CN110506661A - A kind of pet that prevents based on machine learning is had a fist fight method - Google Patents
A kind of pet that prevents based on machine learning is had a fist fight method Download PDFInfo
- Publication number
- CN110506661A CN110506661A CN201910820242.8A CN201910820242A CN110506661A CN 110506661 A CN110506661 A CN 110506661A CN 201910820242 A CN201910820242 A CN 201910820242A CN 110506661 A CN110506661 A CN 110506661A
- Authority
- CN
- China
- Prior art keywords
- pet
- wearable device
- identity information
- data
- machine 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.)
- Granted
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000006399 behavior Effects 0.000 claims abstract description 67
- 230000002996 emotional effect Effects 0.000 claims abstract description 50
- 238000004891 communication Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims description 27
- 238000005516 engineering process Methods 0.000 claims description 23
- 241000700605 Viruses Species 0.000 claims description 19
- 230000036387 respiratory rate Effects 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 13
- 230000007958 sleep Effects 0.000 claims description 12
- 230000001580 bacterial effect Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 10
- 230000036772 blood pressure Effects 0.000 claims description 9
- 230000036760 body temperature Effects 0.000 claims description 9
- 230000008451 emotion Effects 0.000 claims description 9
- 230000001755 vocal effect Effects 0.000 claims description 9
- 230000035479 physiological effects, processes and functions Effects 0.000 claims description 7
- 230000036651 mood Effects 0.000 claims description 4
- 210000000629 knee joint Anatomy 0.000 claims description 3
- 241000894006 Bacteria Species 0.000 claims 2
- 241001269238 Data Species 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 230000009385 viral infection Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000010010 raising Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000003612 virological effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K15/00—Devices for taming animals, e.g. nose-rings or hobbles; Devices for overturning animals in general; Training or exercising equipment; Covering boxes
- A01K15/02—Training or exercising equipment, e.g. mazes or labyrinths for animals ; Electric shock devices ; Toys specially adapted for animals
- A01K15/021—Electronic training devices specially adapted for dogs or cats
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Environmental Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physical Education & Sports Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Zoology (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Toys (AREA)
Abstract
Pet has a fist fight method that the invention discloses a kind of based on machine learning prevent, which comprises the first pet wearable device, which detects, whether there is the second pet wearable device in pre-determined distance；The first pet wearable device and the second pet wearable device establish communication connection；The first pet wearable device and the second pet wearable device open video recording, recording；Every preset time, emotional state, behavior intention and the physiological data of first pet, emotional state, behavior intention and the physiological data of second pet are obtained, executes pre-set commands in conjunction with default rule；The first pet wearable device and the second pet wearable device terminate video recording, recording；Obtain pet order machine learning model training data, pet order machine learning model described in re -training.The method of the present invention combines machine learning and pet physiological status data, finally more effectively prevents the generation of pet fighting incident.
Description
Technical field
The invention belongs to pet control technology field more particularly to a kind of the pet side of having a fist fight is prevented based on machine learning
Method.
Background technique
With the continuous improvement of society being constantly progressive with living standard, more and more family's raisings have pet, and raise
Keep a pet, especially domestic pets, be one and take a lot of trouble laborious work, many people work it is busy etc. due to usually lack when
Between accompany oneself to go outdoor sports by pet, so sometimes people can put pet oneself and go out for a walk, but because people are not doting on
Object at one's side, will worry pet alone in it is outer when situation.Pet as owner not at one's side, mood often becomes anxious
It is hot-tempered, outside, thing can be baited, is jumped everywhere, is arbitrarily made a sound, or even fighting incident occurs with other people pet.Gently
Neighbours' life can be bothered when micro-, communal facility is damaged, can cause accident when serious, cause pet injured, to economic and life zone
To threaten.
So pet how to be prevented to have a fist fight when going out alone with other pets, pet owner how to be made to monitor oneself
The case where when pet is gone out alone, if necessary issues pet and instructs, and is to be worth solving the problems, such as.
Summary of the invention
The purpose of the present invention is to solve the defect of the above-mentioned prior art, provide a kind of prevents from doting on based on machine learning
Object is had a fist fight method.
The second object of the present invention is to provide a kind of wearable device for pet.
Another object of the present invention is to provide a kind of for assigning the client of simulation command.
The purpose of the present invention can be reached by adopting the following technical scheme that:
A kind of pet that prevents based on machine learning is had a fist fight method characterized by comprising
First pet wearable device, which detects, whether there is the second pet wearable device in pre-determined distance.
The first pet wearable device and the second pet wearable device establish communication connection.
The first pet wearable device and the second pet wearable device open video recording, recording.
Every preset time, emotional state, behavior intention and the physiological data of acquisition first pet, described second are doted on
Emotional state, behavior intention and the physiological data of object.
Pre-set commands are executed in conjunction with default rule.
The first pet wearable device and the second pet wearable device terminate video recording, recording.
Preferably, it is set in the first pet wearable device detection pre-determined distance with the presence or absence of the second pet is wearable
It is standby, comprising:
In pre-determined distance, returned if the first pet wearable device can receive the second pet wearable device
The identity information of the preset format returned then judges in pre-determined distance there are the second pet wearable device, if described the
One pet wearable device can not receive the identity information for the preset format that the second pet wearable device returns, then sentence
The second pet wearable device is not present in disconnected pre-determined distance.
The identity information of the preset format, includes at least: can identify the pet ID of pet identity.
Preferably, the first pet wearable device and the second pet wearable device establish communication connection, wrap
It includes:
The first pet wearable device sends the first identity information to the second pet wearable device, and described second dotes on
Object wearable device sends the second identity information to the first pet wearable device.
First identity information is the data of the first pet wearable device storage, is specifically included: pet ID, thin
Bacterial content, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment times
Sleep state baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency.
Second identity information is the data of the second pet wearable device storage, is specifically included: pet ID, thin
Bacterial content, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment times
Sleep state baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency.
The first pet wearable device, according to the second pet image, in conjunction with image recognition technology, described in detection
Whether the pet for including in the second pet image and the pet for including in the pet photo in second identity information are same
Pet, if it is not, by second identity information pet ID, bacterial content, whether carry virus, weight, gender,
Whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment time sleep state is baited power, intellectual level, pet and is shone
Piece, the sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown.
The second pet wearable device, according to the first pet image, in conjunction with image recognition technology, described in detection
Whether the pet for including in the first pet image and the pet for including in the pet photo in first identity information are same
Pet, if it is not, by first identity information pet ID, bacterial content, whether carry virus, weight, gender,
Whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment time sleep state is baited power, intellectual level, pet and is shone
Piece, the sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown.
Preferably, described every preset time, obtain emotional state, behavior intention and the physiology number of first pet
According to emotional state, behavior intention and the physiological data of second pet, comprising:
By the first pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the sound letter of the first pet are obtained
Breath.
The feelings of first pet are obtained in conjunction with sound Emotion identification technology according to the acoustic information of first pet
Not-ready status.
First pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of first pet
Behavior be intended to.
By the second pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the sound of the second pet are obtained
Information.
The feelings of second pet are obtained in conjunction with sound Emotion identification technology according to the acoustic information of second pet
Not-ready status.
Second pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of second pet
Behavior be intended to.
The emotional state, comprising: indignation, it is happy, it is tranquil, it is frightened.
The behavior is intended to, comprising: warning is attacked, entreat.
Preferably, the combination default rule executes pre-set commands, comprising:
By first identity information, second identity information, the emotional state of first pet, behavior intention and physiology
Data, the emotional state of second pet, behavior intention and physiological data are as command determination factor data, in conjunction with preset
Rule executes pre-set commands, specifically includes:
A) judge that the pet ID in second identity information whether there is in the blacklist of the first pet wearable device
In, if it does, executing the voice messaging for playing " escaping ", it is performed simultaneously connection owner's client, requests and holds
Row owner's remote command.
B) when the emotional state of first pet is angry, in first identity information and second identity information
Gender it is identical, in second identity information exist have a fist fight record, second pet have attack behavior be intended to when, execute
The voice messaging for playing " escaping ".
C) when the emotional state of first pet is happiness, the emotional state of second pet is happiness, described the
One identity information is different with the gender in second identity information, executes the voice messaging for playing " going to play with other side ".
D) when the emotional state of first pet is indignation, the emotional state of second pet is indignation, described the
One pet has the behavior of attack to be intended to, and when second pet has the behavior of attack to be intended to, executes the action and stops order, limit
Make the ability to act of first pet.Pet ID in second identity information is piped off, is stored into described
In one pet wearable device.
E) it when the behavior of first pet is intended to attack, and the behavior of second pet is intended to entreat, executes
The electric shock order.
It f) is more than the second threshold more than the respiratory rate of first threshold or first pet when the heart rate of first pet
When value, allows first pet to feel the comfortable vibration frequency by first identity information acquisition, dote on according to described first
Object feels that the comfortable vibration frequency executes the vibration order, until the heart rate of first pet is restored to lower than first threshold
Or the respiratory rate of first pet then stops the vibration order when being restored to lower than second threshold.
G) judge whether the pet in second identity information carries Virus Info, if carrying virus, described in execution
The voice messaging for playing " escaping ", is performed simultaneously connection owner's client, requests and execute owner's remote command.
H) it when the behavior of first pet is intended to alert, executes the sound for obtaining other side pet fear and threatens pair
Side's order.
Preferably, the combination default rule executes pre-set commands, further includes:
Obtain command determination factor data sample set and corresponding simulation command sample set；Based on command determination factor data sample
Collection and the corresponding simulation command sample set training pet order machine learning model；The command determination factor data is defeated
Enter to pet order machine learning model, determines output order corresponding with the command determination factor data, described first dotes on
Object wearable device executes the pet order machine learning model output order.
Preferably, the acquisition command determination factor data sample set and corresponding simulation command sample set, comprising:
Video data, the recording data of first pet wearable device described in the first owner client terminal playing, described in displaying
Emotional state, behavior intention and the physiology of first pet of the first identity information, second identity information and each time point
Data, the emotional state of the second pet, behavior intention and physiological data receive the owner's of the first pet wearable device
Simulation command operation, obtains the command determination factor data sample set and the corresponding simulation command sample set.
Preferably, the first pet wearable device and the second pet wearable device terminate video recording, recording, packet
It includes:
Whether detect the first pet wearable device at a distance from the second pet wearable device is more than described default
Distance, if it does, the first pet wearable device and the second pet wearable device terminate video recording, recording.
The second object of the present invention can be reached by adopting the following technical scheme that:
A kind of wearable device for pet, it is characterised in that the wearable device includes:
Wireless communication device is wirelessly connected for establishing with other electronic equipments.
Parameter acquisition devices, in-built motion sensor and pressure sensor, mainly for detection of described wearable set is worn
The heart rate of standby pet, blood pressure, body temperature, respiratory rate.
Recording apparatus, for recording the image data in front of the pet for wearing the wearable device.
Recording device, for recording the audio data around the pet for wearing the wearable device.
Processor is mainly used for executing image recognition operation, the operation of sound Emotion identification, vocal behavior intention assessment fortune
The logical operation calculate, issued an order in conjunction with default rule to pet.
Audio playing apparatus, boombox are mainly used for playing audio-frequency information.
Shaking device, built-in vibrator are mainly used for executing the vibration order, and the vibration frequency of the shaking device is
It can be according to parameter regulation.
Electric shock device, built-in electric shock device are mainly used for executing the electric shock order.
Action arrester, built-in knee joint blown jacket are mainly used for limiting the ability to act of pet dog.
Another object of the present invention can be reached by adopting the following technical scheme that:
It is a kind of for assigning the client of simulation command, which is characterized in that the client includes:
Data show device, for playing Video data, the recording data of the first pet wearable device, show described the
Emotional state, behavior intention and the physiology number of first pet of one identity information, second identity information and each time point
According to, the emotional state of the second pet, behavior intention and physiological data.
Simulation command device, the simulation command operation of the owner for receiving the first pet wearable device, obtains
Machine learning model training data.
Remote command device, for sending remote command to the first pet wearable device.
The technical solution that the disclosure proposes has the advantages that
The disclosure establishes communication connection by bluetooth or wifi technology, automation, transmits identity information, the judgement for making pet meet
In time, keep interactive command more accurate by identity information；By image recognition technology, the body of wearable device storage is avoided
Part information and the unmatched situation of pet for currently dressing the wearable device, reduce error rate.By preset order, certainly
Dynamicization assigns instruction to pet, reduces and intervenes number to owner, saves the time of owner, compared to directly connection owner, energy
Issuing an order to pet much sooner, avoids conflict；By simulation command, continuous reinforcement machine learning model is accurate
Rate；Automatic video recording disconnects automatically, only carries out video recording at the crucial necessary moment and saves electric power and equipment storage space；Pass through blacklist
Mechanism reminds owner in time, assigns to pet and avoids order, reduces conflict, and owner oneself can select removal black
List more improves the chance got along to pet, more flexible.
Detailed description of the invention
Fig. 1 is that a kind of pet that prevents based on machine learning of embodiment according to the present invention is had a fist fight the flow chart of method.
Specific embodiment
Step the S101: the first pet wearable device, in conjunction with bluetooth or wifi technology, detecting whether there is in pre-determined distance
Second pet wearable device, if it does, the first pet wearable device is built with the second pet wearable device
Vertical communication connection, the first pet wearable device send the first identity information, institute to the second pet wearable device
It states the second pet wearable device and sends the second identity information to the first pet wearable device, enter step S201.
It whether there is the second pet wearable device in the detection pre-determined distance, specifically include: in pre-determined distance, such as
First pet wearable device described in fruit can receive the identity for the preset format that the second pet wearable device returns
Information then judges in pre-determined distance there are the second pet wearable device, if the first pet wearable device without
Method receives the identity information for the preset format that the second pet wearable device returns, then judges to be not present in pre-determined distance
The second pet wearable device.
The identity information of the preset format, includes at least: can identify the pet ID of pet identity.
First identity information, comprising: the pet data of the first pet wearable device storage, such as: pet
ID, bacterial content, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segments
Time sleep state baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency
Deng.
Second identity information, comprising: the pet data of the second pet wearable device storage, such as: pet
ID, bacterial content, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segments
Time sleep state baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency
Deng.
Step S201: the first pet wearable device and the second pet wearable device open video recording, recording.
Step S301: the first pet wearable device, according to the second pet image obtained by step S201, in conjunction with image
Identification technology detects the pet for including in the second pet image and includes in the pet photo in second identity information
Pet whether be same pet, if it is not, by second identity information pet ID, bacterial content, whether take
Band virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment time sleep state is baited
Power, intellectual level, pet photo, the sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown.
The second pet wearable device, according to the first pet image obtained by step S201, in conjunction with image recognition technology,
The pet for detecting the pet for including in the first pet image and including in the pet photo in first identity information is
It is no be same pet, if it is not, by first identity information pet ID, bacterial content, whether carry virus, body
Weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, proximal segment time sleep state baits power, intelligence water
Flat, pet photo, the sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown.
The first pet image is the second pet wearable device according to resulting picture of recording a video in step S201
In, the picture containing first pet.
The second pet image is the first pet wearable device according to resulting picture of recording a video in step S201
In, the picture containing second pet.
Step S401: every preset time, obtain first pet emotional state, behavior intention and physiological data,
Emotional state, behavior intention and the physiological data of second pet execute pre-set commands in conjunction with default rule, specific to wrap
It includes:
By the first pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the acoustic information of the first pet are obtained
Equal physiological datas.
The feelings of first pet are obtained in conjunction with sound Emotion identification technology according to the acoustic information of first pet
Not-ready status.
First pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of first pet
Behavior be intended to.
By the second pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the sound of the second pet are obtained
The physiological datas such as information.
The feelings of second pet are obtained in conjunction with sound Emotion identification technology according to the acoustic information of second pet
Not-ready status.
Second pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of second pet
Behavior be intended to.
First pet is the pet for dressing the first pet wearable device.
Second pet is the pet for dressing the second pet wearable device.
The emotional state, comprising: indignation, happy, tranquil, fear etc..
The behavior is intended to, comprising: warning, attack, entreat etc..
By first identity information, second identity information, the emotional state of first pet, behavior be intended to and
Physiological data, the emotional state of second pet, behavior intention and physiological data are as command determination factor data, in conjunction with pre-
If rule execute pre-set commands.
The pre-set commands, specifically include: vibration, plays " escaping " at the sound threatening other side for obtaining other side pet fear
Voice messaging, play " attentioning " voice messaging, play " low roars of laughter " voice messaging, action stop, electric shock, play " go with
The voice messaging of other side's object for appreciation ", connection owner's client, request and execute owner's remote command.
The vibration, specifically includes: the shaking device of starting the first pet wearable device makes first pet
Vibration is experienced, the vibration frequency of the shaking device can be according to parameter regulation.
The sound for obtaining other side pet fear threatens other side, specifically includes: obtaining the second pet identity information
The sound of middle pet fear starts the audio playing apparatus of the first pet wearable device, plays the second pet body
The sound of pet fear in part information.
The voice messaging for playing " escaping ", specifically includes: the audio of starting the first pet wearable device is broadcast
Device is put, the voice messaging of " escaping " is played.
The voice messaging for playing " attentioning ", specifically includes: the audio of starting the first pet wearable device is broadcast
Device is put, the voice messaging of " attentioning " is played.
The voice messaging for playing " low roars of laughter ", specifically includes: the audio of starting the first pet wearable device is broadcast
Device is put, the voice messaging of " low roars of laughter " is played.
The action stops, and specifically includes: the action arrester of starting the first pet wearable device limits institute
State the ability to act of the first pet.
The electric shock, specifically includes: the first pet wearable device obtains the weather feelings of current time current location
Condition does not enable the electric shock device of the first pet wearable device then if it is rainy weather, if not rainy weather,
The electric shock device for then enabling the electric shock device of the first pet wearable device makes first pet experience electric shock.Pass through
Weather condition judgement, can lead to the case where accidentally injuring first pet of leaking electricity to avoid rainwater.
The voice messaging for playing " going to play with other side ", specifically includes: starting the first pet wearable device
Audio playing apparatus plays the voice messaging of " going to play with other side ".
Connection owner's client, requests and execute owner's remote command, specifically include: first pet is wearable
Equipment sends to first owner's client and requests first identity information, second identity information and each time point
Emotional state, behavior intention and the physiological data of first pet, the emotional state of the second pet, behavior intention and physiological data,
According to Video data, the recording data of the first pet wearable device obtained by step S201；Receive and perform described first
The remote command that owner's client returns, the remote command include: vibration, the sound threatening pair for obtaining other side pet fear
Side, the voice messaging of broadcasting " escaping ", the voice messaging for playing " attentioning ", the voice messaging of broadcasting " low roars of laughter ", action stop, are electric
Hit, play the voice messaging of " going to play with other side ".
Pre-set commands are executed in conjunction with default rule, comprising:
A) judge that the pet ID in second identity information whether there is in the blacklist of the first pet wearable device
In, if it does, executing the voice messaging for playing " escaping ", it is performed simultaneously connection owner's client, requests and holds
Row owner's remote command.
In the present embodiment, by executing the voice messaging order for playing " escaping " connection is executed while to lead
People's client is requested and is executed owner's remote command, obtains owner's remote command, so that pet be prevented not accept " escaping " life
Surprisingly occur caused by order, also increases the controlling extent of owner.
B) when the emotional state of first pet is indignation, first identity information and second identity information
In gender it is identical, in second identity information exist have a fist fight record, second pet have attack behavior be intended to when, hold
The row voice messaging for playing " escaping ".
In the present embodiment, it when second pet has the behavior of attack to be intended to, while detecting described first and doting on
The emotional state of object, the gender in first identity information and second identity information, the bucket of second identity information
Record is beaten up, more can accurately judge danger at this time, just it is necessary to assign " escaping " order.
C) when the emotional state of first pet is happiness, the emotional state of second pet is happiness, described the
One identity information is different with the gender in second identity information, executes the voice messaging for playing " going to play with other side ".
In the present embodiment, in conjunction with the emotional state and gender of first pet and second pet, appropriate
In the case where play the voice messaging of " going to play with other side ", promote a possibility that making friends between pet.
D) when the emotional state of first pet is indignation, the emotional state of second pet is indignation, described the
One pet has the behavior of attack to be intended to, and when second pet has the behavior of attack to be intended to, executes the action and stops order, limit
Make the ability to act of first pet.Pet ID in second identity information is piped off, is stored into described
In one pet wearable device.
In the present embodiment, it is intended to by the emotional state and behavior that detect first pet and second pet,
When the two pet has attack to be inclined to and is angry, timely restriction ability avoids unexpected generation.
E) it when the behavior of first pet is intended to attack, and the behavior of second pet is intended to entreat, executes
The electric shock order.
In the present embodiment, it is intended to by detecting the behavior of first pet and second pet, to described first
The attack tendency of pet carries out electric shock punishment.
It f) is more than the second threshold more than the respiratory rate of first threshold or first pet when the heart rate of first pet
When value, allows first pet to feel the comfortable vibration frequency by first identity information acquisition, dote on according to described first
Object feels that the comfortable vibration frequency executes the vibration order, until the heart rate of first pet is restored to lower than first threshold
Or the respiratory rate of first pet then stops the vibration order when being restored to lower than second threshold.
In the present embodiment, according to the physiological data of first pet, when the physiological data of first pet is in
When abnormal, such as heart rate when breathing rate is more than 15 beats/min of second threshold, is incited somebody to action more than 150 beats/min of first threshold
The vibration frequency of shaking device, which is transferred to, to be made first pet feel the comfortable vibration frequency and shakes, and pet is made to feel pleased
It is happy, until restoring normal physiological status.
G) judge whether the pet in second identity information carries Virus Info, if there is virus, described in execution
The voice messaging for playing " escaping ", is performed simultaneously connection owner's client, requests and execute owner's remote command.
In the present embodiment, when determining the pet carrying Virus Info in second identity information, in time to described
First pet plays the voice messaging of " escaping " while connecting owner's client of the first pet wearable device, and request is simultaneously
Owner's remote command is executed, the probability for the virus infection that first pet is carried by second pet is reduced, is made described
The owner of first pet can monitor first pet and the case where viral source contacts, for after main occurrences in human life to virus infection
Judgement processing provides foundation.
H) it when the behavior of first pet is intended to alert, executes the sound for obtaining other side pet fear and threatens pair
Side's order.
Pre-set commands are executed in conjunction with default rule, further includes:
The command determination factor data is input to pet order machine learning model, the determining and command determination is because of prime number
It is ordered according to corresponding output, is exported and ordered according to the pet order machine learning model, execute the output order.
Wherein the pet order machine learning model is to be in advance based on command determination factor data sample set and corresponding
The training of simulation command sample set is formed.
The training algorithm of the pet order machine learning model can be svm, deep learning, naive Bayesian etc., institute
The training process for stating pet order machine learning model is technology known in those skilled in the art, and this will not be repeated here.
The pet order machine learning model output order, comprising: vibration, the sound threatening for obtaining other side pet fear
Other side, broadcasting " escaping " voice messaging, play " attentioning " voice messaging, play " low roars of laughter " voice messaging, action stop,
Electric shock, the voice messaging for playing " going to play with other side ", connection owner's client, request and execute owner's remote command.
Step S501: detection the first pet wearable device at a distance from the second pet wearable device whether
More than the pre-determined distance, if it does, the first pet wearable device and the second pet wearable device terminate
Video recording, recording.
Step S601: pet order machine learning model training data, pet order engineering described in re -training are obtained
Practise model:
The acquisition process of the pet order machine learning model training data is as follows:
Obtain a plurality of training data, each training data includes at least command determination factor data sample set, and with each institute
State the corresponding simulation command sample set of command determination factor data sample set.
When obtaining a plurality of training data, command determination factor data sample set is obtained first, specifically, command determination factor
The mode of data acquisition includes by the first pet of first identity information, second identity information and each time point
Emotional state, behavior intention and physiological data, the emotional state of the second pet, behavior intention and physiological data be as the life
Enable factor of judgment set of data samples.Further, simulation command corresponding with each command determination factor data sample is obtained,
To obtain simulation command sample set corresponding with the command determination factor data sample set；Specifically, in the first owner visitor
Family end plays Video data, the recording data according to the first pet wearable device obtained by step S201, shows the life
Enable factor of judgment data.Meanwhile the simulation command operation of the first owner client is received, therefore command determination can obtained
Corresponding simulation command is obtained while factor data sample data, and then is obtained corresponding with command determination factor data sample set
Simulation command sample set, wherein simulation command include vibration, obtain other side pet fear sound threatening other side, broadcasting " escape
The voice messaging of race ", the voice messaging for playing " attentioning ", the voice messaging for playing " low roars of laughter ", action are stopped, are shocked by electricity, playing and " go
With other side play " voice messaging, connection owner's client, execute owner's remote command.
The first owner client is the client of the owner of the first pet wearable device, is mainly used for receiving
And the data that the first pet wearable device is sent are presented, it sends remote command and is set to first pet is wearable
The simulation command operation of owner that is standby, receiving the first pet wearable device, obtains machine learning model training data.
The first owner client is also used to: the blacklist of editor's the first pet wearable device storage, specifically
It include: that pet ID is added into the blacklist of the first pet wearable device storage；It is set from first pet is wearable
Pet ID is deleted in the blacklist of standby storage.
The disclosure also provides a kind of pet wearable device, comprising: wireless communication device is mainly used for setting with other electronics
It is standby to establish communication connection, receive or transmit data；Processor is mainly used for executing image recognition operation, sound Emotion identification fortune
The logical operation for calculating, vocal behavior intention assessment operation, issuing an order in conjunction with default rule to pet；Recording device, it is built-in
Phonographic recorder is mainly used for recording audio information；Recording apparatus, built-in camera are mainly used for recorded video information；Audio plays
Device, boombox are mainly used for playing audio-frequency information；Shaking device, built-in vibrator are mainly used for executing the vibration
Order, the vibration frequency of the shaking device can be according to parameter regulation；Electric shock device, built-in electric shock device are mainly used for holding
The row electric shock order；Action arrester, built-in knee joint blown jacket are mainly used for limiting the ability to act of pet dog；
Parameter acquisition devices, in-built motion sensor and pressure sensor, mainly for detection of the heart rate of pet, blood pressure, body temperature, breathing
The physiological datas such as frequency.
Claims (10)
1. a kind of pet that prevents based on machine learning is had a fist fight method characterized by comprising
First pet wearable device, which detects, whether there is the second pet wearable device in pre-determined distance；
The first pet wearable device and the second pet wearable device establish communication connection；
The first pet wearable device and the second pet wearable device open video recording, recording；
Every preset time, emotional state, behavior intention and the physiological data of first pet, second pet are obtained
Emotional state, behavior intention and physiological data；
Pre-set commands are executed in conjunction with default rule；
The first pet wearable device and the second pet wearable device terminate video recording, recording.
2. a kind of pet that prevents based on machine learning according to claim 1 is had a fist fight method, it is characterised in that: described
One pet wearable device, which detects, whether there is the second pet wearable device in pre-determined distance, comprising:
In pre-determined distance, returned if the first pet wearable device can receive the second pet wearable device
The identity information of the preset format returned then judges in pre-determined distance there are the second pet wearable device, if described the
One pet wearable device can not receive the identity information for the preset format that the second pet wearable device returns, then sentence
The second pet wearable device is not present in disconnected pre-determined distance；
The identity information of the preset format, includes at least: can identify the pet ID of pet identity.
3. a kind of pet that prevents based on machine learning according to claim 1 is had a fist fight method, it is characterised in that: described
One pet wearable device and the second pet wearable device establish communication connection, comprising:
The first pet wearable device sends the first identity information to the second pet wearable device, and described second dotes on
Object wearable device sends the second identity information to the first pet wearable device；
First identity information is the data of the first pet wearable device storage, and specifically include: pet ID, bacterium contain
It measures, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, the sleeps of proximal segment time
Situation baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency；
Second identity information is the data of the second pet wearable device storage, and specifically include: pet ID, bacterium contain
It measures, whether carry virus, weight, gender, whether there is or not record of having a fist fight, age, personality, proximal segment time amount of exercise, the sleeps of proximal segment time
Situation baits power, intellectual level, pet photo, the sound of pet fear, pet and feels the comfortable vibration frequency；
The first pet wearable device, according to the second pet image, in conjunction with image recognition technology, detection described second
Whether the pet for including in pet image and the pet for including in the pet photo in second identity information are same and dote on
Object, if it is not, by second identity information pet ID, bacterial content, whether carry virus, weight, gender, whether there is or not
Have a fist fight record, age, personality, proximal segment time amount of exercise, proximal segment time sleep state, bait power, intellectual level, pet photo,
The sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown；
The second pet wearable device, according to the first pet image, in conjunction with image recognition technology, detection described first
Whether the pet for including in pet image and the pet for including in the pet photo in first identity information are same and dote on
Object, if it is not, by first identity information pet ID, bacterial content, whether carry virus, weight, gender, whether there is or not
Have a fist fight record, age, personality, proximal segment time amount of exercise, proximal segment time sleep state, bait power, intellectual level, pet photo,
The sound of pet fear, pet feel that the comfortable vibration frequency is set as unknown.
4. a kind of pet that prevents based on machine learning according to claim 1 is had a fist fight method, it is characterised in that: described every
Every preset time, emotional state, behavior intention and the physiological data of first pet, the mood shape of second pet are obtained
State, behavior intention and physiological data, comprising:
By the first pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the sound letter of the first pet are obtained
Breath；
The mood shape of first pet is obtained in conjunction with sound Emotion identification technology according to the acoustic information of first pet
State；
The row of first pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of first pet
To be intended to；
By the second pet wearable device, heart rate, blood pressure, body temperature, respiratory rate, the sound letter of the second pet are obtained
Breath；
The mood shape of second pet is obtained in conjunction with sound Emotion identification technology according to the acoustic information of second pet
State；
The row of second pet is obtained in conjunction with vocal behavior intention assessment technology according to the acoustic information of second pet
To be intended to；
The emotional state, comprising: indignation, it is happy, it is tranquil, it is frightened；
The behavior is intended to, comprising: warning is attacked, entreat.
5. a kind of pet that prevents based on machine learning according to claim 1 is had a fist fight method, it is characterised in that: the knot
It closes default rule and executes pre-set commands, comprising:
By first identity information, second identity information, the emotional state of first pet, behavior intention and physiology
Data, the emotional state of second pet, behavior intention and physiological data are as command determination factor data, in conjunction with preset
Rule executes pre-set commands, specifically includes:
A) judge that the pet ID in second identity information whether there is in the blacklist of the first pet wearable device
In, if it does, executing the voice messaging for playing " escaping ", it is performed simultaneously connection owner's client, requests and holds
Row owner's remote command；
B) when the emotional state of first pet is angry, in first identity information and second identity information property
It is not identical, there is record in second identity information, when second pet has the behavior of attack to be intended to, described in execution
Play the voice messaging of " escaping "；
C) when the emotional state of first pet is happiness, the emotional state of second pet is happiness, first body
Part information and the gender difference in second identity information execute the voice messaging for playing " going to play with other side "；
D) when the emotional state of first pet is indignation, the emotional state of second pet is indignation, and described first dotes on
Object has the behavior of attack to be intended to, and when second pet has the behavior of attack to be intended to, executes the action and stops order, limit institute
State the ability to act of the first pet.Pet ID in second identity information is piped off, described first is stored into and dotes on
In object wearable device；
E) when the behavior of first pet is intended to attack, and the behavior of second pet is intended to entreat, described in execution
Electric shock order；
F) when the heart rate of first pet is more than second threshold more than first threshold or the respiratory rate of first pet,
It allows first pet to feel the comfortable vibration frequency by first identity information acquisition, is felt according to first pet
The comfortable vibration frequency executes the vibration and orders, until the heart rate of first pet is restored to lower than first threshold or described
Then stop the vibration order when respiratory rate of first pet is restored to lower than second threshold；
G) judge whether the pet in second identity information carries Virus Info, if carrying virus, execute the broadcasting
The voice messaging of " escaping " is performed simultaneously connection owner's client, requests and execute owner's remote command；
H) when the behavior of first pet is intended to alert, the sound threatening other side for executing the acquisition other side pet fear is ordered
It enables.
6. a kind of pet that prevents based on machine learning according to claim 5 is had a fist fight method, it is characterised in that: the knot
It closes default rule and executes pre-set commands, further includes:
Obtain command determination factor data sample set and corresponding simulation command sample set；Based on command determination factor data sample
Collection and the corresponding simulation command sample set training pet order machine learning model；The command determination factor data is defeated
Enter to pet order machine learning model, determines output order corresponding with the command determination factor data, described first dotes on
Object wearable device executes the pet order machine learning model output order.
7. a kind of pet that prevents based on machine learning according to claim 6 is had a fist fight method, it is characterised in that: described to obtain
Take command determination factor data sample set and corresponding simulation command sample set, comprising:
Video data, the recording data of first pet wearable device described in the first owner client terminal playing, described in displaying
Emotional state, behavior intention and the physiology of first pet of the first identity information, second identity information and each time point
Data, the emotional state of the second pet, behavior intention and physiological data receive the owner's of the first pet wearable device
Simulation command operation, obtains the command determination factor data sample set and the corresponding simulation command sample set.
8. a kind of pet that prevents based on machine learning according to claim 1 is had a fist fight method, it is characterised in that: described
One pet wearable device and the second pet wearable device terminate video recording, recording, comprising:
Whether detect the first pet wearable device at a distance from the second pet wearable device is more than described default
Distance, if it does, the first pet wearable device and the second pet wearable device terminate video recording, recording.
9. a kind of wearable device for pet characterized by comprising
Wireless communication device is wirelessly connected for establishing with other electronic equipments；
Parameter acquisition devices, in-built motion sensor and pressure sensor, mainly for detection of the wearable device is worn
The heart rate of pet, blood pressure, body temperature, respiratory rate.
Recording apparatus, for recording the image data in front of the pet for wearing the wearable device；
Recording device, for recording the audio data around the pet for wearing the wearable device；
Processor is mainly used for executing image recognition operation, the operation of sound Emotion identification, the operation of vocal behavior intention assessment, knot
Close the logical operation that default rule is issued an order to pet；
Audio playing apparatus, boombox are mainly used for playing audio-frequency information；
Shaking device, built-in vibrator are mainly used for executing the vibration order, and the vibration frequency of the shaking device is can root
According to parameter regulation；
Electric shock device, built-in electric shock device are mainly used for executing the electric shock order；
Action arrester, built-in knee joint blown jacket are mainly used for limiting the ability to act of pet dog.
10. a kind of for assigning the client of simulation command characterized by comprising
Data show device, for playing Video data, the recording data of the first pet wearable device, show described the
Emotional state, behavior intention and the physiology number of first pet of one identity information, second identity information and each time point
According to, the emotional state of the second pet, behavior intention and physiological data；
Simulation command device, the simulation command operation of the owner for receiving the first pet wearable device, obtains machine
Learning model training data；
Remote command device, for sending remote command to the first pet wearable device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910820242.8A CN110506661B (en) | 2019-09-01 | 2019-09-01 | Method for preventing pet fighting based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910820242.8A CN110506661B (en) | 2019-09-01 | 2019-09-01 | Method for preventing pet fighting based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110506661A true CN110506661A (en) | 2019-11-29 |
| CN110506661B CN110506661B (en) | 2021-12-21 |
Family
ID=68630049
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910820242.8A Expired - Fee Related CN110506661B (en) | 2019-09-01 | 2019-09-01 | Method for preventing pet fighting based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110506661B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112450116A (en) * | 2020-12-14 | 2021-03-09 | 深圳科帮电子有限公司 | Pet management method, device, system, equipment and storage medium |
| CN116391630A (en) * | 2023-04-24 | 2023-07-07 | 重庆长安汽车股份有限公司 | In-vehicle pet management method, system, electronic equipment and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004042670A2 (en) * | 2002-11-08 | 2004-05-21 | Dbs Security Ltd. | Canine security system |
| CN105123551A (en) * | 2015-08-28 | 2015-12-09 | 广州中胜物联网络科技有限公司 | Intelligent pet wearable device and control method thereof |
| CN106534477A (en) * | 2016-08-30 | 2017-03-22 | 深圳市沃特沃德股份有限公司 | Method, device and system for managing living habits of pet |
| CN107635400A (en) * | 2016-05-09 | 2018-01-26 | 深圳市欸阿技术有限公司 | Pet wearable device and its method for supervising pet |
| CN107926747A (en) * | 2018-01-02 | 2018-04-20 | 合肥淘云科技有限公司 | A kind of Combined pet instructs and guides system |
| CN109064012A (en) * | 2018-07-30 | 2018-12-21 | 合肥东恒锐电子科技有限公司 | A kind of pet continues tracing management monitoring method and system |
| WO2019010494A1 (en) * | 2017-07-07 | 2019-01-10 | James Chad R | Animal training device that controls stimulus using proportional pressure-based input |
| CN109360115A (en) * | 2018-11-07 | 2019-02-19 | 中山乐心电子有限公司 | Pet friend-making control method, device and wearable device |
| CN109757395A (en) * | 2018-09-24 | 2019-05-17 | 天津大学 | A kind of pet behavioral value monitoring system and method |
| KR20190081599A (en) * | 2017-12-29 | 2019-07-09 | (주)씽크웨이브 | Companion dog management apparatus |
| CN110063267A (en) * | 2019-06-11 | 2019-07-30 | 深圳派特科技有限公司 | A kind of system and method for automatic detection and correction animal behavior |
| CN110122364A (en) * | 2019-05-13 | 2019-08-16 | 安徽三品技术服务有限公司 | Multifunctional pet clothes based on smart machine interconnection |
-
2019
- 2019-09-01 CN CN201910820242.8A patent/CN110506661B/en not_active Expired - Fee Related
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004042670A2 (en) * | 2002-11-08 | 2004-05-21 | Dbs Security Ltd. | Canine security system |
| CN105123551A (en) * | 2015-08-28 | 2015-12-09 | 广州中胜物联网络科技有限公司 | Intelligent pet wearable device and control method thereof |
| CN107635400A (en) * | 2016-05-09 | 2018-01-26 | 深圳市欸阿技术有限公司 | Pet wearable device and its method for supervising pet |
| CN106534477A (en) * | 2016-08-30 | 2017-03-22 | 深圳市沃特沃德股份有限公司 | Method, device and system for managing living habits of pet |
| WO2019010494A1 (en) * | 2017-07-07 | 2019-01-10 | James Chad R | Animal training device that controls stimulus using proportional pressure-based input |
| KR20190081599A (en) * | 2017-12-29 | 2019-07-09 | (주)씽크웨이브 | Companion dog management apparatus |
| CN107926747A (en) * | 2018-01-02 | 2018-04-20 | 合肥淘云科技有限公司 | A kind of Combined pet instructs and guides system |
| CN109064012A (en) * | 2018-07-30 | 2018-12-21 | 合肥东恒锐电子科技有限公司 | A kind of pet continues tracing management monitoring method and system |
| CN109757395A (en) * | 2018-09-24 | 2019-05-17 | 天津大学 | A kind of pet behavioral value monitoring system and method |
| CN109360115A (en) * | 2018-11-07 | 2019-02-19 | 中山乐心电子有限公司 | Pet friend-making control method, device and wearable device |
| CN110122364A (en) * | 2019-05-13 | 2019-08-16 | 安徽三品技术服务有限公司 | Multifunctional pet clothes based on smart machine interconnection |
| CN110063267A (en) * | 2019-06-11 | 2019-07-30 | 深圳派特科技有限公司 | A kind of system and method for automatic detection and correction animal behavior |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112450116A (en) * | 2020-12-14 | 2021-03-09 | 深圳科帮电子有限公司 | Pet management method, device, system, equipment and storage medium |
| CN116391630A (en) * | 2023-04-24 | 2023-07-07 | 重庆长安汽车股份有限公司 | In-vehicle pet management method, system, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110506661B (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10136619B2 (en) | Intelligent pet monitoring system | |
| CN101198277B (en) | Systems for physiological and psycho-physiological monitoring | |
| US8305220B2 (en) | Monitoring and displaying activities | |
| CN111975772B (en) | Robot control method, device, electronic device and storage medium | |
| CN110506661A (en) | A kind of pet that prevents based on machine learning is had a fist fight method | |
| JPWO2016021236A1 (en) | Information processing system, information processing apparatus, information processing program, and information processing method | |
| JPWO2016171077A1 (en) | Information processing system | |
| US20200045929A1 (en) | Hog health monitor | |
| US11635816B2 (en) | Information processing apparatus and non-transitory computer readable medium | |
| CN112188296A (en) | Interaction method, device, terminal and television | |
| JP2018092227A (en) | Presentation control device, presentation control method and program | |
| JP2023129585A (en) | Information processing method, information processing apparatus, and information processing program | |
| WO2021011784A1 (en) | Remote physiological data sensing robot | |
| US20050197546A1 (en) | Security system | |
| JP6914597B2 (en) | Pet / person friendship measuring device and pet / person friendship measuring program | |
| CN106650288A (en) | Temperature control charging remote medical system parameter configuration system | |
| JP2021089635A (en) | Information processing device and program | |
| KR102473443B1 (en) | Intelligent companion animal management system and method based on psychological and behavioral analysis of indoor companion animals through iot | |
| JP7429416B2 (en) | Information processing device and program | |
| CN113031758A (en) | Information processing apparatus and computer readable medium | |
| CN114366029A (en) | Wearable pet equipment and pet monitoring method | |
| WO2023187865A1 (en) | Information processing device, information processing method, and recording medium | |
| CN110427155B (en) | Nursing method and device | |
| EP4293646A1 (en) | A baby monitoring system and method | |
| US20230095345A1 (en) | Non-contact systems and methods for monitoring and addressing breathing events in neonates |
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: 20211112 Address after: 453000 No. 51, South District, Taoyuan Village, Doumen Township, Yuanyang County, Xinxiang City, Henan Province Applicant after: Wu Shukuan Address before: 512026 room f101-12, incubation production building 1, guanshao innovation and innovation (equipment) center, Huake City, No. 42, Baiwang Avenue, Wujiang District, Shaoguan City, Guangdong Province Applicant before: Shaoguan Qizhi Information Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20211221 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |