Summary of the invention
In view of this, the invention provides a kind of control method of information automatic asking and a kind of control device of information automatic asking, can need not the user and manually import identity information, can discern the different user that uses same computer automatically.
The control method of a kind of information automatic asking provided by the invention is provided with the sensitive component that links to each other with TV, and described sensitive component is used to take the face images of users of watching this TV, and this method comprises:
A, the facial image of taking is carried out people's face detect;
B, detected target people's face is mated with each faceform of having made up;
The faceform that c, basis set in advance and the corresponding relation of identify label are determined the pairing identify label of faceform with detected target people's face coupling;
The identify label that d, basis are determined is searched corresponding information and is obtained record, is used to initiate asking for automatically of corresponding informance;
E, know the program category of the current broadcast of TV, belong to same type, process ends then, otherwise execution in step f if wait the corresponding informance of initiating to ask for automatically and the program category of the current broadcast of TV by the central control system of TV inside;
The information of f, this identify label correspondence that will find is obtained record and is sent to remote server in the network, and record pushes corresponding information and the mode by roll titles presents so that the remote server in the network can obtain according to the information that receives.
After detected target people's face mated with each faceform of having made up, if do not find faceform with detected target people's face coupling, then this method further comprises: the faceform who makes up a coupling for this target people face, and be this target people face definition unique corresponding identify label, process ends then.
This method is further searched the information of identify label correspondence and is obtained record in described TV.
This method is further searched the information of identify label correspondence and is obtained record in described remote server.
The control device of a kind of information automatic asking provided by the invention, this device comprises:
People's face detecting unit, the facial image that sensitive component is exported carries out the detection of people's face, and described sensitive component links to each other with TV and is used to take the face images of users of watching this TV;
The Model Matching unit mates each faceform of detected target people's face of described people's face detecting unit and storage inside;
Identity determination unit behind the faceform who obtains mating with described detected target people's face in described Model Matching unit, according to the faceform of storage inside and the corresponding relation of identify label, is determined the pairing identify label of this faceform;
The record search unit obtains in the record in all information of adding up in advance, searches the information of the definite identify label correspondence of described identity determination unit and obtains record, is used to initiate asking for automatically of corresponding informance;
Communication enables the unit, be used for knowing the program category of the current broadcast of TV by the central control system of TV inside, belong to same type if wait the corresponding informance of initiating to ask for automatically and the program category of the current broadcast of TV, forbid that then the telecommunication unit obtains record with information and is sent to remote server in the network, otherwise, allow the telecommunication unit that information is obtained and write down the remote server that is sent in the network;
The telecommunication unit, be used for that the information that described record search unit finds is obtained record and be sent to the remote server of network, record pushes corresponding information and the mode by roll titles presents so that the remote server in the network can obtain according to the information that receives.
Further comprise: the modelling unit, be used for after described Model Matching unit does not match the faceform of mating with described detected target people's face, make up with the faceform of this target people face coupling, feed back to the storage of described Model Matching unit, and for the unique corresponding identify label of definition of this target people face and export described identity determination unit storage to.
Further comprise physical storage medium in the described main frame, be used to store described all information of statistics in advance and obtain record.Described all information of statistics are in advance obtained recording storage in described remote server.
As seen from the above technical solution, the present invention utilizes sensitive component to take user's facial image of current use main frame, utilize the identity of people's face testing result of facial image being discerned this user then, and utilize the user identity that identifies to control asking for automatically of corresponding informance, thereby can discern the different user that uses same main frame, also need not the user and manually import identity information.
Embodiment
For making purpose of the present invention, technical scheme and advantage clearer, below with reference to the accompanying drawing embodiment that develops simultaneously, the present invention is described in more detail.
One sensitive component is set in the present invention and it is linked to each other with main frame that for example computer etc. is connected to network, this sensitive component will photograph the user's facial image that uses this main frame and be sent in the main frame, make the application program that is carried on processor in the main frame carry out facial image after people's face detects, can identify the user identity that uses this main frame according to people's face testing result, and the information automatic asking that utilizes the user identity that identifies to initiate this user's correspondence.
Below, be example with computer as main frame earlier, describe for an embodiment.
Embodiment one
Fig. 1 is the schematic flow sheet of the control method of information automatic asking in the embodiment of the invention one.As shown in Figure 1, being used to of linking to each other with computer be set take after the sensitive component of user's facial image of use computer, the control method of information automatic asking comprises in the present embodiment:
Step 101, the facial image that sensitive component is exported carries out the detection of people's face.
In this step, can utilize self-adaptation to strengthen various existing algorithms such as (AdaBoost) and realize that people's face detects, concrete testing process is that prior art can realize, does not repeat them here.
Step 102 is mated detected target people's face successively with each faceform who has made up, obtain the faceform with detected target people's face coupling.
In this step, also can realize, not repeat them here for prior art with faceform's matching process.
In this step, if do not find faceform with detected target people's face coupling, then can make up the faceform of a coupling for this target people face that does not mate, be used for preserving the image feature of this target people face, and be this target people face definition unique corresponding identify label, process ends then.Wherein, the faceform who makes up coupling can realize for prior art, does not repeat them here.
After this, can be according to the identify label of definition, add up the address link of the pairing user capture of this target people's face, the information such as key word in the information obtained are obtained record, so that after this user is connected to network, it is realized information automatic asking next time.
That is to say that each faceform who has made up described in the step 102 is meant each user who uses this computer, the faceform who makes up when using this computer for the first time.Certainly, among each faceform who has made up described in the step 102, also can be included in the user and use before this computer the faceform who makes up in advance and store for the first time.
Step 103 according to the corresponding relation of faceform who sets in advance and identify label, determines and the pairing identify label of faceform of detected target people's face coupling that the user of current use computer is promptly represented in this identify label.
Step 104 is searched corresponding information according to the identify label of determining and is obtained record, and the pairing information of the user of promptly current use computer is obtained record.
In the practical application, information is obtained record and can be recorded in the local computing, also can be recorded in the remote server.Therefore, in this step,, can in computer, search the information of identify label correspondence and obtain record, realize that promptly this locality searches for the previous case; And for latter event, then can in remote server, search the information of identify label correspondence and obtain record, promptly realize long-range searching.
Step 105 is obtained record with the information of this identify label correspondence of finding and is sent to remote server in the network, initiates asking for automatically of corresponding informance.
In this step, the initiation process of asking for automatically can be identical with existing mode, do not repeat them here.
So far, this flow process finishes.
After finishing above-mentioned flow process, the remote server in the network can obtain record according to the information that receives, and pushes corresponding information and presents to the user by computer.
Fig. 2 is the structural representation of the control device of information automatic asking in the embodiment of the invention one.As shown in Figure 2, the control device of information automatic asking comprises in the present embodiment:
Sensitive component 200 is used to take face images of users.
People's face detecting unit 201 is used for the facial image of sensitive component 200 outputs is carried out the detection of people's face.
Model Matching unit 202, be used for people's face detecting unit 201 detected target people's faces are mated with each faceform of storage inside successively, if find faceform with people's face detecting unit 201 detected target people's face couplings, then notify identity determination unit 203, and export identity determination unit 203 to after this faceform duplicated, maybe the preset model sign with this faceform exports identity determination unit 203 to.
Identity determination unit 203, be used for after Model Matching unit 202 finds the faceform of coupling, according to the faceform of storage inside and the corresponding relation of identify label, determine the pairing identify label of faceform of the coupling that Model Matching unit 202 finds, the user of current use computer is promptly represented in this identify label.
Record search unit 204 is used for obtaining record in all information of statistics in advance, searches the information of the identify label correspondence that identity determination unit 203 determines and obtains record, and the pairing information of the user of promptly current use computer is obtained record.
Telecommunication unit 205, can realize information interaction by any one default existing communication agreement and remote server, be used for the information that record search unit 204 finds is obtained the remote server that record is sent to network, ask for corresponding information automatically.
In the present embodiment, device as shown in Figure 2 can also comprise: modelling unit 206.
In this case, if Model Matching unit 202 does not find the faceform with people's face detecting unit 201 detected target people's face couplings, the identify label of not writing down the user of this target people face correspondence before the expression, promptly this user is new user, then notification model is set up unit 206, and exports target people face to modelling unit 206.
Modelling unit 206, the target people's face that is used for 202 outputs from the Model Matching unit extracts characteristics of image, the faceform of structure and this target people face coupling also feeds back to 202 storages of Model Matching unit, also defines a unique corresponding identify label for this target people face simultaneously and export identity determination unit 203 to store.
Like this, increased new user's identify label in the identity determination unit 203 newly, be used in local computing or the remote server statistical information obtain the application program of record can be at newly-increased identify label, information such as key word in the information that the address of adding up new user capture links, obtains are obtained record, so that after this new user is connected to network next time, it is realized information automatic asking.
In the practical application, Tong Ji all information are obtained in the physical storage medium (not shown among Fig. 2) that record can be stored in local computing, also can be recorded in the remote server in advance.Therefore, for the previous case, record search unit 204 can be searched the information of identify label correspondence and obtain record in the physical storage medium of computer, realizes that promptly this locality searches; And for latter event, then record search unit 204 can be mutual by telecommunication unit 205 and remote server, in remote server, search the information of identify label correspondence and obtain record, promptly realize long-range searching, certainly, if record search unit 204 has the network communication function, then also can be directly and remote server mutual.
Above-mentioned other functional units except that sensitive component 200 all can be arranged in the processor in the computer, are realized by the application program that processor carried in the computer.
As seen from the above-described embodiment, the present invention utilizes sensitive component to take user's facial image of current use computer, utilize identity then to this user of IP Address Recognition of people's face testing result of facial image rather than computer, and utilize the user identity that identifies to control asking for automatically of corresponding informance, thereby can discern the different user that uses same computer, also need not the user and manually import identity information.
For example, some email service provider, by Email content that analysis user sent and received or key word wherein, know this interested field of user indirectly, as the finance and economics dynamic dispatching, then the user need not to login e-mail page and whether computer that no matter this user uses has changed the IP address, this service provider all can pass through the mode of propelling movement wittingly, in any page that this user opened, place the news links of some and finance and economics dynamical correlation, with effective increase user clicking rate, and lifting user's service satisfaction.
The foregoing description one only is to be example with computer as the main frame that is connected to network, in the practical application, can also utilize TV as main frame and be connected in the network, and this has just constituted the embodiment two among the present invention.
Embodiment two
Need to prove, utilize TV as main frame and be connected in the network, need in TV, add a processor, and this processor and sensitive component link to each other with remote server in the network.
Fig. 3 is the schematic flow sheet of the control method of information automatic asking in the embodiment of the invention two.As shown in Figure 2, the sensitive component link to each other with TV is set, is used to take after user's facial image of watching TV, the control method of information automatic asking comprises in the present embodiment:
Step 301, the facial image that sensitive component is exported carries out the detection of people's face.
In this step, can utilize various existing algorithms realization people faces detections such as AdaBoost, concrete testing process is that prior art can realize, does not repeat them here.
Step 302 is mated detected target people's face successively with each faceform who has made up, obtain the faceform with detected target people's face coupling.
In this step, also can realize, not repeat them here for prior art with faceform's matching process.
In this step, if do not find faceform with detected target people's face coupling, then can make up the faceform of a coupling for this target people face that does not mate, be used for preserving the image feature of this target people face, and be this target people face definition unique corresponding identify label, process ends then.Wherein, the faceform who makes up coupling can realize for prior art, does not repeat them here.
After this, promptly can be according to the identify label of definition, add up the address link of the pairing user capture of this target people's face, the information such as key word in the information obtained are obtained record, so that after this user is connected to network, it is realized information automatic asking next time.
That is to say that each faceform who has made up described in the step 302 is meant each user who watches this TV, the faceform who when watching this TV for the first time, makes up.Certainly, among each faceform who has made up described in the step 302, also can be included in the user and watch before this TV the faceform who makes up in advance and store for the first time.
Step 303 according to the corresponding relation of faceform who sets in advance and identify label, determines and the pairing identify label of faceform of detected target people's face coupling that the current user who watches TV is promptly represented in this identify label.
Step 304 is searched corresponding information according to the identify label of determining and is obtained record, and the promptly current pairing information of user of TV of watching is obtained record.
In the practical application, information is obtained record and can be recorded in the TV, also can be recorded in the remote server.Therefore, in this step,, can in TV, search the information of identify label correspondence and obtain record, realize that promptly this locality searches for the previous case; And for latter event, then can in remote server, search the information of identify label correspondence and obtain record, promptly realize long-range searching.
Step 305 is obtained record with the information of this identify label correspondence of finding and is sent to remote server in the network, initiates asking for automatically of corresponding informance.
In this step, the initiation process of asking for automatically can be identical with existing mode, do not repeat them here.
In the practical application, alternatively, before initiating the asking for automatically of corresponding informance, can also pass through the central control system of TV inside earlier, know the program category of current broadcast in the TV, belong to same type, for example belong to finance and economic or sport category if wait the program of current broadcast in the corresponding informance of initiating to ask for automatically and the TV, represent that then the active user is watching its favorite program, can not initiate to ask for automatically and direct process ends; Belong to dissimilar if wait the program of current broadcast in the corresponding informance of initiating to ask for automatically and the TV, represent that then the program that the active user watches not is its inclination, thereby initiate asking for automatically of corresponding informance, and after treating to ask for corresponding informance automatically, the mode by roll titles presents.
So far, this flow process finishes.
After finishing above-mentioned flow process, the remote server in the network can obtain record according to the information that receives, and pushes corresponding information and presents to the user by TV.Wherein, for the situation of TV, also comprise user's TV program watching record in the requestor message record usually as main frame.
Fig. 4 is the structural representation of the control device of information automatic asking in the embodiment of the invention two.As shown in Figure 4, in the present embodiment control device of information automatic asking comprise with embodiment one in identical following components and parts and functional unit: sensitive component 200, and people's face detecting unit 201, Model Matching unit 202, identity determination unit 203, record search unit 204, telecommunication unit 205, modelling unit 206 in the processor.
Wherein, obtain the situation of recording storage in TV, need physical storage medium be set in TV inside for information.
Than embodiment one, alternatively, control device in the present embodiment can also comprise in processor: communication enables unit 407, and the output terminal of the central control system in its input end and record search unit 204 and the TV links to each other, output terminal links to each other with the input end of telecommunication unit 205.
Communication enables unit 407, be used for central control system by TV inside, know the program category of current broadcast in the TV, belong to same type if wait the program of current broadcast in the corresponding informance of initiating to ask for automatically and the TV, for example belong to finance and economic or sport category, represent that then the active user is watching its favorite program, forbid that telecommunication unit 205 obtains record with information and is sent to remote server in the network, promptly forbids initiating asking for automatically of corresponding informance; Belong to dissimilar if wait the program of current broadcast in the corresponding informance of initiating to ask for automatically and the TV, represent that then the program that the active user watches not is its inclination, thereby allow telecommunication unit 205 that information is obtained record to be sent to remote server in the network, promptly to allow to initiate asking for automatically of corresponding informance.
Like this, when the multi-user watches TV jointly, processor in the TV can go out to watch all user identity of TV according to the target recognition of face that sensitive component photographs, and initiate asking for automatically of corresponding informance according to the user identity that identifies, the information that will ask for is automatically presented to the user by the mode of roll titles then.
As seen from the above-described embodiment, the present invention utilizes sensitive component to take current user's facial image of watching TV, utilize identity then to this user of IP Address Recognition of people's face testing result of facial image rather than TV, and utilize the user identity that identifies to control asking for automatically of corresponding informance, thereby can discern the different user of watching same TV, also need not the user and manually import identity information.
Suppose, the cartoon that father and son watches the son that plays in the TV to like at home jointly, but father dynamically relatively is concerned about finance and economics.At this moment, processor detects the beholder and comprises father and these two users of son, and the central control system by coupled TV inside is known, the program category of current broadcast is cartoon, and write down the TV program watching record that comprises by the requestor message of searching father and son and judge, son is interested in the cartoon, father is interested in the finance and economics program, but not it is interested in the cartoon, therefore, processor in the TV can be asked for the information of some finance and economics to remote server, broadcasts some financial and economic news by roll titles, confession fathers such as quotations on the stock market browse.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to and replace and improvement etc., all should be included within protection scope of the present invention.