CN108508974A - Wearable device processor and its data processing method and wearable device - Google Patents
Wearable device processor and its data processing method and wearable device Download PDFInfo
- Publication number
- CN108508974A CN108508974A CN201810294037.8A CN201810294037A CN108508974A CN 108508974 A CN108508974 A CN 108508974A CN 201810294037 A CN201810294037 A CN 201810294037A CN 108508974 A CN108508974 A CN 108508974A
- Authority
- CN
- China
- Prior art keywords
- data
- wearable device
- data processing
- user
- control unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/16—Constructional details or arrangements
- G06F1/1613—Constructional details or arrangements for portable computers
- G06F1/163—Wearable computers, e.g. on a belt
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/163—Interprocessor communication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a kind of wearable device processors, including：Data processing micro control unit is connected with each sensor of wearable device, and the motion-sensing data for acquiring each sensor carry out corresponding data operation processing by data processing algorithm, obtains operation result, identifies the behavior act of user；Multimedia micro control unit is connected with data processing micro control unit, the multi-media module for controlling wearable device；It is additionally operable to be learnt according to the motion-sensing data that data processing micro control unit transmits, and updates the data the data processing algorithm of processing micro control unit according to learning outcome.In addition, the invention also discloses the data processing method of the wearable device processor and wearable devices.Wherein, wearable device includes the wearable device processor of the present invention.The wearable device processor of the present invention has artificial intelligence function, recognition result can be kept more real-time, be more in line with the needs of displaying with the demand of some artificial intelligence of processing locality.
Description
Technical field
The present invention relates to intelligent wearable device field more particularly to wearable device processor and its data processing sides
Method, wearable device.
Background technology
The processor of current wearable device, as just data collector, can not deep learning people behavioural habits,
And carry out personalized adapt to according to custom.Therefore, the function of many action recognitions and deduction can not all effectively improve identification
Rate, efficiency is low, and such system architecture can seriously affect usage experience.
And current AI is mainly used in cloud server end, data carry out acquired original in equipment end, then are transferred to high in the clouds
Learnt；Transmission range is long, and delay is big, can not scene where effectively consersion unit is practical feature, real-time and use
The behavioural habits at family.
Invention content
In order to solve drawbacks described above, the present invention provides a kind of wearable device processor and its data processing method and can
Wearable device.Specifically, technical scheme is as follows：
On the one hand, the invention discloses a kind of wearable device processors, including：Data processing micro control unit, and can wear
Each sensor for wearing equipment is connected, for carrying out the motion-sensing data of each sensor acquisition by data processing algorithm
Corresponding data operation processing, obtains operation result, identifies the behavior act of user；Multimedia micro control unit, with the data
It handles micro control unit to be connected, the multi-media module for controlling the wearable device；It is additionally operable to micro- according to the data processing
The motion-sensing data of control unit transmission are learnt, and the data of the data processing micro control unit are updated according to learning outcome
Processing Algorithm.
Preferably, the data processing micro control unit includes：Sensing data receiving module, for receiving wearable device
The motion-sensing data of each sensor acquisition；Data processing module is used for according to the mathematical algorithm of itself, by the motion-sensing
Data carry out fusion operation processing, obtain operation result data；Action recognition module, for being obtained according to the data processing module
The operation result data taken, identify the behavior act of user；Data transmission module is used for the operation result data and identification
The behavior act data of corresponding user be transferred to multimedia micro control unit.
Preferably, the multimedia micro control unit includes：Receiving module, requirement command for receiving user and described
The behavior act data of the operation result data and user of the transmission of data processing micro control unit；Multimedia main control module is used for root
According to the requirement command of the user or the behavior act data of the user, controls each multimedia and exported accordingly；Intelligence
Module, for passing through neural network, in the behavior act data of the operation result data and user that are received from the receiving module
The behavioural characteristic for learning user updates the calculation that in the data processing micro control unit motion-sensing data are carried out with calculation process
Method.
Preferably, the multimedia main control module is integrated with RAM, ROM, level cache, GPU, PMU and multimedia periphery
Interface.
Preferably, the data processing micro control unit and/or micro- place that multimedia micro control unit is the CM4 with floating-point operation
Manage device.
Second aspect, the invention also discloses a kind of data processing methods based on wearable device processor, including：
S200 obtains the motion-sensing data of each sensor acquisition of wearable device；
S300 carries out data processing using the data processing algorithm of the user to the motion-sensing data, obtains operation
Result data；
S400 identifies the behavior act of user according to the operation result data；
S500 controls the wearable device and is responded accordingly according to the behavior act of the user of identification.
Preferably, further include before the step S100：S100 learns the behavioural characteristic of user, forms the user's
Data processing algorithm.
Preferably, the step S100 includes：
Motion-sensing data of the S110 by each sensor acquisition of wearable device for study；
S120 carries out data processing using initial data processing algorithm to the motion-sensing data, obtains operation result
Data；
S130 identifies the behavior act of user according to the operation result data；
S140 is moved the motion-sensing data, the operation result data and corresponding user behavior by neural network
Make data and carries out self study as sample；
After the sample learning that S150 passes through specified quantity, the initial data processing algorithm is updated, the user is formed
Data processing algorithm.
Preferably, after the step S130, before the step S140, further include：S135 judges the operation knot
Whether the accuracy of fruit data reaches preset percentage, if so, entering step S140.
The third aspect, the invention also discloses a kind of wearable devices, including any one of them of the present invention is wearable sets
Standby processor.
The present invention at least has with the next item down technique effect：
(1) the multimedia micro control unit in wearable device processor of the invention also has other than to multimedia control
Standby learning functionality, can learn the behavioural characteristic of user, to renolation algorithm so that according to being differently formed for user's individual
Itself unique algorithm reaches the accurate identification of action etc..
(2) two micro control units (MCU) are integrated on the processor of a wearable device by the present invention, reduce occupancy
Area is suitable for wearable device.
(3) the multimedia micro control unit in wearable device processor of the invention and data processing micro control unit can be adopted
With the CM4 microprocessors with floating-point operation, the DSP operation ability in CM4 microprocessors is strong, to by making full use of DSP numbers
According to processing capacity, power consumption is reduced, achievees the purpose that power saving.
(4) present invention realizes acquisition and study processing of exercise data etc. and is run in local operation, relative in cloud
Server end carries out data processing and study, this motion timeliness are stronger.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is the block diagram of the embodiment of wearable device processor of the present invention；
Fig. 2 is the block diagram of another embodiment of wearable device processor of the present invention；
Fig. 3 is the internal structure schematic diagram of another embodiment of wearable device processor of the present invention；
Fig. 4 is the flow chart of the embodiment of the data processing method of wearable device processor of the present invention；
Fig. 5 is the flow chart of another embodiment of the data processing method of wearable device processor of the present invention；
Fig. 6 is the flow chart of another embodiment of the data processing method of wearable device processor of the present invention.
Specific implementation mode
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into
It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation
Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts
All other embodiment, shall fall within the protection scope of the present invention.
The invention discloses a kind of wearable device processors, and embodiment is as shown in Figure 1, include：Data processing micro-control list
Member 100, is connected with each sensor of wearable device, for the motion-sensing data of each sensor acquisition to be passed through data
Processing Algorithm carries out corresponding data operation processing, obtains operation result, identifies the behavior act of user；Multimedia micro control unit
200, it is connected with the data processing micro control unit 100, the multi-media module for controlling the wearable device；It is additionally operable to root
The motion-sensing data transmitted according to the data processing micro control unit 100 are learnt, and update the number according to learning outcome
According to the data processing algorithm of processing micro control unit 100.
Wearable device processor in the present embodiment is integrated with two micro control units (i.e. MCU), reduces occupied space,
It is relatively specific for wearable device.Wherein, the movement that data processing micro control unit acquires each sensor of wearable device passes
Feel data and carry out fusion operation, the algorithm that operation uses is currently existing data processing algorithm (if note that the algorithm passes through
Multimedia micro control unit is updated, then existing data processing algorithm is then newer data processing algorithm, that is to say, that
Only newest data processing algorithm can be stored in data processing micro control unit always).Multimedia micro control unit is then used for controlling
Every multi-media module of wearable device, the multimedia micro control unit are equipped with several peripheral interfaces, being capable of plug-in memory, wheat
Gram wind, screen etc..For example the multimedia micro control unit can control the screen output of wearable device, obtain language by microphone
Message breath etc..In addition, mostly important, the multimedia micro control unit in the present embodiment can also be used to learn motion-sensing number
According to learning the motion feature etc. of user, the exclusive data processing algorithm of the user gradually formed after constantly learning, to right
Data processing algorithm in former data processing micro control unit is updated.
The wearable device processor of the present embodiment can be according to the difference of user's individual after study after a period of time
(there may be differences for different user behavioural characteristic) and gradually form the data processing algorithm for belonging to the user, certainly, formed should
Be not after the data processing algorithm of user yet it is unalterable, the wearable device processor can by constantly learning,
The data processing algorithm also can be continuous perfect, slowly, to the behavior of the user judge also can be more it is accurate.In addition, this
The wearable device processor of embodiment by motion-sensing data that wearable device acquires locally carried out data processing and
Study etc. needs, beyond the clouds come for carrying out data processing, timeliness is stronger, to know the behavior act of user relative to existing
Not faster, to just the wearable device be allowed to make corresponding response according to the behavioural characteristic of user faster.
Another embodiment of wearable device processor of the present invention, as shown in Fig. 2, on the basis of the above embodiments, institute
Stating data processing micro control unit 100 includes：Sensing data receiving module 110, each sensor for receiving wearable device are adopted
The motion-sensing data of collection；Data processing module 120, for according to the mathematical algorithm of itself, by the motion-sensing data into
The processing of row fusion operation, obtains operation result data；Action recognition module 130, for being obtained according to the data processing module 120
The operation result data taken, identify the behavior act of user；Data transmission module 140, for by the operation result data and
The behavior act data of the corresponding user of identification are transferred to multimedia micro control unit 200.
Preferably, in the embodiment of any of the above-described wearable device processor, the multimedia micro control unit 200 includes：
Receiving module 210, the operation result number of requirement command and the data processing micro control unit 100 transmission for receiving user
According to the behavior act data with user；Multimedia main control module 220 is used for the requirement command according to the user or the user
Behavior act data, control each multimedia and exported accordingly；Intelligent object 230, for by neural network, from described
The behavioural characteristic at family is commonly used in the behavior act data middle school of operation result data and user that receiving module 210 receives, updates institute
State the algorithm that in data processing micro control unit 100 motion-sensing data are carried out with calculation process.
Specifically, above-mentioned multimedia main control module may include：Graphics process submodule, power management submodule, audio
Decoder module, peripheral interface module, master control submodule etc..For example, graphics process submodule is for carrying out image operation processing
Work controls the screen picture output of the wearable device；Audio decoder module, for what is received to the wearable device
Audio speech signal is decoded, and obtains audio digital signals；Master control submodule, for controlling in the multimedia control module
Each submodule co-ordination etc..
Preferably, the multimedia main control module in above-described embodiment be integrated with RAM, ROM, level cache, GPU, PMU and
Multimedia peripheral interface etc..
In addition, the data processing micro control unit in any of the above-described wearable device processor embodiment, multimedia are micro-
Control the microprocessor that unit can be the CM4 with floating-point operation.DSP operation ability in CM4 microprocessors is strong, to by filling
Divide and utilize DSP data-handling capacities, reduces power consumption, achieve the purpose that power saving.
Another embodiment of wearable device processor of the present invention, wearable device processor includes two MCU, wherein one
MCU is linked with each sensor, the exercise data progress calculation process for acquire to each sensor；Another MCU is as more
Media master control, and neural network module is equipped in the MCU, the progress such as motion-sensing data of user for being acquired to sensor
Study so that the wearable device, gradually can be according to optimization algorithm the characteristics of user itself, to make after constantly learning
The data that must be exported are more accurate, for example can accurately identify the action behavior etc. of the user.Specifically, as shown in figure 3,
Have two MCU in this wearable device processor, one, as multimedia master control (the ARM CM4 on the left side), is internally integrated
RAM, ROM, L1Cache (level cache), GPU (graphics processor), PMU (Power Management Unit) and a variety of peripheral interfaces,
Can plug-in WIFI, BT/BLE, extend out memory, analog/digital microphone, high-resolution screen；Another is used as sensor collection
Knot device is used (the ARM CM4 on the right), it can constantly electricity works for a long time, and power consumption is extremely low.I2C chains can be passed through
Various sensors are connect, GPS is linked by Uart, using DSP core+neural network/machine learning algorithm inside CM4, are come real
When learn the physical trait and behavioural characteristic of user, identify motor pattern and scene perception, and constantly train, optimize oneself
Algorithm.
Two MCU in above-mentioned wearable device processor embodiment can be micro- place of the CM4 with floating-point operation
Device is managed, 24MHz and 48MHz is run.For example, as the MCU (the right) of sensor collector using with floating-point operation in Fig. 3
CM4 microprocessors, it may be possible to which the data of each sensor are melted in the operation for meeting the multiple Data Fusion of Sensor in periphery
It closes, using the DSP and software algorithm inside CM4, carries out artificial intelligence operation, can be inferred that motor pattern or real-time field
Scape.Multi-frequency can be run using the microprocessor of the CM4 with floating-point operation as the MCU (left side) of multimedia master control for another example,
The screen etc. of highest 320*320 screens can be driven.
Based on the same technical idea, the invention also discloses a kind of data processing sides based on wearable device processor
Method, the data processing method that the present invention can be used in wearable device processor of the invention carry out data processing, specifically, as schemed
Shown in 4, including：
S200 obtains the motion-sensing data of each sensor acquisition of wearable device；
S300 carries out data processing using the data processing algorithm of the user to the motion-sensing data, obtains operation
Result data；
S400 identifies the behavior act of user according to the operation result data；
S500 controls the wearable device and is responded accordingly according to the behavior act of the user of identification.
The present embodiment carries out data processing based on the processor in any of the above-described wearable device processor embodiment, specifically
, by the data processing micro control unit in the wearable device processor to the fortune of each sensor acquisition of the wearable device
Dynamic data carry out calculation process and identify user according further to the operation result data to obtain operation result data
Behavior act, finally, the multimedia main control unit of the wearable device processor can be according to the behavior of the user identified
Action control wearable device is responded accordingly, such as certain wearable device has the function of the lift bright screen of hand, then using
During lifting hand, each sensor of the wearable device will acquire motion-sensing data when user lifts hand, then at family
Data processing micro control unit carries out fusion operation according to the data processing algorithm of the user to these motion-sensing data, obtains one
Then a operation result judges that the lift of user is made manually according to this operation result, finally, multimedia main control module can root
Make manually according to the lift of the user identified, control the bright screen of screen of wearable device, to realize the function of the lift bright screen of hand.Its
In, the data processing algorithm for the user that data processing micro control unit uses refers to being formed for the user of the wearable device
Algorithm, that is to say, that different users, since its behavioural characteristic has differences, then the algorithm of the data processing eventually formed is
Different.Therefore, the behavior act of the user also more can be accurately identified using the exclusive data processing algorithm of the user.
Due to acquiring the motion-sensing data of user by each sensor, to the calculation process of motion-sensing data and identification, even lead to
Crossing intelligence learning and forming the data processing algorithm of user is can be completed in local operation, apparent to be obtained in timeliness
It is promoted, sensing data is only acquired compared to existing wearable device, and processing of sensing data etc. is completed beyond the clouds, this implementation
The response that the wearable device processor of example is provided for the action of user can more rapid, user experience higher.
Another data processing method embodiment of the present invention, on the basis of above method embodiment, in above-mentioned steps S100
Further include before：S100 learns the behavioural characteristic of user, forms the data processing algorithm of the user.Specifically, this method is real
It applies for example shown in Fig. 5, specifically includes：
Motion-sensing data of the S110 by each sensor acquisition of wearable device for study；
S120 carries out data processing using initial data processing algorithm to the motion-sensing data, obtains operation result
Data；
S130 identifies the behavior act of user according to the operation result data；
S140 is moved the motion-sensing data, the operation result data and corresponding user behavior by neural network
Make data and carries out self study as sample；
After the sample learning that S150 passes through specified quantity, the initial data processing algorithm is updated, the user is formed
Data processing algorithm；
S200 obtains the motion-sensing data of each sensor acquisition of wearable device；
S300 carries out data processing using the data processing algorithm of the user to the motion-sensing data, obtains operation
Result data；
S400 identifies the behavior act of user according to the operation result data；
S500 controls the wearable device and is responded accordingly according to the behavior act of the user of identification.
The present embodiment on the basis of the above embodiments, increases specific intelligence learning step, and wearable device is most opened
It is at the motion-sensing data acquired to each sensor according to built-in primary data Processing Algorithm when beginning to use
Reason, still, with continuing on for user, then the artificial intelligence module in the wearable device processor then can constantly be gone
Study is taken from the motion-sensing data of acquisition, slowly the behavioural characteristic of study to user, to which constantly improve optimizes initial number
According to Processing Algorithm, the exclusive data processing algorithm of user is slowly formed.The data processing algorithm of the user then can be more accurate
The motion-sensing data to the user handle, identify the action of user.In addition, the data processing algorithm of the user
It is not unalterable, with continuous study, the data processing algorithm of the user also can constantly update and perfect.
Preferably, on the basis of above method embodiment, the judgement selection step of learning data is increased, specifically,
Another embodiment of present aspect method, as shown in fig. 6, including：
Motion-sensing data of the S110 by each sensor acquisition of wearable device for study；
S120 carries out data processing using initial data processing algorithm to the motion-sensing data, obtains operation result
Data；
S130 identifies the behavior act of user according to the operation result data；
S135 judges whether the accuracy of the operation result data reaches preset percentage, if so, entering step
S140；
S140 is moved the motion-sensing data, the operation result data and corresponding user behavior by neural network
Make data and carries out self study as sample；
After the sample learning that S150 passes through specified quantity, the initial data processing algorithm is updated, the user is formed
Data processing algorithm；
S200 obtains the motion-sensing data of each sensor acquisition of wearable device；
S300 carries out data processing using the data processing algorithm of the user to the motion-sensing data, obtains operation
Result data；
S400 identifies the behavior act of user according to the operation result data；
S500 controls the wearable device and is responded accordingly according to the behavior act of the user of identification.
This method embodiment increases the selection of learning sample on the basis of above method embodiment, avoids study to not
Accurate sample is not enough bonded the behavioural characteristic of the user so as to cause the algorithm of formation.Specifically, in the fortune acquired to user
After dynamic sensing data is handled using currently existing data processing algorithm, can to the accuracy of treated operational data into
The judgement of row one can just go the movement that will be acquired in the case that the accuracy of only operation result has reached preset percentage
Sensing data, corresponding operation result and action recognition data go to learn as sample, then by the study of great amount of samples after
Existing data processing algorithm is slowly improved, the exclusive data processing algorithm of the user is formed.Such as the preset percentage set
It is 80%, then, that is to say, that judge that user lifts ability of the accuracy rate of the action of hand 80% or more after calculation process
Sample can be used as to carry out learning training.
Finally, the invention also discloses a kind of wearable device, which, which includes any one of the invention, to wear
Wear device handler.Since the wearable device processor in the present invention is constituted by two micro control units are integrated, occupancy is reduced
Space meets the size requirement of wearable device.In addition, by the multimedia micro control unit of wearable device processor to passing
The study of nyctinastic movement data learns the physical trait and behavioural characteristic of user, to renolation data processing micro control unit
In data processing algorithm so that updated algorithm more accurately identifies the behavior act of user.
Equally, we are for lifting the bright screen of hand, since the sensing data that each user lift is made manually is slightly different, then such as
If fruit carries out calculation process using same algorithm, it is possible that the situation that action recognition is not accurate, and use this hair
Bright wearable device then can be to avoid this problem.For example, the lift of standard is made manually, at initial data processing algorithm
It is obtained after managing the motion-sensing data that standard lift is made manually, the lift that operation result is 100% is made manually, due to being not each
The lift of user is all the lift with standard is made manually manually, and the difference of user's individual, corresponding characteristic can not yet
Equally.If obtaining certain user after using initial data processing algorithm to handle lifts the corresponding operation result of action of hand as 80%
Lift make manually, the wearable device of the present embodiment then can constantly learn the lift hand characteristic of the user by neural network
According to, and then optimize existing data processing algorithm, the data processing algorithm of the user is formed, to which the follow-up user lifts again
When making manually, then can accurately it be identified using the data processing algorithm of the user, specifically, using the data processing of the user
Algorithm, which calculates the operation result obtained after the lift hand sensing data of the user, will be more than 80%, even 100%.In this way, just
It can accurately judge that the lift of user is made manually according to the operation result, and then the wearable device can be according to the judging result
Quickly execute bright screen action.
The wearable device processor of the present invention, the data processing method and wearable device of wearable device processor
Technical concept it is almost the same, therefore, the technical detail of three is mutually applicable in, and repeats to repeat no more to reduce.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
God and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of wearable device processor, which is characterized in that including：
Data processing micro control unit is connected with each sensor of wearable device, the movement for acquiring each sensor
Sensing data carries out corresponding data operation processing by data processing algorithm, obtains operation result, identifies that the behavior of user is dynamic
Make；
Multimedia micro control unit is connected with the data processing micro control unit, the multimedia for controlling the wearable device
Module；It is additionally operable to be learnt according to the motion-sensing data that the data processing micro control unit transmits, and according to learning outcome
Update the data processing algorithm of the data processing micro control unit.
2. a kind of wearable device processor according to claim 1, which is characterized in that the data processing micro control unit
Including：
Sensing data receiving module, the motion-sensing data that each sensor for receiving wearable device acquires；
Data processing module, for according to the mathematical algorithm of itself, the motion-sensing data being carried out fusion operation processing, are obtained
Take operation result data；
Action recognition module, the operation result data for being obtained according to the data processing module identify that the behavior of user is dynamic
Make；
Data transmission module, for being transferred to the behavior act data of the operation result data and the corresponding user of identification
Multimedia micro control unit.
3. a kind of wearable device processor according to claim 1, which is characterized in that the multimedia micro control unit packet
It includes：
Receiving module, the operation result data of requirement command and data processing micro control unit transmission for receiving user
With the behavior act data of user；
Multimedia main control module, for according to the requirement command of the user or the behavior act data of the user, control to be each
Multimedia is exported accordingly；
Intelligent object, for by neural network, the behavior of the operation result data and user that are received from the receiving module to be dynamic
Make the behavioural characteristic that family is commonly used in data middle school, updates in the data processing micro control unit and motion-sensing data are carried out at operation
The algorithm of reason.
4. a kind of wearable device processor according to claim 3, which is characterized in that the multimedia main control module collection
At RAM, ROM, level cache, GPU, PMU and multimedia peripheral interface.
5. according to a kind of wearable device processor of claim 1-4 any one of them, which is characterized in that the data processing
Micro control unit and/or the microprocessor that multimedia micro control unit is the CM4 with floating-point operation.
6. a kind of data processing method based on wearable device processor, which is characterized in that including：
S200 obtains the motion-sensing data of each sensor acquisition of wearable device；
S300 carries out data processing using the data processing algorithm of the user to the motion-sensing data, obtains operation result
Data；
S400 identifies the behavior act of user according to the operation result data；
S500 controls the wearable device and is responded accordingly according to the behavior act of the user of identification.
7. a kind of data processing method based on wearable device processor according to claim 6, which is characterized in that
Further include before the step S100：
S100 learns the behavioural characteristic of user, forms the data processing algorithm of the user.
8. a kind of data processing method based on wearable device processor according to claim 7, which is characterized in that institute
Stating step S100 includes：
Motion-sensing data of the S110 by each sensor acquisition of wearable device for study；
S120 carries out data processing using initial data processing algorithm to the motion-sensing data, obtains operation result number
According to；
S130 identifies the behavior act of user according to the operation result data；
The motion-sensing data, the operation result data and corresponding user behavior are acted number by S140 by neural network
Self study is carried out according to as sample；
After the sample learning that S150 passes through specified quantity, the initial data processing algorithm is updated, the number of the user is formed
According to Processing Algorithm.
9. a kind of data processing method based on wearable device processor according to claim 8, which is characterized in that
After the step S130, before the step S140, further include：
S135 judges whether the accuracy of the operation result data reaches preset percentage, if so, entering step S140.
10. a kind of wearable device, which is characterized in that including claim 1-5 any one of them wearable device processors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810294037.8A CN108508974A (en) | 2018-03-30 | 2018-03-30 | Wearable device processor and its data processing method and wearable device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810294037.8A CN108508974A (en) | 2018-03-30 | 2018-03-30 | Wearable device processor and its data processing method and wearable device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108508974A true CN108508974A (en) | 2018-09-07 |
Family
ID=63380323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810294037.8A Pending CN108508974A (en) | 2018-03-30 | 2018-03-30 | Wearable device processor and its data processing method and wearable device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108508974A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784364A (en) * | 2018-12-06 | 2019-05-21 | 南京蓝思信息科技有限公司 | Head movement data processing method and device |
| CN109919034A (en) * | 2019-01-31 | 2019-06-21 | 厦门大学 | A kind of identification of limb action with correct auxiliary training system and method |
| CN110888535A (en) * | 2019-12-05 | 2020-03-17 | 上海工程技术大学 | AR system capable of improving on-site reality |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104698831A (en) * | 2015-03-03 | 2015-06-10 | 惠州Tcl移动通信有限公司 | Intelligent wrist watch capable of automatically rotating interface along with gesture and control method thereof |
| CN104793866A (en) * | 2015-05-05 | 2015-07-22 | 陈王胜 | Control unit and control method for backlight of intelligent wrist type wearable equipment |
| WO2016131256A1 (en) * | 2015-08-03 | 2016-08-25 | 中兴通讯股份有限公司 | Screen control method and apparatus for wearable device |
| CN107071605A (en) * | 2015-12-30 | 2017-08-18 | 杭州赛泫科技有限公司 | Intelligent 3D earphones |
| CN107530004A (en) * | 2015-02-20 | 2018-01-02 | Mc10股份有限公司 | The automatic detection and construction of wearable device based on personal situation, position and/or orientation |
-
2018
- 2018-03-30 CN CN201810294037.8A patent/CN108508974A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107530004A (en) * | 2015-02-20 | 2018-01-02 | Mc10股份有限公司 | The automatic detection and construction of wearable device based on personal situation, position and/or orientation |
| CN104698831A (en) * | 2015-03-03 | 2015-06-10 | 惠州Tcl移动通信有限公司 | Intelligent wrist watch capable of automatically rotating interface along with gesture and control method thereof |
| CN104793866A (en) * | 2015-05-05 | 2015-07-22 | 陈王胜 | Control unit and control method for backlight of intelligent wrist type wearable equipment |
| WO2016131256A1 (en) * | 2015-08-03 | 2016-08-25 | 中兴通讯股份有限公司 | Screen control method and apparatus for wearable device |
| CN107071605A (en) * | 2015-12-30 | 2017-08-18 | 杭州赛泫科技有限公司 | Intelligent 3D earphones |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784364A (en) * | 2018-12-06 | 2019-05-21 | 南京蓝思信息科技有限公司 | Head movement data processing method and device |
| CN109919034A (en) * | 2019-01-31 | 2019-06-21 | 厦门大学 | A kind of identification of limb action with correct auxiliary training system and method |
| CN110888535A (en) * | 2019-12-05 | 2020-03-17 | 上海工程技术大学 | AR system capable of improving on-site reality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108508974A (en) | Wearable device processor and its data processing method and wearable device | |
| WO2021179852A1 (en) | Image detection method, model training method, apparatus, device, and storage medium | |
| CN110956202B (en) | Image training method, system, medium and intelligent device based on distributed learning | |
| CN113869482A (en) | Intelligent street lamp self-adaptive energy-saving control method and system based on deep reinforcement learning | |
| CN110288513B (en) | Method, apparatus, device and storage medium for changing face attribute | |
| CN114330699A (en) | Neural network structure searching method and device | |
| CN114065863B (en) | Federal learning method, apparatus, system, electronic device and storage medium | |
| CN113742082A (en) | Application resource allocation method and device, computer readable medium and terminal | |
| CN110958750B (en) | Lighting equipment control method and device | |
| CN111061564A (en) | Server capacity adjusting method and device and electronic equipment | |
| CN109357366B (en) | Regulation control method and device, storage medium and air conditioning system | |
| CN111917999A (en) | Video processing method, mobile terminal and readable storage medium | |
| CN112364253A (en) | Method and system for generating intelligent catering service scheme | |
| CN114265320A (en) | Smart home control method and system for analyzing user habits based on deep learning | |
| CN117193524A (en) | Man-machine interaction system and method based on multi-mode feature fusion | |
| CN114698201A (en) | Illumination control method based on artificial illumination and related equipment | |
| CN106681477A (en) | Method for controlling low power consumption motion mode and intelligent terminal | |
| CN117762032A (en) | Intelligent equipment control system and method based on scene adaptation and artificial intelligence | |
| CN116704553B (en) | Human body characteristic identification auxiliary system based on computer vision technology | |
| CN117031971A (en) | Intelligent furniture equipment adjusting method, device, equipment and medium based on intelligent mirror | |
| CN112198853A (en) | Control method and device of intelligent household equipment | |
| CN110413106B (en) | Augmented reality input method and system based on voice and gestures | |
| CN115527501A (en) | Backlight value adjusting method, processor, display terminal and storage medium | |
| CN115145525A (en) | Screen brightness adjustment model training method and device, storage medium and electronic equipment | |
| CN112652325B (en) | Remote voice adjustment method based on artificial intelligence and related equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180907 |
|
| WD01 | Invention patent application deemed withdrawn after publication |