CN103517118B - The action identification method of a kind of remote controller and system - Google Patents
The action identification method of a kind of remote controller and system Download PDFInfo
- Publication number
- CN103517118B CN103517118B CN201210582780.6A CN201210582780A CN103517118B CN 103517118 B CN103517118 B CN 103517118B CN 201210582780 A CN201210582780 A CN 201210582780A CN 103517118 B CN103517118 B CN 103517118B
- Authority
- CN
- China
- Prior art keywords
- remote controller
- data
- action
- intelligent television
- angular
- 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.)
- Expired - Fee Related
Links
- 230000009471 action Effects 0.000 title claims abstract description 122
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000013499 data model Methods 0.000 claims abstract description 54
- 230000001133 acceleration Effects 0.000 claims abstract description 51
- 238000000605 extraction Methods 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 5
- 241001269238 Data Species 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000004321 preservation Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 8
- 238000004364 calculation method Methods 0.000 abstract description 5
- 230000003068 static effect Effects 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Abstract
The action identification method of a kind of remote controller and system, high in the clouds is that intelligent television pre-builds remote controller action data model and downloads to corresponding intelligent television, and intelligent television receives and carries out data feature extraction analysis formation data characteristics vector after the 3-axis acceleration value of acceleration transducer and three axis angular rate values of angular-rate sensor on remote controller；And carry out matching judgment execution identification with the remote controller action data model preserved and export judged result, data characteristics vector is sent to high in the clouds simultaneously and is updated remote controller action data model forming the remote controller action data model after updating.The present invention is static and the setting up remote controller action data model remote controller action data model can be made to be continued to optimize of dynamic bind, make intelligent television more accurate to the identification of remote controller action, modeling is transferred to high in the clouds and is carried out alleviating the amount of calculation of intelligent television with the process of intelligent television execution identification, improves the intelligent television efficiency to remote controller action recognition.
Description
Technical field
The present invention relates to Intelligent TV remote controller, particularly relate to action identification method and the system of a kind of remote controller.
Background technology
Development along with science and technology, intelligent sensing technology is the most gradually applied in people's daily life, the specific product of intelligent sensing technology performance includes: can sense the Intelligent TV remote controller of people's action, be mounted with gravity and the smart mobile phone of angular-rate sensor, can sense the intelligent wrist band of people's hand motion, and the intelligent shoe of people's motion can be sensed etc., these panoramic intelligent sensing equipment is required to gather a large amount of sensing data analysis and uses during using.
At present, Intelligent TV remote controller the most all controls operation simply by button, and the various sensing datas underusing its collection are further analyzed utilization.Technical intelligence TV remote controller has the most all possessed the function of " empty Mus " now, and this function needs to increase acceleration transducer on a remote control and two sensors of angular-rate sensor realize.Intelligent TV set receives remote controller and is sent to the acceleration transducer of intelligent television end and the data of angular-rate sensor, correspondence calculates the position of mouse, but, major part intelligent television be merely by both sensors will " empty Mus " function do more accurate, more perfect, and these data being transmitted through from intelligent remote controller can also not carried out Intelligent Recognition further, intellectual analysis, to realize the application of more function.
Therefore, prior art could be improved and develop.
Summary of the invention
In place of above-mentioned the deficiencies in the prior art, the present invention solves prior art defect and deficiency, propose a kind of can the action identification method of Intelligent Recognition TV remote controller action and system, by static and dynamic modeling, the model identified can be optimized, enable intelligent television to identify the action of remote controller more accurately.
It is as follows that the present invention solves the technical scheme that technical problem used:
The action identification method of a kind of remote controller, comprises the steps:
The intelligent television that high in the clouds is different pre-builds the remote controller action data model of correspondence and downloads to the preservation of corresponding intelligent television；
Intelligent television receives 3-axis acceleration value and the three axis angular rate values of angular-rate sensor of the acceleration transducer being arranged on remote controller；
Intelligent television carries out data characteristics extraction according to described 3-axis acceleration value and three axis angular rate values and analyzes formation data characteristics vector；
Intelligent television carries out matching judgment execution identification according to described data characteristics vector with the remote controller action data model preserved and exports judged result, data characteristics vector is sent to simultaneously high in the clouds remote controller action data model is updated formed update after remote controller action data model, then will update after remote controller action data model download to corresponding intelligent television preserve.
As improving further, described high in the clouds pre-builds that remote controller action data model is the probability of each classification each characteristic attribute lower in statistics known remote control classification action data set, and calculates what all characteristic attributes completed about the conditional probability value of each classification according to NB Algorithm.
Described 3-axis acceleration value includes the accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space；Three axis angular rate values of angular-rate sensor include the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.
It is to carry out signal statistics analysis according to the 3-axis acceleration value received and three axis angular rate Value Datas time series data in fixed length window that the extraction of described data characteristics is analyzed, and extracted data feature forms data characteristics vector.
The time series data that the extraction of described data characteristics is extracted in fixed length window in analyzing is the average power of extracted data waveform, the standard deviation of data waveform, the amplitude of data waveform, the crest number of data waveform, the trough number of data waveform, the parameter of Fourier transformation and the parameter of Chebyshev transformation.
Described action recognition is to calculate described data characteristics vector probability distribution situation in remote controller action data model according to NB Algorithm to judge its described action classification.
Described action classification includes picking up remote controller, puts down remote controller, clicks on remote controller buttons, picks up remote controller and walk about and hold remote controller shake.
The present invention also provides for a kind of remote controller motion recognition system, including cloud server, intelligent television and remote controller；
Cloud server includes cloud database, and in cloud database, storage has remote controller action data model；
Intelligent television includes transmission module in remote controller action data model module, data feature values extraction analysis module, remote controller action recognition module and the data characteristics Tong Bu with cloud server；
Remote controller includes acceleration transducer, angular-rate sensor；
Described remote controller is for obtaining 3-axis acceleration value by acceleration transducer, obtain three axis angular rate Value Datas by angular-rate sensor and be sent to intelligent television, intelligent television receives 3-axis acceleration value and the three axis angular rate Value Datas of angular-rate sensor of remote controller acceleration transducer, analyze module extracted data feature by data feature values extraction and form data characteristics vector, the remote controller action recognition module of intelligent television carries out probability distribution situation analysis according to this data characteristics vector with the remote controller action data model of storage in remote controller action data model module and judges the remote controller action belonging to this data characteristics vector；In described data characteristics, data feature values extraction is analyzed the data characteristics vector of module extraction simultaneously and is uploaded to cloud server and preserve to cloud database and update remote controller action data model by transmission module.
Described 3-axis acceleration value includes the accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space；Three axis angular rate values of angular-rate sensor include the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.
nullCompared with prior art，The recognition methods of remote controller action of the present invention arranges remote controller action data model the most beyond the clouds，Compare with remote controller action data model Tong Bu with high in the clouds in intelligent television after receiving the 3-axis acceleration value of remote controller and three axis angular rate Value Data extracted data features by intelligent television and obtain the action of remote controller，Upload to the data characteristics received again form new remote controller action data model after the remote controller action data model that high in the clouds stores by high in the clouds dynamically updates，The process setting up remote controller action data model of this static state and dynamic bind can make remote controller action data model be continued to optimize，Make follow-up intelligent television more accurate to the identification of remote controller action，Simultaneously，The workload of modeling is transferred to high in the clouds carry out，The process of intelligent television execution identification，Alleviate the amount of calculation of intelligent television，Improve the intelligent television efficiency to remote controller action recognition.
Accompanying drawing explanation
Fig. 1 is the workflow diagram of the action identification method of a kind of remote controller of the present invention.
Fig. 2 is the theory structure block diagram of the motion recognition system of a kind of remote controller of the present invention.
Detailed description of the invention
For making the purpose of the present invention, technical scheme and advantage clearer, clear and definite, the present invention is described in more detail for the embodiment that develops simultaneously referring to the drawings.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
The sensing motion that the present invention is directed to remote controller is studied, by the sensing data of remote controller is modeled, by user behavior is analyzed by the action collecting user operation remote controller on sensor, preferably to apply it in intelligent television field, the application for intelligent television provides data source.
The remote controller sensing data that the inventive method is gathered is remote controller based on prior art intelligent television, or the acceleration transducer being used widely on other mobile terminal devices and angular-rate sensor realize, for acceleration transducer and angular-rate sensor, it all produces data on three direction of principal axis, i.e. accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space, the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.High in the clouds is that cloud computing technology based on prior art realizes, and intelligent television keeps network to be connected, so that realizing uploading and downloading of data between high in the clouds and intelligent television with high in the clouds.
It is the workflow diagram of the action identification method of a kind of remote controller of the present invention shown in Fig. 1, as it is shown in figure 1, the inventive method comprises the following steps:
S100, the intelligent television that high in the clouds is different pre-builds the remote controller action data model of correspondence and downloads to the preservation of corresponding intelligent television；In order to realize the intelligent television identification to remote controller action, it is necessary first to data base beyond the clouds saves as the remote controller action data model that intelligent television pre-builds.High in the clouds pre-builds that remote controller action data model needs to add up the probability of each classification each characteristic attribute lower in known remote control classification action data set.Generally speaking, user includes picking up remote controller, puts down remote controller, clicks on remote controller buttons, holds the classification action that remote controller is walked about, hold remote controller shake the action of remote controller, the various classification action datas of remote controller to be obtained, acceleration transducer and the angular-rate sensor data waveform change in fixed length time window in the various classification actions to remote controller is needed to carry out statistical analysis, rule of thumb, the data waveform variation characteristic of the classification action of above-mentioned various remote controller is as shown in table 1:
The data waveform variation characteristic of table 1 remote controller classification action
| Remote controller classification action | Data waveform variation characteristic |
| Pick up remote controller | The a certain number of axle of remote controller, according in obvious spininess, is the most motionless waveform before spine |
| Put down remote controller | The a certain number of axle of remote controller, according in obvious spininess, is steady motionless waveform after spine |
| Click on remote controller buttons | Timing waveform is little amplitude spine, and the identification of this action can coordinate the button of remote controller to be modeled |
| Hold remote controller to walk about | Acceleration change is cycle shape, and angular velocity change is substantially |
| Hold remote controller shake | Acceleration change period frequency is obvious, and amplitude is less |
The action of above-mentioned various remote controller all has acceleration transducer and angular-rate sensor change of data waveform on three direction of principal axis of correspondence, the process of modeling needs first to form training sample set according to the data of sorting item known to these, then the conditional probability adding up each characteristic attribute under each action classification is estimated, the conditional probability for the characteristic attribute of known remote control classification action has calculated according to NB Algorithm.Such as, the characteristic set of above-mentioned 5 remote controller classifications of motion is respectively by y0、y1、y2、y3、y4Represent, a1、a2、a3、amRepresent the feature value vector of an action respectively, add up feature set incompatible calculating each eigenvalue a of the corresponding each remote controller classification of motion of feature value vector of each action respectivelymThe conditional probability of each remote controller classification of motion corresponding, i.e. eigenvalue a1The conditional probability of each type corresponding is P (a1/y0),P(a1/y1),P(a1/y2),P(a1/y3),P(a1/y4), eigenvalue a2The conditional probability of each type corresponding is P (a2/y0),P(a2/y1),P(a2/y2),P(a2/y3),P(a2/y4), eigenvalue a3The conditional probability of each type corresponding is P (a3/y0),P(a3/y1),P(a3/y2),P(a3/y3),P(a3/y4), eigenvalue amThe conditional probability of each type corresponding is P (am/y0),P(am/y1),P(am/y2),P(am/y3),P(am/y4), above-mentioned all conditions probability record is completed modeling process.
S200, intelligent television receives 3-axis acceleration value and the three axis angular rate values of angular-rate sensor of the acceleration transducer being arranged on remote controller；3-axis acceleration value includes the accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space；Three axis angular rate values of angular-rate sensor include the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.
S300, intelligent television carries out data characteristics extraction according to described 3-axis acceleration value and three axis angular rate values and analyzes formation data characteristics vector；It is to carry out signal statistics analysis according to the 3-axis acceleration value received and three axis angular rate Value Datas time series data in fixed length window that the extraction of described data characteristics is analyzed, and extracted data feature forms data characteristics vector.The time series data that the extraction of described data characteristics is extracted in fixed length window in analyzing is the average power of extracted data waveform, the standard deviation of data waveform, the amplitude of data waveform, the crest number of data waveform, the trough number of data waveform, the parameter of Fourier transformation and the parameter of Chebyshev transformation.
Table 2
| Characteristic parameter | Feature calculation method |
| The average power of data waveform | All waveform point value superpositions are divided by counting |
| The standard deviation of data waveform | Each data point deduct average square divided by sqrt after counting |
| The amplitude of data waveform | The maximum of waveform deducts minima |
| The crest number of data waveform | Traveling through each point, the point that current point is bigger than front n point and rear n point is crest, and n is generally higher than 3 |
| The trough number of data waveform | Traveling through each point, the point that current point is less than front n point and rear n point is trough, and n is generally higher than 3 |
| The parameter of Fourier transformation | Use mathematics bag that wave sequence carries out Fourier transformation, intercept the component sine waves on the bigger front k rank of amplitude as parameter |
| The parameter of Chebyshev transformation | Using mathematics bag that wave sequence carries out Chebyshev transformation, before intercepting, m Chebyshev polynomials coefficient is as characteristic parameter |
S400, intelligent television carries out matching judgment execution identification according to described data characteristics vector with the remote controller action data model preserved and exports judged result；Described action recognition is to calculate described data characteristics vector probability distribution situation in remote controller action data model according to NB Algorithm to judge its described action classification.Described action classification includes picking up remote controller, puts down remote controller, clicks on remote controller buttons, picks up remote controller and walk about and hold remote controller shake.Still with the characteristic value collection in above-mentioned modeling process, feature value vector, identification process being described, process is as follows:
1, X={a is set1, a2, a3... amIt is the feature value vector set of the remote controller that intelligent television receives, a1, a2, a3... amFor the eigenvalue attribute of each action, these eigenvalue attributes are the characteristic parameters extracted in the 3-axis acceleration value according to remote controller and three axis angular rate values；
2, remote controller action classification set C={y is set0, y1, y2, y3, y4……yn, wherein y0、y1、y2、y3、y4For above-mentioned 5 remote controller action classifications, in addition to these 5 actions, it is also possible to define other classes as required, do not limit；
3, calculate, i.e. calculate each action classification and concentrate the probability occurred at training sample, if each eigenvalue attribute is conditional sampling, then has according to Bayes theorem and derive as follows:
Because denominator is constant for everything classification, as long as molecule is maximized, again because each eigenvalue attribute is conditional sampling, so having:
Wherein P (aj|yi) it is to be calculated by the probability distribution situation of canned data in model.
If 4, thenIf type of action is maximum relative to X conditional probability, then be judged to this type of action.
S500, data characteristics vector is sent to high in the clouds and is updated remote controller action data model forming the remote controller action data model after updating by intelligent television；Bigger owing to remote controller action data model to be modeled again the amount of calculation of renewal, in order to reduce the amount of calculation of intelligent television end, intelligent television also needs to the data characteristics vector of the remote controller received is sent to high in the clouds, by high in the clouds according to this data characteristics vector to original remote controller action data model, i.e. training sample set is updated, and so can realize dynamically being updated remote controller action data model.
S600, the remote controller action data model after high in the clouds will update downloads to the preservation of corresponding intelligent television so that intelligent television follow-up remote controller action recognition is more accurate.Intelligent television i.e. keeps Tong Bu with the remote controller action data model in high in the clouds updating remote controller action data model, follow-up can be more accurate when being remotely controlled device action recognition.
The present invention also proposes a kind of remote controller motion recognition system, including cloud server, intelligent television and remote controller.As in figure 2 it is shown,
Cloud server 10 includes cloud database 11, and in cloud database 11, storage has remote controller action data model；
Intelligent television 20 includes transmission module 24 in remote controller action data model module 21, data feature values extraction analysis module 22, remote controller action recognition module 23 and the data characteristics Tong Bu with cloud server；
Remote controller 30 includes acceleration transducer 31, angular-rate sensor 32；
Described remote controller 30 is for obtaining 3-axis acceleration value by acceleration transducer 31, obtain three axis angular rate Value Datas by angular-rate sensor 32 and be sent to intelligent television 20, intelligent television 20 receives the 3-axis acceleration value of remote controller acceleration transducer 31 and three axis angular rate Value Datas of angular-rate sensor 32, analyze module extracted data feature by data feature values extraction and form data characteristics vector, the remote controller action recognition module 23 of intelligent television 20 carries out probability distribution situation analysis according to this data characteristics vector with the remote controller action data model of storage in remote controller action data model module 21 and judges the remote controller action belonging to this data characteristics vector；In described data characteristics, the data characteristics vector that module extraction is analyzed in data feature values extraction is uploaded to cloud server 10 and preserves to cloud database 11 and update remote controller action data model simultaneously by transmission module 24.
Its application operation principle is identical with the action identification method of aforesaid remote controller, does not repeats.
Should be understood that; the foregoing is only presently preferred embodiments of the present invention; it is not sufficient to limit technical scheme; for those of ordinary skills; within the spirit and principles in the present invention; can be increased and decreased according to the above description, replaced, converted or be improved, and all these increase and decrease, replace, convert or improve after technical scheme, all should belong to the protection domain of claims of the present invention.
Claims (7)
1. the action identification method of a remote controller, it is characterised in that comprise the steps:
The intelligent television that high in the clouds is different pre-builds the remote controller action data model of correspondence and downloads to the preservation of corresponding intelligent television；
Intelligent television receives 3-axis acceleration value and the three axis angular rate values of angular-rate sensor of the acceleration transducer being arranged on remote controller；
Intelligent television carries out data characteristics extraction according to described 3-axis acceleration value and three axis angular rate values and analyzes formation data characteristics vector；
Intelligent television carries out matching judgment execution identification according to described data characteristics vector with the remote controller action data model preserved and exports judged result, data characteristics vector is sent to simultaneously high in the clouds remote controller action data model is updated formed update after remote controller action data model, then will update after remote controller action data model download to corresponding intelligent television preserve；
It is to carry out signal statistics analysis according to the 3-axis acceleration value received and three axis angular rate Value Datas time series data in fixed length window that the extraction of described data characteristics is analyzed, and extracted data feature forms data characteristics vector；
The time series data that the extraction of described data characteristics is extracted in fixed length window in analyzing is the average power of extracted data waveform, the standard deviation of data waveform, the amplitude of data waveform, the crest number of data waveform, the trough number of data waveform, the parameter of Fourier transformation and the parameter of Chebyshev transformation.
The action identification method of remote controller the most according to claim 1, it is characterized in that, described high in the clouds pre-builds that remote controller action data model is the probability of each classification each characteristic attribute lower in statistics known remote control classification action data set, and calculates what all characteristic attributes completed about the conditional probability value of each classification according to NB Algorithm.
The action identification method of remote controller the most according to claim 1, it is characterised in that described 3-axis acceleration value includes the accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space；Three axis angular rate values of angular-rate sensor include the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.
The action identification method of remote controller the most according to claim 1, it is characterized in that, described action recognition is to calculate described data characteristics vector probability distribution situation in remote controller action data model according to NB Algorithm to judge its described action classification.
The action identification method of remote controller the most according to claim 4, it is characterised in that described action classification includes picking up remote controller, puts down remote controller, clicks on remote controller buttons, picks up remote controller and walk about and hold remote controller shake.
6. a remote controller motion recognition system, it is characterised in that include cloud server, intelligent television and remote controller；
Cloud server includes cloud database, and in cloud database, storage has remote controller action data model；
Intelligent television includes transmission module in remote controller action data model module, data feature values extraction analysis module, remote controller action recognition module and the data characteristics Tong Bu with cloud server；
Remote controller includes acceleration transducer, angular-rate sensor；
Described remote controller is for obtaining 3-axis acceleration value by acceleration transducer, obtain three axis angular rate Value Datas by angular-rate sensor and be sent to intelligent television, intelligent television receives 3-axis acceleration value and the three axis angular rate Value Datas of angular-rate sensor of remote controller acceleration transducer, analyze module extracted data feature by data feature values extraction and form data characteristics vector, the remote controller action recognition module of intelligent television carries out probability distribution situation analysis according to this data characteristics vector with the remote controller action data model of storage in remote controller action data model module and judges the remote controller action belonging to this data characteristics vector；In described data characteristics, data feature values extraction is analyzed the data characteristics vector of module extraction simultaneously and is uploaded to cloud server and preserve to cloud database and update remote controller action data model by transmission module；
The extraction of described data feature values analyzes module for carrying out signal statistics analysis, extracted data feature formation data characteristics vector according to the 3-axis acceleration value received and three axis angular rate Value Datas time series data in fixed length window；
The time series data that the extraction of described data characteristics is extracted in fixed length window in analyzing is the average power of extracted data waveform, the standard deviation of data waveform, the amplitude of data waveform, the crest number of data waveform, the trough number of data waveform, the parameter of Fourier transformation and the parameter of Chebyshev transformation.
Remote controller motion recognition system the most according to claim 6, it is characterised in that described 3-axis acceleration value includes the accekeration on acceleration transducer X, Y, Z axis direction in three-dimensional space；Three axis angular rate values of angular-rate sensor include the magnitude of angular velocity on angular-rate sensor X, Y, Z axis direction in three-dimensional space.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210582780.6A CN103517118B (en) | 2012-12-28 | 2012-12-28 | The action identification method of a kind of remote controller and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210582780.6A CN103517118B (en) | 2012-12-28 | 2012-12-28 | The action identification method of a kind of remote controller and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103517118A CN103517118A (en) | 2014-01-15 |
| CN103517118B true CN103517118B (en) | 2016-08-03 |
Family
ID=49899025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210582780.6A Expired - Fee Related CN103517118B (en) | 2012-12-28 | 2012-12-28 | The action identification method of a kind of remote controller and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103517118B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105187928B (en) * | 2015-08-11 | 2019-01-01 | 深圳Tcl数字技术有限公司 | Using the method and television set of display |
| CN105160851A (en) * | 2015-09-01 | 2015-12-16 | 胡磊 | Individualized design type intelligent wireless controller |
| CN105582669A (en) * | 2015-12-21 | 2016-05-18 | 上海交通大学 | Motion sensing game bracelet system based on motion template feature matching |
| CN106406516A (en) * | 2016-08-26 | 2017-02-15 | 兰州理工大学 | Local real-time movement trajectory characteristic extraction and identification method for smartphone |
| CN106647398A (en) * | 2016-12-23 | 2017-05-10 | 广东美的制冷设备有限公司 | Remote controller, operation control method and device |
| CN107797666A (en) * | 2017-11-21 | 2018-03-13 | 出门问问信息科技有限公司 | Gesture identification method, device and electronic equipment |
| CN109388669A (en) * | 2018-08-31 | 2019-02-26 | 上海奥孛睿斯科技有限公司 | The acquisition of internet-of-things terminal motion state and analysis system and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101620769A (en) * | 2008-06-30 | 2010-01-06 | 鼎亿数码科技(上海)有限公司 | Remote controller |
| CN101788861A (en) * | 2009-01-22 | 2010-07-28 | 华硕电脑股份有限公司 | Method and system for identifying three-dimensional motion |
| CN201699853U (en) * | 2010-02-03 | 2011-01-05 | 姜开淦 | Remote controller with action recognition function |
| CN102014255A (en) * | 2010-11-30 | 2011-04-13 | 广东星海数字家庭产业技术研究院有限公司 | Digital television remote control equipment with embedded acceleration transducer |
| CN102024151A (en) * | 2010-12-02 | 2011-04-20 | 中国科学院计算技术研究所 | Training method of gesture motion recognition model and gesture motion recognition method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008269174A (en) * | 2007-04-18 | 2008-11-06 | Fujifilm Corp | Control device, method, and program |
-
2012
- 2012-12-28 CN CN201210582780.6A patent/CN103517118B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101620769A (en) * | 2008-06-30 | 2010-01-06 | 鼎亿数码科技(上海)有限公司 | Remote controller |
| CN101788861A (en) * | 2009-01-22 | 2010-07-28 | 华硕电脑股份有限公司 | Method and system for identifying three-dimensional motion |
| CN201699853U (en) * | 2010-02-03 | 2011-01-05 | 姜开淦 | Remote controller with action recognition function |
| CN102014255A (en) * | 2010-11-30 | 2011-04-13 | 广东星海数字家庭产业技术研究院有限公司 | Digital television remote control equipment with embedded acceleration transducer |
| CN102024151A (en) * | 2010-12-02 | 2011-04-20 | 中国科学院计算技术研究所 | Training method of gesture motion recognition model and gesture motion recognition method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103517118A (en) | 2014-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103517118B (en) | The action identification method of a kind of remote controller and system | |
| Li et al. | Dynamic gesture recognition in the internet of things | |
| CN104899561A (en) | Parallelized human body behavior identification method | |
| CN104780217B (en) | Detect method, system and the client of user job efficiency | |
| CN107016342A (en) | A kind of action identification method and system | |
| CN105528613A (en) | Behavior identification method based on GPS speed and acceleration data of smart phone | |
| CN105797350A (en) | Intelligent method and system for body building posture recognition, evaluation, early-warning and intensity estimation | |
| CN103546789B (en) | A kind of TV programme suggesting method, system and intelligent terminal | |
| CN110610158A (en) | Human body posture identification method and system based on convolution and gated cyclic neural network | |
| Divya et al. | Docker-based intelligent fall detection using edge-fog cloud infrastructure | |
| CN105807903A (en) | Control method and device of intelligent equipment | |
| CN110069715A (en) | A kind of method of information recommendation model training, the method and device of information recommendation | |
| Helmi et al. | Human activity recognition using marine predators algorithm with deep learning | |
| WO2017004880A1 (en) | Method, device for behavior recognition and computer storage medium | |
| CN101694521A (en) | Target predicting and tracking method based on probability graph model | |
| CN106406518A (en) | Gesture control device and gesture recognition method | |
| CN103514240B (en) | A kind of kinsfolk's relation excavation method and system based on remote controller | |
| Zhang et al. | Gaitid: Robust wi-fi based gait recognition | |
| CN103065330A (en) | Method and device for particle filter target tracking based on pipeline parallel processing technique | |
| CN108960430A (en) | The method and apparatus for generating personalized classifier for human body motor activity | |
| Li | Dance motion capture based on data fusion algorithm and wearable sensor network | |
| CN104792327A (en) | Motion track contrasting method based on mobile device | |
| JP2016010562A (en) | Data analyzer, data analysis method, and data analysis program | |
| Ameliasari et al. | An evaluation of svm in hand gesture detection using imu-based smartwatches for smart lighting control | |
| CN107092882A (en) | A kind of Activity recognition system and its method of work perceived based on sub- action |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160803 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |