CN108670196B - Method and device for monitoring sleep state of infant - Google Patents
Method and device for monitoring sleep state of infant Download PDFInfo
- Publication number
- CN108670196B CN108670196B CN201810362787.4A CN201810362787A CN108670196B CN 108670196 B CN108670196 B CN 108670196B CN 201810362787 A CN201810362787 A CN 201810362787A CN 108670196 B CN108670196 B CN 108670196B
- Authority
- CN
- China
- Prior art keywords
- infant
- value
- characteristic
- baby
- sleep state
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Anesthesiology (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a method and a device for monitoring the sleep state of an infant, wherein the method comprises the following steps: acquiring biological characteristics of the infant to generate characteristic values; judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state; if so, the infant is judged to be in the non-sleep state. The invention judges whether the baby is in the sleep state according to the biological characteristics of the baby, does not need to artificially monitor the sleep state of the baby, and saves the energy of parents or monitoring people. After the baby is not asleep, a message is sent to a preset mobile phone to remind parents or guardians of pacifying, or a preset audio file is automatically played to pacify the baby, so that the pressure of the parents or guardians is reduced.
Description
Technical Field
The invention relates to the field of application of infant monitoring, in particular to a method and a device for monitoring the sleep state of an infant.
Background
The biometric characteristic is a physiological characteristic (fingerprint, iris, voice, facial phase, DNA, etc.) or a behavioral characteristic (gait, behavioral posture, etc.) inherent to a living organism. For example, the voice uttered by a person, the size of the voice, and the content of the voice are all the biometric features of the person; the posture of a person while walking and the posture of the person while sleeping are also the biometric features of the person. Although the biological characteristics of each person are different, the same biological characteristics have some similarities, for example, when a plurality of persons walk, the height of each person lifting the leg, the stride length, the swinging amplitude of the arm and the like are different, but the postures of the persons are similar to each other by the way that two legs are crossed to step and the arm swings.
The infant is often in a sleeping state, and parents still need to work or work at home, and the like, and cannot always accompany the infant.
At present, parents remotely monitor the sleeping state of an infant by arranging a camera in a room of the infant and connecting the camera with a mobile phone of the parents. Parents or guardians monitor the sleep state of the infant by looking at the mobile phone.
Disclosure of Invention
The invention mainly aims to provide a method and a device for automatically identifying the sleep state of an infant, so that a user has better brightness when using a mobile phone and using each application program.
The invention provides a method for monitoring the sleep state of an infant, which comprises the following steps:
acquiring biological characteristics of the infant to generate characteristic values;
judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state;
if so, the infant is judged to be in the non-sleep state.
Further, after the step of determining that the infant is in the non-sleep state, the method includes:
sending information to a designated terminal; and/or
And controlling the loudspeaker to play the specified audio file.
Further, the step of determining whether the feature value exceeds a preset range of the feature value in the sleep state of the infant includes:
and judging whether the sound volume value exceeds a preset volume threshold value.
Further, the step of determining that the infant is in the non-sleep state includes:
if the sound volume value exceeds a preset volume threshold value, acquiring the source position of the sound;
judging whether the source position is a preset position for placing the baby;
if so, the infant is judged to be in the non-sleep state.
Further, the acquiring the biological feature of the infant and generating the feature value includes:
acquiring a picture corresponding to the current body posture of the baby;
inputting the picture into a model trained based on a neural network for calculation to obtain the characteristic vector value;
the step of judging whether the characteristic value exceeds a preset range of the characteristic value in the sleep state comprises the following steps:
and judging whether the characteristic vector value exceeds a preset characteristic vector threshold value.
The invention also provides a device for monitoring the sleep state of an infant, which comprises:
the acquisition module is used for acquiring the biological characteristics of the infant and generating a characteristic value;
the judging module is used for judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state;
and the judging module is used for judging that the infant is in the non-sleeping state if the characteristic value exceeds the preset range of the characteristic value in the infant sleeping state.
Further, the device for monitoring the sleep state of the infant further comprises:
the sending module is used for sending information to the appointed terminal;
and/or
And the playing module is used for controlling the loudspeaker to play the specified audio file.
Further, the biometric feature is a sound, the feature value is a sound volume value, and the determining module includes:
the first judging unit is used for judging whether the sound volume value exceeds a preset volume threshold value.
Further, the determination module includes:
the position unit is used for acquiring the source position of the sound if the sound volume value exceeds a preset volume threshold;
the second judgment unit is used for judging whether the source position is a preset position for placing the baby;
and the judging unit is used for judging that the infant is in the non-sleeping state if the source position is a preset position for placing the infant.
Further, the biometric feature is a body posture of the infant, the feature value is a feature vector value corresponding to the body posture, and the obtaining module includes:
the acquisition unit is used for acquiring a picture corresponding to the current body posture of the baby;
the calculation unit is used for inputting the picture into a model trained on the basis of a neural network to calculate to obtain the characteristic vector value;
the judging module comprises:
and the third judging unit is used for judging whether the characteristic vector value exceeds a preset characteristic vector threshold value.
Compared with the prior art, the invention has the beneficial effects that: whether the baby is in the sleep state or not is judged according to the biological characteristics of the baby, the sleep state of the baby is not required to be monitored manually, and the energy of parents or monitoring people is saved. After the baby is not asleep, a message is sent to a preset mobile phone to remind parents or guardians of pacifying, or a preset audio file is automatically played to pacify the baby, so that the pressure of the parents or guardians is reduced.
Drawings
FIG. 1 is a schematic flow chart of a method for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating a method for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of a method for monitoring the sleep state of an infant according to another embodiment of the present invention;
FIG. 4 is a flow chart illustrating a method for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 5 is a flow chart illustrating a method for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 6 is a flow chart illustrating a method for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an apparatus for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of an apparatus for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of an apparatus for monitoring the sleep state of an infant according to another embodiment of the present invention;
FIG. 10 is a schematic diagram of an apparatus for monitoring the sleep state of an infant according to an embodiment of the present invention;
FIG. 11 is a schematic diagram of an apparatus for monitoring the sleep state of an infant according to an embodiment of the present invention;
fig. 12 is a schematic structural diagram of an apparatus for monitoring a sleep state of an infant according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, a method for monitoring a sleep state of an infant according to an embodiment of the present invention includes the steps of:
s1, acquiring the biological characteristics of the baby and generating a characteristic value;
s2, judging whether the characteristic value exceeds a preset range of the characteristic value in the sleep state;
and S3, if yes, determining that the baby is in the non-sleep state.
In this embodiment, as described in S1, the biometric characteristic refers to a physiological characteristic or a behavioral characteristic of the infant. Infants exhibit different biological characteristics when awake and asleep, respectively. For example, an infant may be left open with eyes awake and may be sleeping with eyes closed; or the baby wakes up from sleep and sounds crying to attract the attention of the parents, and is quite quiet when in the sleep state. The monitoring system collects the biological characteristics of the facial expression of the infant through the camera, and then calculates characteristic values corresponding to the facial expression through related models, wherein each characteristic value represents the dynamic state of different organs of the infant, such as 61 represents the eye opening, 62 represents the eye closing, 81 represents the mouth opening, 82 represents the mouth closing and the like. The preset range of the characteristic value in the sleep state is a critical value which is preset by a user and is used for judging that the infant is in the sleep state; or a critical range, where the critical range is a set of one or more values. Specifically, the critical ranges are collectively expressed as {61, 81 }. If the feature value 62 is generated, it is not in the set, it is determined that the infant is in a sleeping state, and if it is generated 81, it is determined that the infant is in an unssomnia state within the set.
Referring to fig. 2, the step of determining that the infant is in the non-sleep state includes:
and S4, sending the information to the appointed terminal.
In this embodiment, the monitoring system determines that the infant is in an unsnap state, and sends information that the infant is not asleep to the designated terminal. The appointed terminal is preset by a user, can be a mobile phone of a parent or a guardian, and can also be an APP server corresponding to the monitoring system, and the parent or the guardian receives the information that the baby is not asleep through the APP. The transmitted information may also be a photograph including a state where the infant is not sleeping. The monitoring system is provided with an SIM card and can send messages to a specified mobile phone in the forms of short messages, calls, mails and the like. Or, the monitoring system sends a message to a server through the SIM card or the WiFi connection network, and the server sends a push message to the specified mobile phone.
Referring to fig. 3, in another embodiment, the step of determining that the infant is in the non-sleep state includes the following steps:
and S5, controlling the loudspeaker to play the designated audio file.
In another embodiment, the speaker is hardware that emits sound in one of the subject monitoring systems that perform the method. The monitoring system also comprises a memory for storing the specified audio file. When the baby is not sleeping, and may be crying, it is necessary to find the parent or other device to go to sleep again. The monitoring system calls the audio file in the storage device and outputs the audio file through the loudspeaker, so that the baby hears the audio file and goes to sleep again. The content of the audio file can be music which is helpful for the baby to sleep, or can be preset parent voice which is helpful for the baby to enter a sleep state.
In one embodiment, when the baby is determined not to be in the sleep state, the steps S4 and S5 are executed simultaneously, the notification is sent to the client on the mobile phone of the parent, and the preset voice or music of the parent is played simultaneously to calm the crying and screaming baby in time.
Referring to fig. 4, the biometric characteristic is sound, the feature value is a sound volume value, and the step of determining whether the feature value exceeds a preset range of the feature value in the sleep state of the infant includes:
and S21, judging whether the sound volume value exceeds a preset volume threshold value.
In this embodiment, the infant wakes up from the sleeping state, and no people familiar with the infant are around the infant, and the infant generally makes crying sound to attract other people. Therefore, the monitoring system acquires the sound of the baby, and the sound of breathing is very small when the baby is in a sleep state; when the baby wakes up and cries, the sound emitted by the baby is bigger. The monitoring system acquires the sound of the baby, detects the size of the sound volume value, and generates the sound volume value according to the size of the sound, wherein the unit of the sound volume value is decibel. Whether the baby is crying or not is judged by judging the volume value of the sound, so that whether the baby is in a sleeping state or not is judged.
Referring to fig. 5, the step of determining that the infant is in the non-sleep state includes:
s31, if the sound volume value exceeds a preset volume threshold value, acquiring the source position of the sound;
s32, judging whether the source position is a preset position for placing the baby or not;
and S33, if yes, determining that the baby is in the non-sleep state.
In this embodiment, the infant generally sleeps in a quiet environment. However, if other objects in the environment emit larger sound, or other places emit larger sound to cause the volume value acquired by the monitoring system to exceed the preset volume threshold, some misjudgments may be caused. Therefore, a plurality of microphones are added on the monitoring system to form a microphone array so as to acquire the source position of the sound. In the environment of the infant sleeping, the bed for the parent to put the infant sleeping is fixed, the position of the microphone array of the monitoring system is also fixed, and therefore the relative position of the infant and the microphone array is fixed, and in a specific embodiment, the preset position refers to that the bed of the infant is positioned 2 meters in the east direction of the microphone array. After the volume values of the obtained sounds by the multiple microphone arrays of the monitoring system exceed the volume threshold, the position of a sound source is calculated, if the position of the sound source is 2 meters in the east direction, the baby is judged to be the sound emitted, and then the baby is judged to be in an unsnap state.
Referring to fig. 6, the step of acquiring the biometric characteristic of the infant and generating the feature value further includes:
s11, acquiring a picture corresponding to the body posture of the current baby;
s12, inputting the picture into a model trained on the basis of a neural network for calculation to obtain the characteristic vector value;
the step of judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state comprises the following steps:
and S22, judging whether the characteristic vector value exceeds a preset characteristic vector threshold value.
In this embodiment, the body posture of the infant is the posture of the whole body of the infant, and includes the movements of the limbs, the torso, the head (including the facial organs), and other parts of the whole body. The baby is photographed, the whole body of the baby is photographed into the picture, and the biological characteristic of the posture of the baby is obtained. Then, inputting the picture into the model trained by the neural network for calculation, and outputting a characteristic value to obtain a characteristic vector value corresponding to the body posture of the baby. The feature vector threshold is a vector feature value output after the picture of the infant sleeping is input into the trained model.
The steps of training the model based on the neural network are as follows:
collecting a large number of sample pictures of awakening and sleeping of infants as training samples;
initializing weights and biases of neuron connections;
input forward transfer: the sample input is transmitted forward through each layer of CNN, and the output of each neuron is calculated;
the output layer outputs the class of the set of training data: comparing the category with expected output of training data, and if the error does not accord with a preset threshold value, starting a back propagation algorithm process; if the error meets a preset threshold, the CNN algorithm is terminated;
if the back propagation algorithm (BPTT) is started, calculating the error of the neuron in the previous layer by using the error of the output layer, and calculating the error on the neurons of all the hidden layers and the input layer progressively from back to front in the mode;
according to the error calculated above, adopting a gradient descent algorithm from an input layer to an output layer, and updating the connected weight and the bias of the neuron layer by layer;
and terminating the loop calculation until the error meets a preset threshold value. Thereby obtaining the trained parameters.
The picture including the body posture generated in the step S11 is input to the CNN recognition system (using the parameters obtained by the above training), and the output layer outputs the characteristic vector value after layer-by-layer calculation.
In summary, the method for monitoring the sleep state of the infant of the present invention determines whether the infant is in the sleep state according to the biological characteristics of the infant, and does not need to manually monitor the sleep state of the infant, thereby saving the energy of parents or guardians. After the baby is not asleep, a message is sent to a preset mobile phone to remind parents or guardians of pacifying, or a preset audio file is automatically played to pacify the baby, so that the pressure of the parents or guardians is reduced.
Referring to fig. 7, the present invention also provides an apparatus for monitoring a sleep state of an infant, including:
the acquisition module 1 is used for acquiring the biological characteristics of the infant and generating a characteristic value;
the judging module 2 is used for judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state;
and the judging module 3 is used for judging that the infant is in the non-sleeping state if the characteristic value exceeds the preset range of the characteristic value in the infant sleeping state.
In this embodiment, the biological characteristics refer to physiological characteristics or behavioral characteristics of the infant. Infants exhibit different biological characteristics when awake and asleep, respectively. For example, an infant may be left open with eyes awake and may be sleeping with eyes closed; or the baby wakes up from sleep and sounds crying to attract the attention of the parents, and is quite quiet when in the sleep state. The acquisition module 1 acquires a biological feature of the facial expression of the infant through a camera, and then calculates feature values corresponding to the facial expression through related models, wherein each feature value represents the dynamics of different organs of the infant, such as 61 represents that eyes are open, 62 represents that eyes are closed, 81 represents that a mouth is open, 82 represents that the mouth is closed, and the like. The preset range of the characteristic value in the sleep state is a critical value which is preset by a user and is used for judging that the infant is in the sleep state; or a critical range, where the critical range is a set of one or more values. Specifically, the critical ranges are collectively expressed as {61, 81 }. If the feature value 62 is generated, the judging module 2 judges that the feature value is not in the set, the judging module 3 judges that the infant is in the sleeping state, and if 81 is generated, the judging module 3 judges that the feature value is in the set range, and the judging module 3 judges that the infant is in the non-sleeping state.
Referring to fig. 8, further, the apparatus for monitoring the sleep state of the infant includes:
and the sending module 4 is used for sending the information to the appointed terminal.
In this embodiment, the determining module 3 determines that the infant is in an unsnap state, and then the sending module 4 sends information that the infant is not asleep to the designated terminal. The appointed terminal is preset by a user, can be a mobile phone of a parent or a guardian, and can also be an APP server corresponding to the monitoring system, and the parent or the guardian receives the information that the baby is not asleep through the APP. The transmitted information may also be a photograph including a state where the infant is not sleeping. The monitoring system is provided with an SIM card and can send messages to a specified mobile phone in the forms of short messages, calls, mails and the like. Or the sending module 4 sends the message to the server through the SIM card or the WiFi connection network, and the server sends the push message to the specified mobile phone.
Referring to fig. 9, in another embodiment, the apparatus for monitoring the sleep state of an infant includes:
and the playing module 5 is used for controlling the loudspeaker to play the specified audio file.
In another embodiment, the playing module 5 further has a memory for storing the specified audio file. When the baby is not sleeping, and may be crying, it is necessary to find the parent or other device to go to sleep again. The playing module 5 calls the audio file in the storage device, and the audio file is output through the loudspeaker, so that the baby hears the audio file and goes to sleep again. The content of the audio file may be some music that helps the infant sleep, and the playing module 5 may also play a preset parent voice to help the infant enter the sleep state.
In a specific embodiment, when the determining module 3 determines that the baby is in the non-sleeping state, the sending module 4 and the playing module 5 work simultaneously, the sending module 4 sends a notification to a client on the mobile phone of the parent, and the playing module 5 plays preset voice or music of the parent at the same time to calm the crying baby in time.
Referring to fig. 10, the biometric characteristic is a voice, the feature value is a voice volume value, and the determining module 2 includes:
the first judging unit 21 is configured to judge whether the sound volume value exceeds a preset volume threshold.
In this embodiment, the infant wakes up from the sleeping state, and no people familiar with the infant are around the infant, and the infant generally makes crying sound to attract other people. Therefore, the acquisition module 1 acquires the sound of the baby, and the baby in a sleep state makes a breathing sound which is very small; when the baby wakes up and cries, the sound emitted by the baby is bigger. When the sound of the baby is acquired, the first judgment unit 21 detects the magnitude of the sound volume value, and generates the sound volume value according to the magnitude of the sound, wherein the unit of the sound volume value is decibel. The first judgment unit 21 judges whether the baby is crying by judging the volume value of the sound, thereby judging whether the baby is in a sleeping state.
Referring to fig. 11, the determining module 3 further includes:
a location unit 31, configured to obtain a source location of the sound if the sound volume value exceeds a preset volume threshold;
a second determining unit 32, configured to determine whether the source location is a preset location for placing an infant;
the determination unit 33 is configured to determine that the infant is in the non-sleep state if the source position is a preset position for placing the infant.
In this embodiment, the infant generally sleeps in a quiet environment. However, if other objects in the environment emit larger sound, or other places emit larger sound to cause the volume value acquired by the monitoring system to exceed the preset volume threshold, some misjudgments may be caused. Thus, the position unit 31 has a plurality of microphones forming a microphone array, and the position unit 31 acquires the source position of the sound. In the environment of the infant sleeping, the bed for the parent to put the infant sleeping is fixed, the position for installing the microphone array is also fixed, therefore, the relative position of the infant and the microphone array is fixed, and in a specific embodiment, the preset position refers to that the bed of the infant is positioned 2 meters in the east direction of the microphone array. After the volume values of the acquired sounds by the multiple microphone arrays of the position unit 31 exceed the volume threshold, the position of the sound source is calculated, and the second judging unit 32 judges that if the position of the sound source is 2 meters in the east direction, the judging unit 33 judges that the baby emits the sounds, and further judges that the baby is in the non-sleep state.
Referring to fig. 12, the biometric characteristic is a body posture of the infant, and the feature value is a feature vector value corresponding to the body posture, and the obtaining module 1 includes:
the acquisition unit 11 is used for acquiring a picture corresponding to the current body posture of the baby;
the calculation unit 12 is configured to input the picture into a model trained based on a neural network to perform calculation, so as to obtain the characteristic vector value;
the judging module 2 includes:
a third determining unit 22, configured to determine whether the feature vector value exceeds a preset feature vector threshold.
In this embodiment, the body posture of the infant is the posture of the whole body of the infant, and includes the movements of the limbs, the torso, the head (including the facial organs), and other parts of the whole body. The acquisition unit 11 takes a picture of the baby, and takes the whole body of the baby into the picture, namely, the biological characteristic of the posture of the baby is acquired. Then, the calculation unit 12 inputs the image into the model trained by the neural network for calculation, and outputs a characteristic value, namely a characteristic vector value corresponding to the body posture of the infant. The feature vector threshold is a vector feature value output after the picture of the infant sleeping is input into the trained model. The third judgment unit 22 judges whether the feature vector value exceeds a preset feature vector threshold value.
The steps of training the model based on the neural network are as follows:
collecting a large number of sample pictures of awakening and sleeping of infants as training samples;
initializing weights and biases of neuron connections;
input forward transfer: the sample input is transmitted forward through each layer of CNN, and the output of each neuron is calculated;
the output layer outputs the class of the set of training data: comparing the category with expected output of training data, and if the error does not accord with a preset threshold value, starting a back propagation algorithm process; if the error meets a preset threshold, the CNN algorithm is terminated;
if the back propagation algorithm (BPTT) is started, calculating the error of the neuron in the previous layer by using the error of the output layer, and calculating the error on the neurons of all the hidden layers and the input layer progressively from back to front in the mode;
according to the error calculated above, adopting a gradient descent algorithm from an input layer to an output layer, and updating the connected weight and the bias of the neuron layer by layer;
and terminating the loop calculation until the error meets a preset threshold value. Thereby obtaining the trained parameters.
The pictures containing the body gestures, which are generated by the acquisition unit 11, are input into a CNN recognition system (parameters obtained by the training are adopted), and the output layer outputs characteristic vector values through layer-by-layer calculation.
In summary, the device for monitoring the sleep state of the infant of the present invention determines whether the infant is in the sleep state according to the biological characteristics of the infant, and does not need to monitor the sleep state of the infant manually, thereby saving the energy of parents or monitoring people. After the baby is not asleep, a message is sent to a preset mobile phone to remind parents or guardians of pacifying, or a preset audio file is automatically played to pacify the baby, so that the pressure of the parents or guardians is reduced.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (6)
1. A method of monitoring the sleep state of an infant, comprising the steps of:
acquiring biological characteristics of the infant to generate characteristic values;
when the biological feature is sound and the feature value is a sound volume value, judging whether the sound volume value exceeds a preset volume threshold value;
if the sound volume value exceeds a preset volume threshold value, acquiring the source position of the sound through a microphone array;
judging whether the source position is a preset position for arranging the baby or not according to the relative position of the source position and the microphone array and the relative position of a preset position for arranging the baby and the microphone array; wherein a relative position of a preset position where the baby is placed and the microphone array is fixed;
if so, judging that the infant is in the non-sleep state;
the monitoring system acquires biological characteristics of facial expressions of infants through the camera, calculates characteristic values corresponding to the facial expressions through related models, and each characteristic value represents the dynamics of different organs of the infants; the characteristic value in the sleep state is a critical range, and the critical range is a set of a plurality of values; if the generated feature value is not in the set, the infant is determined to be in an unsnap state.
2. The method of monitoring the sleep state of an infant as set forth in claim 1, wherein the step of determining that the infant is in the non-sleep state is followed by the steps of:
sending information to a designated terminal; and/or
And controlling the loudspeaker to play the specified audio file.
3. The method of claim 1, wherein the biometric characteristic is a body posture of the infant, the feature value is a feature vector value corresponding to the body posture, and the step of obtaining the biometric characteristic of the infant and generating the feature value comprises:
acquiring a picture corresponding to the current body posture of the baby;
inputting the picture into a model trained based on a neural network for calculation to obtain the characteristic vector value;
the step of judging whether the characteristic value exceeds a preset range of the characteristic value in the sleep state comprises the following steps:
and judging whether the characteristic vector value exceeds a preset characteristic vector threshold value.
4. An apparatus for monitoring the sleep state of an infant, comprising:
the acquisition module is used for acquiring the biological characteristics of the infant and generating a characteristic value;
the judging module is used for judging whether the characteristic value exceeds a preset range of the characteristic value of the infant in the sleep state; the judgment module comprises a first judgment unit and a second judgment unit, wherein the biological feature is sound, the characteristic value is a sound volume value, and the first judgment unit is used for judging whether the sound volume value exceeds a preset volume threshold value;
the judging module is used for judging that the infant is in the non-sleeping state if the characteristic value exceeds the preset range of the characteristic value in the infant sleeping state;
the determination module includes: the position unit is used for acquiring the source position of the sound through a microphone array if the sound volume value exceeds a preset volume threshold;
the second judging unit is used for judging whether the source position is the preset position for arranging the baby or not according to the relative position of the source position and the microphone array and the relative position of the preset position for arranging the baby and the microphone array; wherein a relative position of a preset position where the baby is placed and the microphone array is fixed;
the judging unit is used for judging that the baby is in the non-sleeping state if the source position is a preset position for placing the baby;
the acquisition module acquires biological characteristics of facial expressions of infants through the camera, and calculates characteristic values corresponding to the facial expressions through related models, wherein each characteristic value represents the dynamics of different organs of the infants; the characteristic value in the sleep state is a critical range, and the critical range is a set of a plurality of values; if the judgment module judges that the generated characteristic value is not in the set, the judgment module judges that the infant is in the state of not sleeping.
5. The apparatus for monitoring the sleep state of an infant as in claim 4, further comprising:
the sending module is used for sending information to the appointed terminal;
and/or
And the playing module is used for controlling the loudspeaker to play the specified audio file.
6. The apparatus for monitoring the sleep state of an infant as claimed in claim 4, wherein the biometric characteristic is the body posture of the infant, the characteristic value is a characteristic vector value corresponding to the body posture, and the obtaining module comprises:
the acquisition unit is used for acquiring a picture corresponding to the current body posture of the baby;
the calculation unit is used for inputting the picture into a model trained on the basis of a neural network to calculate to obtain the characteristic vector value;
the judging module comprises:
and the third judging unit is used for judging whether the characteristic vector value exceeds a preset characteristic vector threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810362787.4A CN108670196B (en) | 2018-04-20 | 2018-04-20 | Method and device for monitoring sleep state of infant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810362787.4A CN108670196B (en) | 2018-04-20 | 2018-04-20 | Method and device for monitoring sleep state of infant |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108670196A CN108670196A (en) | 2018-10-19 |
| CN108670196B true CN108670196B (en) | 2021-11-09 |
Family
ID=63802074
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810362787.4A Active CN108670196B (en) | 2018-04-20 | 2018-04-20 | Method and device for monitoring sleep state of infant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108670196B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109819111A (en) * | 2019-01-29 | 2019-05-28 | 上海与德通讯技术有限公司 | A kind of control method, device, equipment and storage medium |
| CN110706449A (en) * | 2019-09-04 | 2020-01-17 | 中移（杭州）信息技术有限公司 | Infant monitoring method and device, camera equipment and storage medium |
| CN111768593A (en) * | 2020-06-29 | 2020-10-13 | 歌尔科技有限公司 | State monitoring method and device, intelligent wearable device and storage medium |
| CN111787278A (en) * | 2020-06-30 | 2020-10-16 | 百度在线网络技术（北京）有限公司 | User state monitoring method, intelligent sound box and electronic equipment |
| CN113505742B (en) * | 2021-07-27 | 2024-05-10 | 宁波星巡智能科技有限公司 | Infant sleep tracking and monitoring method, device, equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103561094A (en) * | 2013-11-04 | 2014-02-05 | 成都数之联科技有限公司 | Intelligent monitoring method for sleep condition of infant |
| WO2015078879A1 (en) * | 2013-11-27 | 2015-06-04 | Universite De Rennes I | Method for constructing an activity index, device, and computer program therefore |
| CN105489217A (en) * | 2015-12-30 | 2016-04-13 | 武传胜 | Method and system of monitoring the aged through sound by means of internet of things |
| CN205211106U (en) * | 2015-08-26 | 2016-05-04 | 马晓阳 | Can realize mobile terminal real -time communication's crib monitoring alarm system |
| CN105678960A (en) * | 2016-03-16 | 2016-06-15 | 上海斐讯数据通信技术有限公司 | Sleep security monitoring method and system |
| CN106094539A (en) * | 2016-06-14 | 2016-11-09 | 李敬 | Device and desk lamp and toy and the client matched are slept in intelligence baby's roars of laughter |
| CN205696853U (en) * | 2016-04-06 | 2016-11-23 | 杭州电子科技大学 | A kind of multifunctional baby cradle |
| CN106510295A (en) * | 2016-10-27 | 2017-03-22 | 乐视控股（北京）有限公司 | Intelligent infant bed for infant nursing, server and method |
| CN107453964A (en) * | 2017-07-21 | 2017-12-08 | 北京小米移动软件有限公司 | Sleep environment management method and device |
| CN206921268U (en) * | 2017-04-14 | 2018-01-23 | 北京克路德人工智能科技有限公司 | Intelligent monitor system based on auditory localization |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996469A (en) * | 2010-11-30 | 2011-03-30 | 付文军 | Early-warning method for infant motion accidents as well as intelligent infant nursing device |
| CN107184297A (en) * | 2017-03-29 | 2017-09-22 | 广西喜爱家饮水设备有限公司 | A kind of purifying drinking water system guarded for old man |
| US9924921B1 (en) * | 2017-06-20 | 2018-03-27 | Qualcomm Incorporated | System for mapping joint performance |
-
2018
- 2018-04-20 CN CN201810362787.4A patent/CN108670196B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103561094A (en) * | 2013-11-04 | 2014-02-05 | 成都数之联科技有限公司 | Intelligent monitoring method for sleep condition of infant |
| WO2015078879A1 (en) * | 2013-11-27 | 2015-06-04 | Universite De Rennes I | Method for constructing an activity index, device, and computer program therefore |
| CN205211106U (en) * | 2015-08-26 | 2016-05-04 | 马晓阳 | Can realize mobile terminal real -time communication's crib monitoring alarm system |
| CN105489217A (en) * | 2015-12-30 | 2016-04-13 | 武传胜 | Method and system of monitoring the aged through sound by means of internet of things |
| CN105678960A (en) * | 2016-03-16 | 2016-06-15 | 上海斐讯数据通信技术有限公司 | Sleep security monitoring method and system |
| CN205696853U (en) * | 2016-04-06 | 2016-11-23 | 杭州电子科技大学 | A kind of multifunctional baby cradle |
| CN106094539A (en) * | 2016-06-14 | 2016-11-09 | 李敬 | Device and desk lamp and toy and the client matched are slept in intelligence baby's roars of laughter |
| CN106510295A (en) * | 2016-10-27 | 2017-03-22 | 乐视控股（北京）有限公司 | Intelligent infant bed for infant nursing, server and method |
| CN206921268U (en) * | 2017-04-14 | 2018-01-23 | 北京克路德人工智能科技有限公司 | Intelligent monitor system based on auditory localization |
| CN107453964A (en) * | 2017-07-21 | 2017-12-08 | 北京小米移动软件有限公司 | Sleep environment management method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108670196A (en) | 2018-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108670196B (en) | Method and device for monitoring sleep state of infant | |
| KR100838099B1 (en) | Automatic system for monitoring independent person requiring occasional assistance | |
| JP2019162714A (en) | Robot recognizing direction of sound source | |
| JP2004531800A (en) | Automated system for monitoring persons requiring monitoring and their caretakers | |
| CN109480868B (en) | Intelligent infant monitoring system | |
| JP2013154458A (en) | Device, program, and method for conversation | |
| US11514269B2 (en) | Identification device, robot, identification method, and storage medium | |
| US20210295668A1 (en) | Alert system | |
| JP6900058B2 (en) | Personal assistant control system | |
| US11219409B2 (en) | Method and system for user interaction through user posture recognition | |
| JP2020089947A (en) | Information processing device, information processing method, and program | |
| WO2016206644A1 (en) | Robot control engine and system | |
| JP2021128350A (en) | Information processing system, information processing method, and recording medium | |
| JP7296626B2 (en) | Information processing device and program | |
| WO2023171708A1 (en) | Information processing system, information processing method, and program | |
| CN114828737A (en) | System and method for alerting a caregiver based on the status of a person in need thereof | |
| KR102196779B1 (en) | Artificial intelligence based necklace type wearable device | |
| JP7254345B2 (en) | Information processing device and program | |
| CN110427155B (en) | Nursing method and device | |
| JP2012010856A (en) | State recognition device, listening interaction continuing system, state recognition program and state recognition method | |
| KR102668931B1 (en) | Apparatus and method for providing customized service | |
| CN115281487B (en) | Sleep monitoring method based on intelligent mattress and related products | |
| JP5602753B2 (en) | A toy showing nostalgic behavior | |
| JP6900089B2 (en) | Personal assistant control system | |
| US20240096239A1 (en) | Remote breathing and vocalization simulating device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210922 Address after: 518000 201, No.26, yifenghua Innovation Industrial Park, Xinshi community, Dalang street, Longhua District, Shenzhen City, Guangdong Province Applicant after: Shenzhen waterward Information Co.,Ltd. Address before: 518000, block B, huayuancheng digital building, 1079 Nanhai Avenue, Shekou, Nanshan District, Shenzhen City, Guangdong Province Applicant before: SHENZHEN WATER WORLD Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |