CN103248955A - Identity recognition method and device based on intelligent remote control system - Google Patents
Identity recognition method and device based on intelligent remote control system Download PDFInfo
- Publication number
- CN103248955A CN103248955A CN2013101409868A CN201310140986A CN103248955A CN 103248955 A CN103248955 A CN 103248955A CN 2013101409868 A CN2013101409868 A CN 2013101409868A CN 201310140986 A CN201310140986 A CN 201310140986A CN 103248955 A CN103248955 A CN 103248955A
- Authority
- CN
- China
- Prior art keywords
- similarity
- identity
- behavior data
- user behavior
- control system
- 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
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Selective Calling Equipment (AREA)
Abstract
The invention discloses an identity recognition method and device based on an intelligent remote control system. The method includes the following steps: receiving behavior data of at least one user obtained from a sensor; calculating the comentropy gain rates of behavior data of all the users, and adopting a decision-making tree to classify the behavior data of all the users according to the comentropy gain rates of behavior data of all the users, so as to enable the behavior data of users of the same type to form a sub-dataset; performing modeling for the sub-datasets by adopting a Gaussian model and/or an SVM model to obtain the result of identity recognition; and displaying the result of identity recognition to be recognized and confirmed by users. According to the invention, the sensor recognition technology, decision-making tree classification technology, and the modeling and recognition technology are adopted to precisely recognize identities of users, so that the precision of identities of users in the remote control system is effectively improved.
Description
Technical field
The present invention relates to TV technology, specially refer to personal identification method and device based on intelligent remote control system.
Background technology
At present, the service that intelligent remote control system provides more and more depends on the mutual of itself and user, for example server and intelligent television terminal are given program push by user interactions user that demand is arranged etc., thus realization system and user become basic function indispensable in the intelligent remote control system alternately gradually.It is mutual that but intelligent remote control system of the prior art all can't be realized according to user's concrete identity, and this has seriously limited the service content that intelligent remote control system provides undoubtedly.Personage's identity such as grandfather, father, child, mother are for example arranged in the one family usually, but server and intelligent television terminal but can't realize suitable program push according to these personage's identity, as: recommend to take care of health program according to grandfather's identity, recommend movie program according to father's identity, recommend juvenile's program according to child's identity, recommend shopping program etc. according to mother's identity.Even some remote control system is applied to the user recognition technology of television system with sensor technology now, still, the general identity of identification user that these technology can only be rough can't be accomplished accurate identification, usually causes the situation of identification error to take place.
Summary of the invention
Main purpose of the present invention is for providing a kind of personal identification method and device based on intelligent remote control system of accurate identification user identity.
The present invention proposes a kind of personal identification method based on intelligent remote control system, comprises step:
At least one user behavior data that steps A, receiving sensor are obtained;
Step B calculates the comentropy ratio of profit increase of each user behavior data, and adopts decision tree according to the comentropy ratio of profit increase of described each user behavior data each user behavior data to be sorted out, and makes the user behavior data of same classification constitute a sub-data set;
Step C adopts Gauss model and/or SVM(Support Vector Machine, SVMs) model is to each subdata collection modeling, obtains the identification result;
Step D shows described identification result, calibrates affirmation for the user.
Preferably, described step B specifically comprises:
Determine the comentropy of each user behavior data;
According to described comentropy, determine the ratio of profit increase of each user behavior data;
Ratio of profit increase according to described each user behavior data is set up decision tree;
Adopt decision tree that the user behavior data of described ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
Preferably, described step C specifically comprises:
Adopt Gauss model and/or SVM model to each subdata collection modeling, obtain the similarity of each identity respectively;
The similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
Preferably, among the described step C, when adopting Gauss model and SVM model to each subdata collection modeling, the similarity of described each identity of ordering, the identity that similarity is the highest specifically comprises as identification result's step:
The similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model was identical, direct that described similarity is the highest identity was as the identification result;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model not simultaneously, the similarity of S1 and S2 relatively;
When the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
When the similarity of S1 and S2 not simultaneously, the identity that similarity among S1 and the S2 is higher is as the identification result.
Preferably, also comprise after the described steps A:
In the user behavior data that described transducer obtains, extract the user behavior data belong in the preset reference scope, and will describedly belong to user behavior data in the preset reference scope as being used for computing information entropy production rate among the step B and for the data of decision tree classification.
The present invention also proposes a kind of identity recognition device based on intelligent remote control system, comprising:
Receiver module is used at least one user behavior data that receiving sensor obtains;
Sort module, be used for calculating the comentropy ratio of profit increase of each user behavior data, and adopt decision tree according to the comentropy ratio of profit increase of described each user behavior data each user behavior data to be sorted out, make the user behavior data of same classification constitute a sub-data set;
Identification module is used for adopting Gauss model and/or SVM model to each subdata collection modeling, obtains the identification result;
Display module is used for showing described identification result, calibrates affirmation for the user.
Preferably, described sort module specifically is used for:
Determine the comentropy of each user behavior data;
According to described comentropy, determine the ratio of profit increase of each user behavior data;
Ratio of profit increase according to described each user behavior data is set up decision tree;
Adopt decision tree that the user behavior data of described ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
Preferably, described identification module specifically is used for:
Adopt Gauss model and/or SVM model to each subdata collection modeling, obtain the similarity of each identity respectively;
The similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
Preferably, when described identification module adopted Gauss model and SVM model to each subdata collection modeling, described identification module specifically was used for:
The similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model was identical, direct that described similarity is the highest identity was as the identification result;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model not simultaneously, the similarity of S1 and S2 relatively;
When the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
When the similarity of S1 and S2 not simultaneously, the identity that similarity among S1 and the S2 is higher is as the identification result.
Preferably, described identity recognition device based on intelligent remote control system also comprises the screening module, specifically is used for, and extracts the user behavior data that belongs in the preset reference scope in the user behavior data that described transducer obtains, and is sent to described sort module.
The present invention has adopted transducer recognition technology, decision tree classification technology, modeling and recognition technology, and the realization user identity is accurately identified, and has effectively improved the accuracy of user identity in the remote control system.
Description of drawings
Fig. 1 is the flow chart of first embodiment that the present invention is based on the personal identification method of intelligent remote control system;
Fig. 2 is the flow chart of second embodiment that the present invention is based on the personal identification method of intelligent remote control system;
Fig. 3 is the flow chart of the 3rd embodiment that the present invention is based on the personal identification method of intelligent remote control system;
Fig. 4 is the flow chart of the 4th embodiment that the present invention is based on the personal identification method of intelligent remote control system;
Fig. 5 is the flow chart of the 5th embodiment that the present invention is based on the personal identification method of intelligent remote control system;
Fig. 6 is the structural representation of first embodiment that the present invention is based on the identity recognition device of intelligent remote control system;
Fig. 7 is the structural representation of second embodiment that the present invention is based on the identity recognition device of intelligent remote control system.
The realization of the object of the invention, functional characteristics and advantage will be in conjunction with the embodiments, are described further with reference to accompanying drawing.
Embodiment
Should be appreciated that specific embodiment described herein only in order to explaining the present invention, and be not used in restriction the present invention.
As shown in Figure 1, Fig. 1 is the flow chart of first embodiment that the present invention is based on the personal identification method of intelligent remote control system.The personal identification method based on intelligent remote control system that present embodiment is mentioned comprises:
At least one user behavior data that steps A, receiving sensor are obtained;
In the present embodiment, because each user's behavioural habits are different, for example the user picks up the speed of remote controller, there is difference the aspects such as angle that remote controller is faced the remote control system receiver, the employing transducer obtains the data of user's behavior act, identifies user's identity according to user's concrete behavior data.The transducer that present embodiment adopts can be acceleration transducer, direction sensor, gyro sensor etc., acceleration transducer can obtain that the user picks up or the acceleration when putting down remote controller can obtain the user and picks up the remote controller deflection angle, the rotational angular velocity when gyro sensor can obtain the user and tilts remote controller.For example, father picks up the accekeration of remote controller: x=5, y=20, z=40, direction tilt value: x=0, y=45 degree, z=90 degree; Grandfather picks up the accekeration of remote controller: x=1, y=4, z=10, direction tilt value: x=0, y=0, z=90.Before transducer obtains user behavior data, can be configured each sensor function earlier, for example, the transducer receiving function that unlatching need be used, shielding does not need the transducer receiving function used, when not needing to use the transducer of a certain classification, can directly fall it by the mode of sensors configured in the software end shield of television system, need not to revise the hardware configuration of remote controller.The concrete configuration mode can be with reference to following table:
| Data function | Function is selected (1-open function, 0-closing function) |
| Gyroscope X-axis data | 1 |
| Gyroscope Y-axis data | 1 |
| Gyroscope Z axis data | 1 |
| Direction X-axis data | 0 |
| Direction Y-axis data | 0 |
| Direction Z axis data | 0 |
| Acceleration X-axis data | 1 |
| Acceleration Y-axis data | 1 |
| Acceleration Z axis data | 1 |
According to the configuration of sensor function, adopting, the transducer of open function obtains user behavior data.
Step B calculates the comentropy ratio of profit increase of each user behavior data, and adopts decision tree according to the comentropy ratio of profit increase of each user behavior data each user behavior data to be sorted out, and makes the user behavior data of same classification constitute a sub-data set;
Present embodiment has adopted the mining algorithm analysis and has sorted out user behavior data for the follow-up Model Identification user identity of setting up, and namely is to select attribute for user behavior data, obtains to be used for creating the optimal parameter of mining model.Present embodiment can adopt decision tree and comentropy ratio of profit increase to realize sorted users behavioral data in the mining algorithm, utilize the ratio of profit increase of comentropy to remove the data of duplication similarity, different data are taken out, do the classification action by decision tree then, namely set up decision tree according to the ratio of profit increase of entropy, utilizing decision tree to find out ratio of profit increase from each user behavior data is in data in the same scope and forms data set of son, finally according to a plurality of different gains rate scopes, form a plurality of subdata collection, utilize Gauss model or SVM model to identify in follow-up.
Step C adopts Gauss model and/or SVM model to each subdata collection modeling, obtains the identification result;
According to above-mentioned classification results, each subdata collection is selected Gauss model or selected the SVM model to carry out modeling, and carry out identification.The user identity that above-mentioned identification can be obtained exports display module to and shows all as the identification result, and perhaps a user identity that similarity is the highest exports display module to and shows as the identification result.
Step D shows the identification result, calibrates affirmation for the user.
Display module can be calibrated affirmation by the user after display screen has shown the identification result, if the recognition result that shows has only one, and this recognition result is correct, and then the user confirms that this recognition result is effective; If it is a plurality of that the recognition result that shows has, then the user need select correct identity; If the recognition result that shows is incorrect, then by user's correct identity of typing voluntarily.After the user calibrated affirmation, remote control system can store the model of this user behavior data and employing in corresponding identity subordinate's the data file, when being convenient to next time identification as the reference data.
Present embodiment has adopted transducer recognition technology, decision tree classification technology, modeling and recognition technology, and the realization user identity is accurately identified, and has effectively improved the accuracy of user identity in the remote control system.
As shown in Figure 2, Fig. 2 is the flow chart of second embodiment that the present invention is based on the personal identification method of intelligent remote control system.Also comprise after the steps A:
Step e extracts the user behavior data that belongs in the preset reference scope in the user behavior data that described transducer obtains.
Belong to the interior user behavior data of preset reference scope after the above-mentioned extraction as the data that are used for computing information entropy production rate and sort out for decision tree among the step B.In the user behavior data that transducer obtains, for fear of some extreme data entire identification process is caused significant error, need extract correct data reliably to the user behavior data that obtains, these correct data reliably can be in kinsfolk's the behavioral data term of reference, as following table:
| Identity | Acceleration X-axis term of reference | Acceleration Y-axis term of reference | Acceleration Z axle term of reference |
| Grandfather | 0-30 | 0-30 | 0-30 |
| Grandmother | 0-25 | 0-25 | 0-25 |
| Father | 35-75 | 35-75 | 35-75 |
| Mother | 35-75 | 35-75 | 35-75 |
| Child | 80-100 | 80-100 | 80-100 |
The behavioral data term of reference that is all kinsfolks in this family is 0-100, should extract authentic data in this term of reference.Above-mentioned term of reference is a relative value with respect to normal data, and also the data that can adopt concrete tape unit are as value range, and for example, grandfather's acceleration X-axis term of reference is 0-20m/s
2
Because present embodiment adopts the correct data reliably of extraction to carry out operations such as follow-up classification, identification, the misleading that can avoid extreme data that entire identification process is caused makes recognition result more accurate, has further improved the accuracy of user identity in the remote control system.
As shown in Figure 3, Fig. 3 is the flow chart of the 3rd embodiment that the present invention is based on the personal identification method of intelligent remote control system.Step B specifically comprises:
Step B01 determines the comentropy of each user behavior data;
Tentation data collection S is the data set of an acceleration transducer, Ci is some user action data, freq (Ci, S) be exactly the number of times that Ci occurs in S, | S| is total number of user action data among the data set S, then the comentropy of S can be expressed as Info (S)=-Σ [(freq (and Ci, S)/| S|] log
2[freq (Ci, S)/| S|)].Simultaneously, tentation data collection T is the data set of the X-axis of acceleration transducer, may be calculated Info according to the comentropy after the X branch so
x(T)=Σ [(| Ti|/| T|) Info (Ti)].
Step B02 according to comentropy, determines the ratio of profit increase of each user behavior data;
Computing information entropy production rate is Gain (X)=Info (S) – Info
x(T), make each branch all maximize ratio of profit increase.
Step B03 sets up decision tree according to the ratio of profit increase of each user behavior data;
Size according to ratio of profit increase is set up decision tree, and the user behavior data ratio of profit increase is more big, and this user behavior data is the closer to the decision tree root node.
Step B04 adopts decision tree that the user behavior data of ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
Present embodiment adopts comentropy and decision tree to realize the user behavior data classification, as follow-up modeling parameters, makes recognition result more accurate sorted subdata collection, has further improved the accuracy of user identity in the remote control system.
As shown in Figure 4, Fig. 4 is the flow chart of the 4th embodiment that the present invention is based on the personal identification method of intelligent remote control system.Step C specifically comprises:
Step C01 adopts Gauss model and/or SVM model to each subdata collection modeling, obtains the similarity of each identity respectively;
Present embodiment disposes any model above-mentioned subdata collection is identified, and can arrange in remote control system, or select according to default setting, or adopt two kinds of models to identify simultaneously.Obtain the identity similarity according to Gauss model or selection SVM model.For example, adopt Gauss model to each subdata collection modeling after, obtain following similarity: grandfather's similarity 50%, grandmother's similarity 40%, father's similarity 10%; After adopting the SVM model to each subdata collection modeling, obtain following similarity: grandfather's similarity 70%, grandmother's similarity 30%.
Step C02, the similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
Above-mentioned similarity is sorted, and the identity that similarity is the highest is the result of this identification.For example, in the above-described embodiments, after employing Gauss model and the SVM Model Identification, the identity that similarity is the highest all is grandfather, then " grandfather " is presented on the display screen as the result of identification, for user's calibration.Present embodiment screens recognition result by sequencing of similarity, makes User Recognition more accurate.
As shown in Figure 5, Fig. 5 is the flow chart of the 5th embodiment that the present invention is based on the personal identification method of intelligent remote control system.Among the step C01, when adopting Gauss model and SVM model to each subdata collection modeling, step C02 specifically comprises:
Step C021, the similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
Step C022 judges that the highest identity S1 of the similarity that obtained by Gauss model with whether identical by the highest identity S2 of the similarity of SVM model acquisition, namely judges whether S1=S2=S; If, execution in step C023 then; If not, execution in step C024 then;
Step C023, directly that similarity is the highest identity S is as the identification result;
The following row data instance of present embodiment, adopt Gauss model to each subdata collection modeling after, obtain following similarity: grandfather's similarity 50%, grandmother's similarity 40%, father's similarity 10%; After adopting the SVM model to each subdata collection modeling, obtain following similarity: in the recognition result that grandfather's similarity 70%, 30%, two model of grandmother's similarity obtain, " grandfather " rank is all the highest, " grandfather " is presented on the display screen as the result of identification, for user's calibration.
Step C024, the relatively similarity of S1 and S2;
Step C025, when the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
Following row data instance, the highest similarity identity that adopts Gauss model to obtain is " grandfather ", similarity 50%, the highest similarity identity that adopts the SVM model to obtain is " father ", similarity 50% is presented at " grandfather " and " father " on the display screen simultaneously, for user's calibration.
Step C026, when the similarity of S1 during greater than the similarity of S2, with S1 as the identification result;
Step C027, when the similarity of S1 during less than the similarity of S2, with S2 as the identification result.
Following row data instance, the highest similarity identity that adopts Gauss model to obtain is " grandfather ", similarity 50%, the highest similarity identity that adopts the SVM model to obtain is " father ", similarity 70% is presented at " father " on the display screen, for user's calibration.
Present embodiment adopts two kinds of models to identify simultaneously, and the inaccurate situation of the recognition result of avoiding single model to cause has effectively improved identification accuracy.
As shown in Figure 6, Fig. 6 is the structural representation of first embodiment that the present invention is based on the identity recognition device of intelligent remote control system.The identity recognition device based on intelligent remote control system that present embodiment is mentioned comprises:
In the present embodiment, because each user's behavioural habits are different, for example the user picks up the speed of remote controller, there is difference the aspects such as angle that remote controller is faced the remote control system receiver, the employing transducer obtains the data of user's behavior act, identifies user's identity according to user's concrete behavior data.The transducer that present embodiment adopts can be acceleration transducer, direction sensor, gyro sensor etc., acceleration transducer can obtain that the user picks up or the acceleration when putting down remote controller can obtain the user and picks up the remote controller deflection angle, the rotational angular velocity when gyro sensor can obtain the user and tilts remote controller.For example, father picks up the accekeration of remote controller: x=5, y=20, z=40, direction tilt value: x=0, y=45 degree, z=90 degree; Grandfather picks up the accekeration of remote controller: x=1, y=4, z=10, direction tilt value: x=0, y=0, z=90.Before transducer obtains user behavior data, can be configured each sensor function earlier, for example, the transducer receiving function that unlatching need be used, shielding does not need the transducer receiving function used, when not needing to use the transducer of a certain classification, can directly fall it by the mode of sensors configured in the software end shield of television system, need not to revise the hardware configuration of remote controller.The concrete configuration mode can be with reference to following table:
| Data function | Function is selected (1-open function, 0-closing function) |
| Gyroscope X-axis data | 1 |
| Gyroscope Y-axis data | 1 |
| Gyroscope Z axis data | 1 |
| Direction X-axis data | 0 |
| Direction Y-axis data | 0 |
| Direction Z axis data | 0 |
| Acceleration X-axis data | 1 |
| Acceleration Y-axis data | 1 |
| Acceleration Z axis data | 1 |
According to the configuration of sensor function, adopting, the transducer of open function obtains user behavior data.
Present embodiment has adopted the mining algorithm analysis and has sorted out user behavior data for the follow-up Model Identification user identity of setting up, and namely is to select attribute for user behavior data, obtains to be used for creating the optimal parameter of mining model.Present embodiment can adopt decision tree and comentropy ratio of profit increase to realize sorted users behavioral data in the mining algorithm, utilize the ratio of profit increase of comentropy to remove the data of duplication similarity, different data are taken out, do the classification action by decision tree then, namely set up decision tree according to the ratio of profit increase of entropy, utilizing decision tree to find out ratio of profit increase from each user behavior data is in data in the same scope and forms data set of son, finally according to a plurality of different gains rate scopes, form a plurality of subdata collection, utilize Gauss model or SVM model to identify in follow-up.According to above-mentioned classification results, each subdata collection is selected Gauss model or selected the SVM model to carry out modeling, and carry out identification.The user identity that above-mentioned identification can be obtained exports display module 40 to and shows all as the identification result, and perhaps a user identity that similarity is the highest exports display module 40 to and shows as the identification result.
Present embodiment has adopted transducer recognition technology, decision tree classification technology, modeling and recognition technology, and the realization user identity is accurately identified, and has effectively improved the accuracy of user identity in the remote control system.
In the embodiment of the invention, sort module 20 specifically is used for:
Determine the comentropy of each user behavior data;
According to comentropy, determine the ratio of profit increase of each user behavior data;
Ratio of profit increase according to each user behavior data is set up decision tree;
Adopt decision tree that the user behavior data of ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
In the present embodiment, tentation data collection S is the data set of an acceleration transducer, Ci is some user action data, freq (Ci, S) be exactly the number of times that Ci occurs in S, | S| is total number of user action data among the data set S, then the comentropy of S can be expressed as Info (S)=-Σ [(freq (and Ci, S)/| S|] log
2[freq (Ci, S)/| S|)].Simultaneously, tentation data collection T is the data set of the X-axis of acceleration transducer, may be calculated Info according to the comentropy after the X branch so
x(T)=Σ [(| Ti|/| T|) Info (Ti)].Computing information entropy production rate is Gain (X)=Info (S) – Info
x(T), make each branch all maximize ratio of profit increase.Size according to ratio of profit increase is set up decision tree, and the user behavior data ratio of profit increase is more big, and this user behavior data is the closer to the decision tree root node.Present embodiment adopts comentropy and decision tree to realize the user behavior data classification, as follow-up modeling parameters, makes recognition result more accurate sorted subdata collection, has further improved the accuracy of user identity in the remote control system.
In the embodiment of the invention, identification module 30 specifically is used for:
Adopt Gauss model and/or SVM model to each subdata collection modeling, obtain the similarity of each identity respectively;
The similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
Present embodiment disposes any model above-mentioned subdata collection is identified, and can arrange in remote control system, or select according to default setting, or adopt two kinds of models to identify simultaneously.Obtain the identity similarity according to Gauss model or selection SVM model.For example, adopt Gauss model to each subdata collection modeling after, obtain following similarity: grandfather's similarity 50%, grandmother's similarity 40%, father's similarity 10%; After adopting the SVM model to each subdata collection modeling, obtain following similarity: grandfather's similarity 70%, grandmother's similarity 30%.Above-mentioned similarity is sorted, and the identity that similarity is the highest is the result of this identification.For example, in the above-described embodiments, after employing Gauss model and the SVM Model Identification, the identity that similarity is the highest all is grandfather, then " grandfather " is presented on the display screen as the result of identification, for user's calibration.Present embodiment screens recognition result by sequencing of similarity, makes User Recognition more accurate.
When the embodiment of the invention adopted Gauss model and SVM model to each subdata collection modeling when identification module 30, identification module 30 specifically was used for:
The similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model was identical, direct that similarity is the highest identity was as the identification result;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model not simultaneously, the similarity of S1 and S2 relatively;
When the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
When the similarity of S1 and S2 not simultaneously, the identity that similarity among S1 and the S2 is higher is as the identification result.
The following row data instance of present embodiment, adopt Gauss model to each subdata collection modeling after, obtain following similarity: grandfather's similarity 50%, grandmother's similarity 40%, father's similarity 10%; After adopting the SVM model to each subdata collection modeling, obtain following similarity: in the recognition result that grandfather's similarity 70%, 30%, two model of grandmother's similarity obtain, " grandfather " rank is all the highest, " grandfather " is presented on the display screen as the result of identification, for user's calibration.Following row data instance, the highest similarity identity that adopts Gauss model to obtain is " grandfather ", similarity 50%, the highest similarity identity that adopts the SVM model to obtain is " father ", similarity 50% is presented at " grandfather " and " father " on the display screen simultaneously, for user's calibration.Following row data instance, the highest similarity identity that adopts Gauss model to obtain is " grandfather ", similarity 50%, the highest similarity identity that adopts the SVM model to obtain is " father ", similarity 70% is presented at " father " on the display screen, for user's calibration.Present embodiment adopts two kinds of models to identify simultaneously, and the inaccurate situation of the recognition result of avoiding single model to cause has effectively improved identification accuracy.
As shown in Figure 7, Fig. 7 is the structural representation of second embodiment that the present invention is based on the identity recognition device of intelligent remote control system.Present embodiment has increased screening module 50 on basis embodiment illustrated in fig. 6, specifically is used for:
Judge that user behavior data that transducer obtains is whether in the preset reference scope;
In the user behavior data that described transducer obtains, extract the user behavior data that belongs in the preset reference scope, and be sent to sort module 20.
In the present embodiment, belong to the data that the user behavior data in the preset reference scope is used for computing information entropy production rate and sorts out for decision tree as sort module 20 after the extraction.In the user behavior data that transducer obtains, for fear of some extreme data to the entire identification process guiding that makes the mistake, need extract correct data reliably to the user behavior data that obtains, these correct data reliably should be in kinsfolk's the behavioral data term of reference, as following table:
| Identity | Acceleration X-axis term of reference | Acceleration Y-axis term of reference | Acceleration Z axle term of reference |
| Grandfather | 0-30 | 0-30 | 0-30 |
| Grandmother | 0-25 | 0-25 | 0-25 |
| Father | 35-75 | 35-75 | 35-75 |
| Mother | 35-75 | 35-75 | 35-75 |
| Child | 80-100 | 80-100 | 80-100 |
The behavioral data term of reference that is all kinsfolks in this family is 0-100, should extract authentic data in this term of reference.Above-mentioned term of reference is a relative value with respect to normal data, and also the data that can adopt concrete tape unit are as value range, and for example, grandfather's acceleration X-axis term of reference is 0-20m/s
2
Because present embodiment adopts the correct data reliably of extraction to carry out operations such as follow-up classification, identification, the misleading that can avoid extreme data that entire identification process is caused makes recognition result more accurate, has further improved the accuracy of user identity in the remote control system.
Below only be the preferred embodiments of the present invention; be not so limit claim of the present invention; every equivalent structure or equivalent flow process conversion that utilizes specification of the present invention and accompanying drawing content to do; or directly or indirectly be used in other relevant technical fields, all in like manner be included in the scope of patent protection of the present invention.
Claims (10)
1. the personal identification method based on intelligent remote control system is characterized in that, comprises step:
At least one user behavior data that steps A, receiving sensor are obtained;
Step B calculates the comentropy ratio of profit increase of each user behavior data, and adopts decision tree according to the comentropy ratio of profit increase of described each user behavior data each user behavior data to be sorted out, and makes the user behavior data of same classification constitute a sub-data set;
Step C adopts Gauss model and/or SVM model to each subdata collection modeling, obtains the identification result;
Step D shows described identification result, calibrates affirmation for the user.
2. the personal identification method based on intelligent remote control system according to claim 1 is characterized in that, described step B specifically comprises:
Determine the comentropy of each user behavior data;
According to described comentropy, determine the ratio of profit increase of each user behavior data;
Ratio of profit increase according to described each user behavior data is set up decision tree;
Adopt decision tree that the user behavior data of described ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
3. the personal identification method based on intelligent remote control system according to claim 1 and 2 is characterized in that, described step C specifically comprises:
Adopt Gauss model and/or SVM model to each subdata collection modeling, obtain the similarity of each identity respectively;
The similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
4. the personal identification method based on intelligent remote control system according to claim 3, it is characterized in that, among the described step C, when adopting Gauss model and SVM model to each subdata collection modeling, the similarity of described each identity of ordering, the identity that similarity is the highest specifically comprises as identification result's step:
The similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model was identical, direct that described similarity is the highest identity was as the identification result;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model not simultaneously, the similarity of S1 and S2 relatively;
When the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
When the similarity of S1 and S2 not simultaneously, the identity that similarity among S1 and the S2 is higher is as the identification result.
5. the personal identification method based on intelligent remote control system according to claim 3 is characterized in that, also comprises after the described steps A:
In the user behavior data that described transducer obtains, extract the user behavior data belong in the preset reference scope, and will describedly belong to user behavior data in the preset reference scope as being used for computing information entropy production rate among the step B and for the data of decision tree classification.
6. the identity recognition device based on intelligent remote control system is characterized in that, comprising:
Receiver module is used at least one user behavior data that receiving sensor obtains;
Sort module, be used for calculating the comentropy ratio of profit increase of each user behavior data, and adopt decision tree according to the comentropy ratio of profit increase of described each user behavior data each user behavior data to be sorted out, make the user behavior data of same classification constitute a sub-data set;
Identification module is used for adopting Gauss model and/or SVM model to each subdata collection modeling, obtains the identification result;
Display module is used for showing described identification result, calibrates affirmation for the user.
7. the identity recognition device based on intelligent remote control system according to claim 6 is characterized in that, described sort module specifically is used for:
Determine the comentropy of each user behavior data;
According to described comentropy, determine the ratio of profit increase of each user behavior data;
Ratio of profit increase according to described each user behavior data is set up decision tree;
Adopt decision tree that the user behavior data of described ratio of profit increase in same default ratio of profit increase scope is categorized in the same subdata collection.
8. according to claim 6 or 7 described identity recognition devices based on intelligent remote control system, it is characterized in that described identification module specifically is used for:
Adopt Gauss model and/or SVM model to each subdata collection modeling, obtain the similarity of each identity respectively;
The similarity of each identity that sorts, the identity that similarity is the highest is as the identification result.
9. the identity recognition device based on intelligent remote control system according to claim 8 is characterized in that, when described identification module adopted Gauss model and SVM model to each subdata collection modeling, described identification module specifically was used for:
The similarity of the similarity of each identity that obtained by Gauss model of sorting respectively and each identity of being obtained by the SVM model;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model was identical, direct that described similarity is the highest identity was as the identification result;
When the highest identity S2 of the highest identity S1 of the similarity that is obtained by Gauss model and the similarity that is obtained by the SVM model not simultaneously, the similarity of S1 and S2 relatively;
When the similarity of S1 and S2 is identical, simultaneously with S1 and S2 as the identification result;
When the similarity of S1 and S2 not simultaneously, the identity that similarity among S1 and the S2 is higher is as the identification result.
10. the identity recognition device based on intelligent remote control system according to claim 8, it is characterized in that, also comprise the screening module, concrete being used for extracted the user behavior data that belongs in the preset reference scope at the user behavior data that described transducer obtains, and is sent to described sort module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310140986.8A CN103248955B (en) | 2013-04-22 | 2013-04-22 | Personal identification method and device based on intelligent remote control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310140986.8A CN103248955B (en) | 2013-04-22 | 2013-04-22 | Personal identification method and device based on intelligent remote control system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103248955A true CN103248955A (en) | 2013-08-14 |
| CN103248955B CN103248955B (en) | 2017-07-28 |
Family
ID=48928136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310140986.8A Active CN103248955B (en) | 2013-04-22 | 2013-04-22 | Personal identification method and device based on intelligent remote control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103248955B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546789A (en) * | 2013-08-23 | 2014-01-29 | Tcl集团股份有限公司 | Television program recommendation method and system, and intelligent terminal |
| CN103543832A (en) * | 2013-10-29 | 2014-01-29 | Tcl集团股份有限公司 | User identification method and device based on unbalanced data |
| CN105357565A (en) * | 2014-08-22 | 2016-02-24 | 深圳Tcl新技术有限公司 | Information push method and system for smart television |
| CN106610991A (en) * | 2015-10-23 | 2017-05-03 | 北京国双科技有限公司 | Data processing method and device |
| CN109146574A (en) * | 2018-09-06 | 2019-01-04 | 深圳市木瓜移动科技有限公司 | Ad click cheating monitoring method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005106760A2 (en) * | 2003-10-29 | 2005-11-10 | Smiths Detection Inc. | Chemical and biological agent sensor array |
| CN102302370A (en) * | 2011-06-30 | 2012-01-04 | 中国科学院计算技术研究所 | Method and device for detecting tumbling |
| CN102368293A (en) * | 2011-09-07 | 2012-03-07 | 常州蓝城信息科技有限公司 | Personal identity authentication system based on stealth consciousness |
| CN102750286A (en) * | 2011-04-21 | 2012-10-24 | 常州蓝城信息科技有限公司 | Novel decision tree classifier method for processing missing data |
| CN102831447A (en) * | 2012-08-30 | 2012-12-19 | 北京理工大学 | Method for identifying multi-class facial expressions at high precision |
| CN103019376A (en) * | 2012-12-04 | 2013-04-03 | 深圳Tcl新技术有限公司 | Identity-recognition-based remote control function configuration method and system |
-
2013
- 2013-04-22 CN CN201310140986.8A patent/CN103248955B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005106760A2 (en) * | 2003-10-29 | 2005-11-10 | Smiths Detection Inc. | Chemical and biological agent sensor array |
| CN102750286A (en) * | 2011-04-21 | 2012-10-24 | 常州蓝城信息科技有限公司 | Novel decision tree classifier method for processing missing data |
| CN102302370A (en) * | 2011-06-30 | 2012-01-04 | 中国科学院计算技术研究所 | Method and device for detecting tumbling |
| CN102368293A (en) * | 2011-09-07 | 2012-03-07 | 常州蓝城信息科技有限公司 | Personal identity authentication system based on stealth consciousness |
| CN102831447A (en) * | 2012-08-30 | 2012-12-19 | 北京理工大学 | Method for identifying multi-class facial expressions at high precision |
| CN103019376A (en) * | 2012-12-04 | 2013-04-03 | 深圳Tcl新技术有限公司 | Identity-recognition-based remote control function configuration method and system |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103546789A (en) * | 2013-08-23 | 2014-01-29 | Tcl集团股份有限公司 | Television program recommendation method and system, and intelligent terminal |
| CN103546789B (en) * | 2013-08-23 | 2017-07-11 | Tcl集团股份有限公司 | A kind of TV programme suggesting method, system and intelligent terminal |
| CN103543832A (en) * | 2013-10-29 | 2014-01-29 | Tcl集团股份有限公司 | User identification method and device based on unbalanced data |
| CN103543832B (en) * | 2013-10-29 | 2017-08-04 | Tcl集团股份有限公司 | A kind of user identification method and device based on unbalanced data |
| CN105357565A (en) * | 2014-08-22 | 2016-02-24 | 深圳Tcl新技术有限公司 | Information push method and system for smart television |
| CN106610991A (en) * | 2015-10-23 | 2017-05-03 | 北京国双科技有限公司 | Data processing method and device |
| CN109146574A (en) * | 2018-09-06 | 2019-01-04 | 深圳市木瓜移动科技有限公司 | Ad click cheating monitoring method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103248955B (en) | 2017-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103248955A (en) | Identity recognition method and device based on intelligent remote control system | |
| CN108256568B (en) | Plant species identification method and device | |
| CN102509088B (en) | Hand motion detecting method, hand motion detecting device and human-computer interaction system | |
| CN104573706A (en) | Object identification method and system thereof | |
| CN103763585B (en) | A kind of user's characteristic information acquisition methods, device and terminal device | |
| CN104239416A (en) | User identification method and system | |
| CN103414930A (en) | Remote control system for identifying and sensing user and method thereof | |
| CN109583470A (en) | A kind of explanation feature of abnormality detection determines method and apparatus | |
| CN103402142A (en) | Program list pushing method and device | |
| US11500364B2 (en) | Index selection device and method | |
| CN107491751A (en) | Sitting posture analysis method and device | |
| CN103546789A (en) | Television program recommendation method and system, and intelligent terminal | |
| CN103246448A (en) | Interactive method and remote device for acquiring user identities to perform interaction | |
| CN108072356A (en) | Height measurement method, device and unmanned plane | |
| Nguyen-Dinh et al. | Robust online gesture recognition with crowdsourced annotations | |
| CN108764369B (en) | Figure identification method and device based on data fusion and computer storage medium | |
| CN107465814B (en) | A kind of user's input recognition method based on mobile phone inertial sensor | |
| CN106778587A (en) | State of flight detection method and device based on unmanned plane image | |
| CN107659845B (en) | Method and device for monitoring user state of Internet Protocol Television (IPTV) | |
| CN106156167A (en) | The method of data collection/process, motion information acquisition equipment and system | |
| CN107454970A (en) | A kind of System and method for of movement locus collection and analysis based on ball game | |
| CN106598230A (en) | Virtual reality headset and method and device for tracking spatial attitude information thereof | |
| CN106971134A (en) | It is a kind of can error correction pattern recognition device and method | |
| CN105204626A (en) | Method and device for controlling grading of users | |
| CN115932144B (en) | Chromatograph performance detection method, chromatograph performance detection device, chromatograph performance detection equipment and computer medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |