent to infantile state for judging whether baby has limb action, and by the result of judgement
Analysis module 7.Specifically, humanoid detection module 2 is realized based on technology contents disclosed in attachment 2, limbs are carried out first
The detection of key point, comprising: then 1, input piece image pass through 7 stage, obtains PCM and PAF.Then raw according to PAF
At a series of even matching, due to the vector property of PAF itself, so that the even matching generated is very correct, a people is finally merged into
Overall skeleton.2, crucial point location: in image, how much how many can identify how many just positioning, and the point having had is carried out individual
Association.Different part in the same body are connected.3, each pixel of confidence level matrix S(is calculated by CNN
The possessed confidence level of point), L (being direction vector, 2 dimensions are respectively the component in the direction x and the direction y) S: each point is set
Reliability confidence;L: the direction vector of each point is in x, the component in the direction y.Then determine whether baby there are limbs dynamic
Make, comprising: 1, use the color image having a size of w × h as inputting, and generate the position 2D of everyone dissection key point in image
It sets as output.2, firstly, feedforward network predicts the one group of two dimension confidence level figure S and part affinity of body part position simultaneously
(PAF) one group of two-dimensional vector field L(vector), which show the correlation degrees between body part.3,S=(S1,S2,...,
Sj) there are j confidence level map, each body part one;L=(L1, L2, L3..., LC) has c vector field, each limbs one
It is a.Position encoded 2D vector in L c.It 4, is image finally, by greedy rational analysis confidence level figure and affinity field
In owner export two-dimentional key point.5, the limbs key point position in the image for getting camera carries out continuous right
Than judging whether baby carries out limb motion.
Sound acquisition module 4 is sent to sound inspection for acquiring the acoustic information around baby, and by the acoustic information
Survey module 5.Specifically, in the present embodiment, the sound acquisition module 4 is microphone.
Sound detection module 5 is sent to infantile state analysis mould for judging whether baby cries and shout, and by the result of judgement
Block 7.Specifically, sound detection module is realized based on technical solution disclosed in attachment 3, including following process: 1, baby is used
Crying training voice, extract MFCC characteristic parameter, establish identification model.2, to voice to be identified and identification model progress
Match, calculate matching error, exports recognition result.
Infantile state analysis module 7, for judging whether baby regains consciousness, when baby blinks, opens one's mouth, there is limb action and cries
When crying out, determine that baby is awake, and issue alerting signal to remote intelligent terminal, so that guardian be enable timely to handle sobbing
Baby;If thering is any state not occur during baby blinks, opens one's mouth, there is limb action and cries and shout, determine baby also
In sleep, then system is without any movement.
The above is only the preferred embodiment of the utility model only, is not intended to limit the utility model, all at this
Within the spirit and principle of utility model, any modification, equivalent replacement, improvement and so on should be included in the utility model
Protection scope within.
Claims (3)
1. the baby monitor based on Face datection and sound detection, which is characterized in that the baby monitor includes:
Image capture module, for acquiring the image information of baby, and by described image information be sent to face detection module and
Humanoid detection module;
Face detection module is sent to infantile state analysis for judging whether baby blinks and open one's mouth, and by the result of judgement
Module;
Humanoid detection module is sent to infantile state analysis for judging whether baby has limb action, and by the result of judgement
Module;
Sound acquisition module is sent to sound detection mould for acquiring the acoustic information around baby, and by the acoustic information
Block;
Sound detection module is sent to infantile state analysis module for judging whether baby cries and shout, and by the result of judgement;
Infantile state analysis module, for judging whether baby regains consciousness, when baby blinks, opens one's mouth, has limb action and cry and shout,
Determine that baby is awake, and issue alerting signal to remote intelligent terminal, in baby blinks, opens one's mouth, has limb action and cry and shout
If thering is any state not occur, determine baby also in sleep;
Power module, for being examined to image capture module, face detection module, humanoid detection module, sound acquisition module, sound
It surveys module and infantile state analysis module provides power supply.
2. baby monitor according to claim 1, which is characterized in that described image acquisition module is camera.
3. baby monitor according to claim 1, which is characterized in that the sound acquisition module is microphone.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201821336191.9U CN208969808U (en) | 2018-08-20 | 2018-08-20 | Baby monitor based on Face datection and sound detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201821336191.9U CN208969808U (en) | 2018-08-20 | 2018-08-20 | Baby monitor based on Face datection and sound detection |
Publications (1)
Publication Number | Publication Date |
---|---|
CN208969808U true CN208969808U (en) | 2019-06-11 |
Family
ID=66753544
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201821336191.9U Active CN208969808U (en) | 2018-08-20 | 2018-08-20 | Baby monitor based on Face datection and sound detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN208969808U (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110460809A (en) * | 2019-07-05 | 2019-11-15 | 苏宁智能终端有限公司 | A kind of vagitus method for detecting, device and intelligent camera head apparatus |
CN110472481A (en) * | 2019-07-01 | 2019-11-19 | 华南师范大学 | A kind of sleeping position detection method, device and equipment |
CN113298062A (en) * | 2021-07-28 | 2021-08-24 | 宁波星巡智能科技有限公司 | Infant sleep activity state detection method, device, equipment and medium |
CN113409543A (en) * | 2021-06-24 | 2021-09-17 | 广西宁达汽车科技有限公司 | System and method for caring for children in vehicle |
-
2018
- 2018-08-20 CN CN201821336191.9U patent/CN208969808U/en active Active
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110472481A (en) * | 2019-07-01 | 2019-11-19 | 华南师范大学 | A kind of sleeping position detection method, device and equipment |
CN110472481B (en) * | 2019-07-01 | 2024-01-05 | 华南师范大学 | Sleeping gesture detection method, device and equipment |
CN110460809A (en) * | 2019-07-05 | 2019-11-15 | 苏宁智能终端有限公司 | A kind of vagitus method for detecting, device and intelligent camera head apparatus |
CN113409543A (en) * | 2021-06-24 | 2021-09-17 | 广西宁达汽车科技有限公司 | System and method for caring for children in vehicle |
CN113298062A (en) * | 2021-07-28 | 2021-08-24 | 宁波星巡智能科技有限公司 | Infant sleep activity state detection method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN208969808U (en) | Baby monitor based on Face datection and sound detection | |
Gong et al. | A real-time fire detection method from video with multifeature fusion | |
Tsai et al. | Implementation of fall detection system based on 3D skeleton for deep learning technique | |
CN109919031A (en) | A kind of Human bodys' response method based on deep neural network | |
CN110147738B (en) | Driver fatigue monitoring and early warning method and system | |
Chen et al. | Fall detection system based on real-time pose estimation and SVM | |
CN110472481B (en) | Sleeping gesture detection method, device and equipment | |
CN101655907A (en) | Trainman driving state monitoring intelligent alarm system | |
CN107301376B (en) | Pedestrian detection method based on deep learning multi-layer stimulation | |
WO2019153175A1 (en) | Machine learning-based occluded face recognition system and method, and storage medium | |
CN114255508A (en) | OpenPose-based student posture detection analysis and efficiency evaluation method | |
CN113065515B (en) | Abnormal behavior intelligent detection method and system based on similarity graph neural network | |
Wei et al. | Real-time facial expression recognition for affective computing based on Kinect | |
CN111062356A (en) | Method for automatically identifying human body action abnormity from monitoring video | |
CN110633624A (en) | Machine vision human body abnormal behavior identification method based on multi-feature fusion | |
CN113920326A (en) | Tumble behavior identification method based on human skeleton key point detection | |
CN107133611A (en) | A kind of classroom student nod rate identification with statistical method and device | |
CN110148092A (en) | The analysis method of teenager's sitting posture based on machine vision and emotional state | |
CN114926957A (en) | Infant monitoring system and monitoring method based on smart home | |
CN115691762A (en) | Autism child safety monitoring system and method based on image recognition | |
CN111178201A (en) | Human body sectional type tracking method based on OpenPose posture detection | |
Liu et al. | The study on human action recognition with depth video for intelligent monitoring | |
WO2024099313A1 (en) | Cloud-edge-end collaborative intelligent infant care system and method | |
CN110287933A (en) | A kind of dynamic human face recognition system and recognition methods based on stereo video streaming | |
CN109745053A (en) | A kind of body state detection device and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230615 Address after: 13C-18, Caihong Building, Caihong Xindu, No. 3002, Caitian South Road, Gangsha Community, Futian Street, Futian District, Shenzhen, Guangdong 518033 Patentee after: Core Computing Integrated (Shenzhen) Technology Co.,Ltd. Address before: 830022 Room 1010, Floor 10, Building A, Chuangzhi Building, Xinjiang Software Park, No. 455, Kanas North Road, UErvmqi Economic and Technological Development Zone Patentee before: XINJIANG AIHUA YINGTONG INFORMATION TECHNOLOGY Co.,Ltd. |
|
TR01 | Transfer of patent right |