CN105989345A - Method and device for discovering friends by image matching - Google Patents

Method and device for discovering friends by image matching Download PDF

Info

Publication number
CN105989345A
CN105989345A CN201510090499.4A CN201510090499A CN105989345A CN 105989345 A CN105989345 A CN 105989345A CN 201510090499 A CN201510090499 A CN 201510090499A CN 105989345 A CN105989345 A CN 105989345A
Authority
CN
China
Prior art keywords
image
match
matched
type
matching
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201510090499.4A
Other languages
Chinese (zh)
Inventor
陈昌洲
韩涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201510090499.4A priority Critical patent/CN105989345A/en
Priority to PCT/CN2016/074650 priority patent/WO2016134666A1/en
Publication of CN105989345A publication Critical patent/CN105989345A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor

Abstract

The embodiment of the invention provides a method and a device for discovering friends by image matching. A client receives the matching type of a to-be-matched image input by a user, and obtains a matching parameter of the to-be-matched image by employing a matching algorithm corresponding to the matching type of the to-be-matched image; then, identifiers of the matching parameter, the matching type and the matching algorithm of the to-be-matched image are sent to a server; the server matches the identifiers of the matching parameter, the matching type and the matching algorithm of the to-be-matched image with images in an image library to find a target image matched with the to-be-matched image; and a matching result is returned to the client, wherein the matching result comprises information of the user to which the matched target image belongs, and the user to which the target image belongs is the friend. The method utilizes the traditional image resources of the user to carry out image matching to find the friend so as to bring convenience for the user to find the friend.

Description

Images match finds the method and apparatus of good friend
Technical field
The present embodiments relate to communication technology, particularly relate to a kind of images match find good friend method and Device.
Background technology
Along with the development of mobile terminal, social fast development, importance promotes day by day, how by closing The guiding of reason is that user sets up social networks, becomes the emphasis that all big enterprises fall over each other to excavate, and industry is well-known Application, all will find good friend's main-inlet separately as product, finds that the means also layer of good friend goes out The poorest.
In prior art, a kind of method be fill according to user account, the pet name, telephone number, work The various information such as office, hobby, Recent Activity are that user recommends potential good friend.Client After finding the good friend that user is potential, the account of good friend, the pet name etc. are presented in user interface, User is scanned for by account, the pet name etc., then adds good friend.Another way is scanning good friend The Quick Response Code generated, quickly obtains the information of good friend, and application is added to good friend.
Method of the prior art is single, very flexible.In the second way, user must get The Quick Response Code of friend just can add as a friend, very flexible, is not easy to user and gives good friend.
Summary of the invention
The embodiment of the present invention provides the method and apparatus that a kind of images match finds good friend, it is possible to flexible Discovery good friend.
First aspect present invention provides a kind of method that images match finds good friend, including:
Client receives image to be matched and the match-type of described image to be matched of user's input；
It is described that described client uses the matching algorithm corresponding with the match-type of described image to be matched to obtain The match parameter of image to be matched；
Described client is by the match parameter of described image to be matched, the match-type of described image to be matched It is sent to server, so that described server is according to described image to be matched with the mark of described matching algorithm Match parameter, the match-type of described image to be matched and described matching algorithm mark and image library in Image mate, find and the target image of described images match to be matched；
Described client receives the matching result that described server returns, and described matching result includes coupling The information of the user belonging to described target image arrived.
In conjunction with first aspect present invention, in the first possible implementation of first aspect present invention, When the match-type of described image to be matched is face match-type, described client uses and treats with described The matching algorithm corresponding to match-type of coupling image obtains the match parameter of described image to be matched, including:
Described client uses the matching algorithm corresponding with described face match-type, extracts described to be matched The face characteristic of image.
In conjunction with first aspect present invention, in the implementation that the second of first aspect present invention is possible, When the match-type of described image to be matched is Background matching type, described client uses and treats with described The matching algorithm corresponding to match-type of coupling image obtains the match parameter of described image to be matched, including:
Described client uses the algorithm corresponding with described Background matching type, extracts described image to be matched The color characteristic of background.
In conjunction with first aspect present invention, in the third possible implementation of first aspect present invention, When the match-type of described image to be matched is place match-type, described client uses and treats with described The matching algorithm corresponding to match-type of coupling image obtains the match parameter of described image to be matched, including:
Described client uses the algorithm corresponding with described place match-type, extracts described image to be matched Textural characteristics.
In conjunction with the third possible implementation of first aspect present invention, in the of first aspect present invention In four kinds of possible implementations, the match parameter of described image to be matched also includes described image to be matched Spot for photography information, the most described method also includes:
Described client obtains the bat of described image to be matched from the File header information of described image to be matched Take the photograph location information.
In conjunction with first aspect present invention, in the 5th kind of possible implementation of first aspect present invention, When the match-type of described image to be matched is author's match-type, described client uses and treats with described The matching algorithm corresponding to match-type of coupling image obtains the match parameter of described image to be matched, including:
Described client uses the algorithm corresponding with described author's match-type, from described image to be matched File header information obtains the information of the shooting author of described image to be matched.
Second aspect present invention provides a kind of method that images match finds good friend, including:
Server receive client send the match parameter of image to be matched, described image to be matched Join the mark of the type matching algorithm corresponding with the match-type of described image to be matched；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image；
Described server obtains the information of the user belonging to described target image, according to described target image institute The information of the user belonged to generates matching result, and described matching result is sent to described client.
In conjunction with second aspect present invention, in the first possible implementation of second aspect present invention, When the match-type of described image to be matched is face match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to face match-type；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The face characteristic of the image in described image library is extracted with the matching algorithm that described face match-type is corresponding；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the face of the face characteristic of described image to be matched with the image in described image library Feature is mated, using the image that the match is successful as described target image.
In conjunction with second aspect present invention, in the implementation that the second of second aspect present invention is possible, When the match-type of described image to be matched is Background matching type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to Background matching type；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The color of the background of image in described image library is extracted with matching algorithm corresponding to described Background matching type Feature；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the color characteristic of the background of described image to be matched and the image in described image library The color characteristic of background mate, using the image that the match is successful as described target image.
In conjunction with the implementation that second aspect present invention the second is possible, in the 3rd of second aspect present invention Planting in possible implementation, described server obtains the information of the user belonging to described target image, root Generate matching result according to the information of the user belonging to described target image, described matching result is sent to institute State client, including:
Described server obtains the information of the user belonging to described target image；
Described server judge in the image of the user belonging to described target image with described image to be matched Whether the image number joined is more than the threshold value pre-set；
If with the image number of described images match to be matched in the image of the user belonging to described target image More than described threshold value, the information of the user belonging to described target image is added to described by the most described server In matching result.
In conjunction with second aspect present invention, in the 4th kind of possible implementation of second aspect present invention, When the match-type of described image to be matched is place match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to place match-type；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted Textural characteristics with the image in the algorithm described image library of extraction corresponding with described place match-type；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that in described image library, the match is successful as described target image, including:
Described server is by the texture of the textural characteristics of described image to be matched with the image in described image library Feature is mated, using the image that the match is successful as described target image.
In conjunction with second aspect present invention, in the 5th kind of possible implementation of second aspect present invention, When the match-type of described image to be matched is place match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to place match-type；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The textural characteristics of image in described image library is extracted with algorithm corresponding to described place match-type, and from The spot for photography obtaining the image in described image library in the File header information of the image in described image library is believed Breath；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the textural characteristics in described image to be matched and spot for photography information, with described figure As textural characteristics and the spot for photography information of the image in storehouse are mated, using the image that the match is successful as Described target image.
In conjunction with second aspect present invention, in the 6th kind of possible implementation of second aspect present invention, When the match-type of described image to be matched is author's match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to author's match-type；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted With algorithm corresponding to described author's match-type, the File header information of the image from described image library obtains Take the information of the shooting author of image in described image library；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the information of the shooting author of described image to be matched and the image in described image library The information of shooting author mate, using the image that the match is successful as described target image.
Third aspect present invention provides a kind of client, including:
Receiver module, for receiving image to be matched and the coupling class of described image to be matched of user's input Type；
Acquisition module, for using the matching algorithm corresponding with the match-type of described image to be matched to obtain The match parameter of described image to be matched；
Sending module, for by the match parameter of described image to be matched, the coupling of described image to be matched The mark of type and described matching algorithm is sent to server, so that described server is according to described to be matched The match parameter of image, the match-type of described image to be matched and the mark of described matching algorithm and image Image in storehouse mates, and finds and the target image of described images match to be matched；
Described receiver module, is additionally operable to receive the matching result that described server returns, described matching result Include the information of the user belonging to described target image matched.
In conjunction with third aspect present invention, in the first possible implementation of third aspect present invention, When the match-type of described image to be matched is face match-type, described acquisition module specifically for: Use the matching algorithm corresponding with described face match-type, extract the face characteristic of described image to be matched.
In conjunction with third aspect present invention, in the implementation that the second of third aspect present invention is possible, When the match-type of described image to be matched is Background matching type, described acquisition module specifically for: Using the algorithm corresponding with described Background matching type, the color of the background extracting described image to be matched is special Levy.
In conjunction with third aspect present invention, in the third possible implementation of third aspect present invention, When the match-type of described image to be matched is place match-type, described acquisition module specifically for: Use the algorithm corresponding with described place match-type, extract the textural characteristics of described image to be matched.
In conjunction with the third possible implementation of third aspect present invention, in the of third aspect present invention In four kinds of possible implementations, the match parameter of described image to be matched also includes described image to be matched Spot for photography information, described acquisition module is additionally operable to: from the File header information of described image to be matched Obtain the spot for photography information of described image to be matched.
In conjunction with third aspect present invention, in the 5th kind of possible implementation of third aspect present invention, When the match-type of described image to be matched is author's match-type, described acquisition module specifically for: Use the algorithm corresponding with described author's match-type, obtain from the File header information of described image to be matched Take the information of the shooting author of described image to be matched.
Fourth aspect present invention provides a kind of server, including:
Receiver module, for receiving the match parameter, described to be matched of the image to be matched that client sends The mark of the matching algorithm that the match-type of image is corresponding with the match-type of described image to be matched；
Acquisition module, for the match-type according to described image to be matched and the mark of described matching algorithm, The match parameter of the image in acquisition image library；
Matching module, for by the match parameter of described image to be matched and the image in described image library Match parameter is mated, using the image that the match is successful as described target image；
Sending module, for obtaining the information of the user belonging to described target image, according to described target figure As the information of affiliated user generates matching result, and described matching result is sent to described client.
In conjunction with fourth aspect present invention, in the first possible implementation of fourth aspect present invention, When the match-type of described image to be matched is face match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to face match-type, described acquisition module specifically for: according to described The match-type of image to be matched and the mark of described matching algorithm, use described face match-type corresponding Matching algorithm extract the face characteristic of the image in described image library；
Described matching module specifically for: by the face characteristic of described image to be matched and described image library The face characteristic of image mate, using the image that the match is successful as described target image.
In conjunction with fourth aspect present invention, in the implementation that the second of fourth aspect present invention is possible, When the match-type of described image to be matched is Background matching type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to Background matching type, described acquisition module specifically for: according to described The match-type of image to be matched and the mark of described matching algorithm, use described Background matching type corresponding Matching algorithm extract the color characteristic of background of image in described image library；
Described matching module specifically for: by the color characteristic of the background of described image to be matched and described figure As the color characteristic of the background of the image in storehouse mates, using the image that the match is successful as described target Image.
In conjunction with the implementation that the second of fourth aspect present invention is possible, in the of fourth aspect present invention In three kinds of possible implementations, described sending module specifically for:
Obtain the information of user belonging to described target image；
Judge the image with described images match to be matched in the image of the user belonging to described target image Whether number is more than the threshold value pre-set；
If with the image number of described images match to be matched in the image of the user belonging to described target image More than described threshold value, then the information of the user belonging to described target image is added in described matching result.
In conjunction with fourth aspect present invention, in the 4th kind of possible implementation of fourth aspect present invention, When the match-type of described image to be matched is place match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to place match-type, described acquisition module specifically for: according to described The match-type of image to be matched and the mark of described matching algorithm, use and described place match-type pair The algorithm answered extracts the textural characteristics of the image in described image library；
Described matching module specifically for: by the textural characteristics of described image to be matched and described image library The textural characteristics of image mate, using the image that the match is successful as described target image.
In conjunction with fourth aspect present invention, in the 5th kind of possible implementation of fourth aspect present invention, When the match-type of described image to be matched is place match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to place match-type, described acquisition module specifically for: according to described The match-type of image to be matched and the mark of described matching algorithm, use described place match-type corresponding Algorithm extract the textural characteristics of the image in described image library, and the literary composition of the image from described image library Part header obtains the spot for photography information of image in described image library；
Described matching module specifically for: the textural characteristics in described image to be matched and spot for photography are believed Breath, mates with textural characteristics and the spot for photography information of the image in described image library, coupling is become The image of merit is as described target image.
In conjunction with fourth aspect present invention, in the 6th kind of possible implementation of fourth aspect present invention, When the match-type of described image to be matched is author's match-type, described matching algorithm be designated institute State the mark of matching algorithm corresponding to author's match-type, described acquisition module specifically for: according to described The match-type of image to be matched and the mark of described matching algorithm, use described author's match-type corresponding Algorithm, the File header information of the image from described image library obtains image in described image library The information of shooting author；
Described matching module specifically for: by information and the described figure of the shooting author of described image to be matched As the information of the shooting author of the image in storehouse is mated, using the image that the match is successful as described target Image.
The images match that the embodiment of the present invention provides finds the method and apparatus of good friend, and client is by receiving The match-type of image to be matched of user's input, uses corresponding with the match-type of image to be matched Join algorithm and obtain the match parameter of image to be matched, then, by match parameter, the coupling of image to be matched The mark of type and matching algorithm is sent to server, server according to the match parameter of image to be matched, The mark of match-type and matching algorithm is mated with the image in image library, finds and image to be matched The target image of coupling, and the matching result returned to client, matching result includes the mesh matched The information of the user belonging to logo image, the user belonging to target image is good friend.Said method is by profit Carry out images match with the existing image resource of user and find good friend, conveniently the user discover that good friend.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to reality Execute the required accompanying drawing used in example or description of the prior art to be briefly described, it should be apparent that under, Accompanying drawing during face describes is some embodiments of the present invention, for those of ordinary skill in the art, On the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 finds the flow chart of the method for good friend for the images match that the embodiment of the present invention one provides；
Fig. 2 finds the flow chart of the method for good friend for the images match that the embodiment of the present invention two provides；
The structural representation of a kind of client that Fig. 3 provides for the embodiment of the present invention three；
The structural representation of a kind of server that Fig. 4 provides for the embodiment of the present invention four；
The structural representation of a kind of terminal that Fig. 5 provides for the embodiment of the present invention five；
The structural representation of a kind of server that Fig. 6 provides for the embodiment of the present invention six.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with this Accompanying drawing in bright embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention, Obviously, described embodiment is a part of embodiment of the present invention rather than whole embodiments.Based on Embodiment in the present invention, those of ordinary skill in the art are obtained under not making creative work premise The every other embodiment obtained, broadly falls into the scope of protection of the invention.
Fig. 1 finds the flow chart of the method for good friend, this reality for the images match that the embodiment of the present invention one provides Executing the method for offer of example by client executing, this client is specifically as follows in terminal the application installed (Application), as it is shown in figure 1, the method for the present embodiment may comprise steps of:
Step 101, client receive image to be matched and the match-type of image to be matched of user's input.
User can select an image as to be matched from the local image library of the terminal at client place Image, or, start photographing unit shooting one photo as image to be matched.User can pass through Select button in user interface selects the match-type of image to be matched, the match-type of image to be matched Including: face match-type, Background matching type, place match-type and author's match-type etc..
Step 102, client use the matching algorithm corresponding with the match-type of image to be matched to obtain and treat The match parameter of coupling image.
In the present embodiment, every kind of the most corresponding different matching algorithm of match-type, client is according to be matched The match-type of image, selects the matching algorithm corresponding with match-type to obtain the coupling ginseng of image to be matched Number, below will be explained in detail:
When the match-type of image to be matched is face match-type, client uses mates class with face The matching algorithm that type is corresponding, extracts the face characteristic of image to be matched.The coupling that face match-type is corresponding Algorithm refers to extract the algorithm that face characteristic uses, and the method extracting face characteristic conventional includes: based on mould Plate coupling method, based on singular value features method, based on singular value features method, locality preserving projections Method, main constituent divide method, Elastic Matching method, eigenfaces (converting based on KL), ANN Network method, support vector machine method, based on integral image characteristic method with based on probabilistic model method etc..
When the match-type of image to be matched is Background matching type, client uses and Background matching class The algorithm that type is corresponding, extracts the color characteristic of the background of image to be matched.The background of image refers to image Mass-tone, the color characteristic of image refers to the integral color to image, is mixed to get one according to the algorithm specified Individual mass-tone value.Algorithm corresponding to Background matching type refers to extract the algorithm that the background characteristics of image is used, The conventional method extracting background characteristics includes: according to pixels color value cluster, color histogram take peak value etc..
When the match-type of image to be matched is place match-type, client uses and ground Point matching class The algorithm that type is corresponding, extracts the textural characteristics of image to be matched.Textural characteristics is that one does not relies on color Or the visual signature of homogeneity phenomenon in the reflection image of brightness.The algorithm that place match-type is corresponding refers to carry Taking the method that the textural characteristics of image is used, the method that texture feature extraction is conventional includes: auto-correlation letter Number, Fourier transformation, haul distance analysis, grey scale difference analysis and gray level co-occurrence matrixes analysis etc..
Alternatively, when the match-type of image to be matched is place match-type, image to be matched Join parameter and not only include the textural characteristics of image to be matched, also include the spot for photography information of image to be matched, Then client obtains the spot for photography information of image to be matched from the File header information of image to be matched.Figure In the File header information of picture, storage has the information of the spot for photography of image and the letter of the shooting author of image Breath, can also include in File header information that the brand and model etc. shooting the camera of people's time, shooting image is believed Breath.Such as, it is interchangeable file format [Amiga] (Exchangeable Image when the file format of image to be matched File, is called for short Exif) time, the data storage of Exif format-pattern is identical with jpeg format, Actually Exif form is exactly the information inserting image at jpeg format head, including light during shooting Circle, shutter, white balance, ISO, focal length, date-time, shooting condition and camera brand, model, The sound recorded when color coding, shooting and thumbnail etc..Briefly, Exif=JPEG+ shooting ginseng Number.
When the match-type of image to be matched is author's match-type, client uses mates class with author The algorithm that type is corresponding, obtains the shooting author's of image to be matched from the File header information of image to be matched Information.
Step 103, client by the match parameter of image to be matched, the match-type of image to be matched and The mark of matching algorithm is sent to server so that server according to the match parameter of image to be matched, treat The coupling match-type of image and the mark of matching algorithm are mated with the image in image library, find and The target image of images match to be matched.
Client after getting the match parameter of image to be matched, by the match parameter of image to be matched, The match-type of image to be matched and the mark of matching algorithm are sent to server.Client specifically can be led to Cross following message format and send the match parameter of image to be matched, the match-type of image to be matched and coupling The mark of algorithm: the mark of application protocol head+match-type+upload data type+match parameter+matching algorithm Knowing+End or Not, wherein, application protocol head includes but not limited to HTML (Hypertext Markup Language) (HyperText Transfer Protocol, be called for short HTTP), based on the HTML (Hypertext Markup Language) on SSL) Hyper Text Transfer Protocol over Secure Socket Layer, is called for short HTTPS) etc..On Pass data type for defining the type of the match parameter uploaded.
In the present embodiment, when the match-type of image to be matched is face match-type, image to be matched The face characteristic that match parameter is image to be matched.When the match-type of image to be matched is Background matching During type, the match parameter of image to be matched is the background characteristics of image to be matched.When image to be matched When match-type is place match-type, the match parameter of image to be matched is that the texture of image to be matched is special Levy, or, the match parameter of image to be matched is textural characteristics and the spot for photography information of image to be matched. When the match-type of image to be matched is author's match-type, the match parameter of image to be matched is for treating The information of the shooting author of figure picture.When the match-type of image to be matched is other types, to be matched The match parameter of image is entire image.
Server receives the match parameter of image to be matched, the match-type of image to be matched and matching algorithm Mark after, select and client phase according to the mark of the match-type of image to be matched and matching algorithm Same matching algorithm extracts the match parameter of the image in image library, then, by the coupling of image to be matched Parameter is mated with the match parameter of the image in image library, using the image that the match is successful as target figure Picture.
In the method for the present embodiment, by extracting the match parameter of image to be matched according to match-type, greatly In most cases, client has only to the match parameter of image to be matched is sent to server, by servicing Device carries out images match according to the match parameter of image to be matched, without entire image is sent to clothes Business device, owing to the match parameter of image to be matched is less than entire image, such that it is able to save the network taken Resource.
Step 104, client receive the matching result that server returns, and matching result includes matching The information of the user belonging to target image.
Server includes user's belonging to the target image that matches to the matching result that client returns Information, the user belonging to target image is good friend, and the information of the user belonging to target image includes: use The image etc. of the account at family, the pet name and user.When target image is multiple, server can generate user Information list, is sent to client by user profile list.After client receives matching result, with list Form or other forms be presented to user, user can by select add button the user of needs is added Add as a friend.
In the present embodiment, client, by receiving the match-type of the image to be matched of user's input, uses The match parameter of the matching algorithm acquisition to be matched image corresponding with the match-type of image to be matched, then, The mark of match parameter, match-type and the matching algorithm of image to be matched is sent to server, so that Server according to the mark of match parameter, match-type and the matching algorithm of image to be matched with in image library Image mate, find and the target image of images match to be matched, and to client return Joining result, matching result includes the information of the user belonging to target image matched, target image institute The user belonged to is good friend.Said method is looked for by utilizing the existing image resource of user to carry out images match To good friend, conveniently the user discover that good friend.
Fig. 2 finds the flow chart of the method for good friend, this reality for the images match that the embodiment of the present invention two provides Execute example to describe from the angle of server, as in figure 2 it is shown, the method that the present embodiment provides can include following Step:
Step 201, server receive the match parameter of image to be matched, the figure to be matched that client sends The mark of the matching algorithm that the match-type of picture is corresponding with the match-type of image to be matched.
Match-type according to image to be matched of step 202, server and the mark of matching algorithm, obtain The match parameter of the image in image library.
Coupling corresponding with the match-type of image to be matched for the match-type of image to be matched is calculated by client The mark of method is sent to server, is to make server use identical matching algorithm to extract in image library The match parameter of image.
When the match-type of image to be matched is face match-type, matching algorithm be designated face Joining the mark of matching algorithm corresponding to type, server is calculated according to match-type and the coupling of image to be matched The mark of method, the match parameter of the image in acquisition image library, particularly as follows: server is according to figure to be matched The match-type of picture and the mark of matching algorithm, use the matching algorithm that face match-type is corresponding to extract figure Face characteristic as the image in storehouse.
When the match-type of image to be matched is Background matching type, matching algorithm be designated background Joining the mark of matching algorithm corresponding to type, server is calculated according to match-type and the coupling of image to be matched The mark of method, the match parameter of the image in acquisition image library, particularly as follows: server is according to figure to be matched The match-type of picture and the mark of matching algorithm, the matching algorithm using Background matching type corresponding extracts figure Color characteristic as the background of the image in storehouse.
When the match-type of image to be matched is place match-type, matching algorithm be designated place Joining the mark of matching algorithm corresponding to type, server is calculated according to match-type and the coupling of image to be matched The mark of method, the match parameter of the image in acquisition image library, particularly as follows: server is according to figure to be matched The match-type of picture and the mark of matching algorithm, use the algorithm corresponding with place match-type to extract image The texture of the image in storehouse.Or, server is according to the match-type of image to be matched and matching algorithm Mark, uses the algorithm that place match-type is corresponding to extract the texture of the image in image library, and from image The spot for photography information of the image in acquisition image library in the File header information of the image in storehouse.
When the match-type of image to be matched is author's match-type, matching algorithm be designated author Joining the mark of matching algorithm corresponding to type, server is calculated according to match-type and the coupling of image to be matched The mark of method, the match parameter of the image in acquisition image library, particularly as follows: server is according to figure to be matched The match-type of picture and the mark of matching algorithm, use the algorithm that author's match-type is corresponding, from image library In image File header information in obtain the information of shooting author of image in image library.
Step 203, server mate ginseng by the match parameter of image to be matched and the image in image library Number mates, using the image that the match is successful as described target image.
When the match-type of image to be matched is face match-type, server is by the people of image to be matched Face feature is mated with the face characteristic of the image in image library, using the image that the match is successful as target Image.
When the match-type of image to be matched is Background matching type, server is by the back of the body of image to be matched The color characteristic of scape mates with the color characteristic of the background of the image in image library, by what the match is successful Image is as target image.
When the match-type of image to be matched is place match-type, server is by the stricture of vagina of image to be matched Reason feature is mated with the textural characteristics of the image in image library, using the image that the match is successful as target Image.Or, server is by the textural characteristics in image to be matched and spot for photography information, with image library In the textural characteristics of image and spot for photography information mate, using the image that the match is successful as target Image.
When the match-type of image to be matched is author's match-type, server is by the bat of image to be matched Take the photograph the information of author to mate, by what the match is successful with the information of the shooting author of the image in image library Image is as target image.
Step 204, server obtain the information of the user belonging to target image, belonging to target image The information of user generate matching result, and matching result is sent to client.
Server, after matching target image, obtains the information of user belonging to target image, when Join type when being face match-type, place match-type and author's match-type, by mesh between server The information of the user belonging to logo image is sent to client.When match-type is Background matching type, clothes After business device gets the information of the user belonging to target image, it is judged that the figure of the user belonging to target image In Xiang, whether the image number with images match to be matched is more than the threshold value pre-set, if target image institute In the image of the user belonged to, the image number with images match to be matched is more than this threshold value, then server is by mesh The information of the user belonging to logo image is added in matching result, say, that the user belonging to target image With when having multiple images and images match to be matched, server just using this user as good friend.Such as, should Threshold value is 10, then in the image that certain user uploads, and has the color characteristic of 10 images above and treats The color characteristic of coupling image differs within positive and negative 3, it is determined that this user is good friend, by this user's Information is added in matching result.According to this rule, all users meeting condition are found out.
In the present embodiment, server receives the match parameter of the image to be matched that client sends, coupling class The mark of the matching algorithm that type is corresponding with the match-type of image to be matched, according to the coupling of image to be matched Type and the mark of matching algorithm, the match parameter of the image in acquisition image library, by image to be matched Match parameter is mated with the match parameter of the image in image library, using the image that the match is successful as mesh Logo image, the finally information of acquisition user belonging to target image, according to the user's belonging to target image Information generates matching result, and matching result is sent to client.Said method is by utilizing user Some image resources carry out images match and find good friend, conveniently the user discover that good friend.
The structural representation of a kind of client that Fig. 3 provides for the embodiment of the present invention three, as it is shown on figure 3, The client that the present embodiment provides includes: receiver module 11, acquisition module 12 and sending module 13.
Wherein, receiver module 11, for receiving the image to be matched of user's input and described image to be matched Match-type；
Acquisition module 12, for using the matching algorithm corresponding with the match-type of described image to be matched to obtain Take the match parameter of described image to be matched；
Sending module 13, for by the match parameter of described image to be matched, described image to be matched The mark joining type and described matching algorithm is sent to server so that described server according to described in treat The match parameter of figure picture, the match-type of described image to be matched and the mark of described matching algorithm and figure As the image in storehouse mates, find and the target image of described images match to be matched；
Described receiver module 11, is additionally operable to receive the matching result that described server returns, and described coupling is tied Fruit includes the information of the user belonging to described target image matched.
When the match-type of described image to be matched is face match-type, described acquisition module 12 is concrete For: use the matching algorithm corresponding with described face match-type, extract the people of described image to be matched Face feature.
When the match-type of described image to be matched is Background matching type, described acquisition module 12 is concrete For: use the algorithm corresponding with described Background matching type, extract the background of described image to be matched Color characteristic.
When the match-type of described image to be matched is place match-type, described acquisition module 12 is concrete For: using the algorithm corresponding with described place match-type, the texture extracting described image to be matched is special Levy.Optionally, the match parameter of described image to be matched also includes the spot for photography of described image to be matched Information, described acquisition module 12 is additionally operable to: obtain described from the File header information of described image to be matched The spot for photography information of image to be matched.
When the match-type of described image to be matched is author's match-type, described acquisition module 12 is concrete For: use the algorithm corresponding with described author's match-type, believe from the file header of described image to be matched Breath obtains the information of the shooting author of described image to be matched.
The client of the present embodiment, the method that can be used for performing embodiment one, specific implementation and technology Effect is similar to, and repeats no more here.
The structural representation of a kind of server that Fig. 4 provides for the embodiment of the present invention four, as shown in Figure 4, The present embodiment provide server include: receiver module 21, acquisition module 22, matching module 23 and send out Send module 23.
Wherein, receiver module 21, for receiving the match parameter of image to be matched, the institute that client sends State the mark of the match-type of the image to be matched matching algorithm corresponding with the match-type of described image to be matched Know；
Acquisition module 22, for the match-type according to described image to be matched and the mark of described matching algorithm Know, the match parameter of the image in acquisition image library；
Matching module 23, for by the match parameter of described image to be matched and the image in described image library Match parameter mate, using the image that the match is successful as described target image；
Sending module 24, for obtaining the information of the user belonging to described target image, according to described target The information of the user belonging to image generates matching result, and described matching result is sent to described client.
When the match-type of described image to be matched is face match-type, the mark of described matching algorithm For the mark of matching algorithm corresponding to described face match-type, described acquisition module 22 is specifically for root According to match-type and the mark of described matching algorithm of described image to be matched, described face is used to mate class The matching algorithm that type is corresponding extracts the face characteristic of the image in described image library.Accordingly, described coupling Module 23 specifically for: by the people of the image in the face characteristic of described image to be matched and described image library Face feature is mated, using the image that the match is successful as described target image.
When the match-type of described image to be matched is Background matching type, the mark of described matching algorithm For the mark of matching algorithm corresponding to described Background matching type, described acquisition module 22 is specifically for root According to match-type and the mark of described matching algorithm of described image to be matched, use described Background matching class The matching algorithm that type is corresponding extracts the color characteristic of the background of the image in described image library.Accordingly, institute State matching module 23 specifically for: by the color characteristic of the background of described image to be matched and described image library In the color characteristic of background of image mate, using the image that the match is successful as described target image.
When the match-type of described image to be matched is Background matching type, optionally, described transmission mould Block 24 specifically for: obtain the information of user belonging to described target image；Judge described target image institute In the image of the user belonged to, whether the image number with described images match to be matched is more than the threshold pre-set Value；If with the image number of described images match to be matched in the image of the user belonging to described target image More than described threshold value, then the information of the user belonging to described target image is added in described matching result.
When the match-type of described image to be matched is place match-type, the mark of described matching algorithm For the mark of matching algorithm corresponding to described place match-type, described acquisition module 22 is specifically for root According to match-type and the mark of described matching algorithm of described image to be matched, use and described ground Point matching The textural characteristics of the image in the algorithm described image library of extraction that type is corresponding.Accordingly, described coupling mould Block 23 specifically for: by the texture of the image in the textural characteristics of described image to be matched and described image library Feature is mated, using the image that the match is successful as described target image.
When the match-type of described image to be matched is place match-type, the mark of described matching algorithm For the mark of matching algorithm corresponding to described place match-type, described acquisition module 22 is specifically for root According to match-type and the mark of described matching algorithm of described image to be matched, use described ground Point matching class The algorithm that type is corresponding extracts the textural characteristics of the image in described image library, and the figure from described image library The File header information of picture obtains the spot for photography information of image in described image library.Accordingly, described Matching module 23 specifically for: by the textural characteristics in described image to be matched and spot for photography information, with Textural characteristics and the spot for photography information of the image in described image library are mated, by the figure that the match is successful As described target image.
When the match-type of described image to be matched is author's match-type, the mark of described matching algorithm For the mark of matching algorithm corresponding to described author's match-type, described acquisition module 22 is specifically for root According to match-type and the mark of described matching algorithm of described image to be matched, described author is used to mate class The algorithm that type is corresponding, obtains in described image library in the File header information of the image from described image library The information of the shooting author of image.Accordingly, described matching module 23 specifically for: by described to be matched The information of the shooting author of image is mated with the information of the shooting author of the image in described image library, Using the image that the match is successful as described target image.
The server of the present embodiment, the method that can be used for performing embodiment two, specific implementation and technology Effect is similar to, and repeats no more here.
The structural representation of a kind of terminal that Fig. 5 provides for the embodiment of the present invention five, as it is shown in figure 5, this The terminal 300 that embodiment provides includes: processor 31, memorizer 32, communication interface 33 with communicate always Line 34, wherein, memorizer 32 and communication interface 33 are connected with processor 31 by communication bus 34, Memorizer 32 is used for storing computer instruction, communication connect 33 mouthfuls for communicating with other equipment, processor 31 for performing the computer instruction of storage in memorizer 32, to perform method as described below:
Receive image to be matched and the match-type of described image to be matched of user's input；
The matching algorithm corresponding with the match-type of described image to be matched is used to obtain described image to be matched Match parameter；
By the match parameter of described image to be matched, the match-type of described image to be matched and described coupling The mark of algorithm is sent to server so that described server according to the match parameter of described image to be matched, The match-type of described image to be matched and the mark of described matching algorithm and the image in image library are carried out Join, find and the target image of described images match to be matched；
Receiving the matching result that described server returns, described matching result includes the described mesh matched The information of the user belonging to logo image.
When the match-type of described image to be matched is face match-type, described employing is treated with described The matching algorithm that the match-type of figure picture is corresponding obtains the match parameter of described image to be matched, including: Use the matching algorithm corresponding with described face match-type, extract the face characteristic of described image to be matched.
When the match-type of described image to be matched is Background matching type, described employing is treated with described The matching algorithm that the match-type of figure picture is corresponding obtains the match parameter of described image to be matched, including: Using the algorithm corresponding with described Background matching type, the color of the background extracting described image to be matched is special Levy.
When the match-type of described image to be matched is place match-type, described employing is treated with described The matching algorithm that the match-type of figure picture is corresponding obtains the match parameter of described image to be matched, including: Use the algorithm corresponding with described place match-type, extract the textural characteristics of described image to be matched.Can Choosing, the match parameter of described image to be matched also includes the spot for photography information of described image to be matched, The most described processor 31 is additionally operable to: obtain described to be matched from the File header information of described image to be matched The spot for photography information of image.
When the match-type of described image to be matched is author's match-type, described employing is treated with described The matching algorithm that the match-type of figure picture is corresponding obtains the match parameter of described image to be matched, including: Use the algorithm corresponding with described author's match-type, obtain from the File header information of described image to be matched Take the information of the shooting author of described image to be matched.
The terminal of the present embodiment, the method that can be used for performing embodiment one, specific implementation and technology effect Fruit seemingly, repeats no more here.
The structural representation of a kind of server that Fig. 6 provides for the embodiment of the present invention six, as shown in Figure 6, The server 400 that the present embodiment provides includes: processor 41, memorizer 42, communication interface 43 and logical Letter bus 44, wherein, memorizer 42 and communication interface 43 are by communication bus 44 with processor 41 even Connecing, memorizer 42 is used for storing computer instruction, communication connect 43 mouthfuls for communicating with other equipment, place Reason device 41 is for performing the computer instruction of storage in memorizer 42, to perform method as described below:
Receive the match parameter of image to be matched, the match-type of described image to be matched that client sends The mark of the matching algorithm corresponding with the match-type of described image to be matched；
Match-type according to described image to be matched and the mark of described matching algorithm, obtain in image library The match parameter of image；
The match parameter of the image in the match parameter of described image to be matched and described image library is carried out Join, using the image that the match is successful as described target image；
Obtain the information of user belonging to described target image, according to the user's belonging to described target image Information generates matching result, and described matching result is sent to described client.
When the match-type of described image to be matched is face match-type, the mark of described matching algorithm Mark for matching algorithm corresponding to described face match-type.Described according to described image to be matched Join type and the mark of described matching algorithm, obtain the match parameter of image in image library, including: root According to match-type and the mark of described matching algorithm of described image to be matched, described face is used to mate class The matching algorithm that type is corresponding extracts the face characteristic of the image in described image library.Described by described to be matched The match parameter of image is mated, by what the match is successful with the match parameter of the image in described image library Image as described target image, including: by the face characteristic of described image to be matched and described image library In the face characteristic of image mate, using the image that the match is successful as described target image.
When the match-type of described image to be matched is Background matching type, the mark of described matching algorithm Mark for matching algorithm corresponding to described Background matching type.Described according to described image to be matched Join type and the mark of described matching algorithm, obtain the match parameter of image in image library, including: root According to match-type and the mark of described matching algorithm of described image to be matched, use described Background matching class The matching algorithm that type is corresponding extracts the color characteristic of the background of the image in described image library.Described by described The match parameter of image to be matched is mated with the match parameter of the image in described image library, will coupling Successfully image is as described target image, including: by the color characteristic of the background of described image to be matched Mate with the color characteristic of the background of the image in described image library, using the image that the match is successful as Described target image.
When the match-type of described image to be matched is Background matching type, described acquisition described target figure As the information of affiliated user, generate matching result according to the information of the user belonging to described target image, Described matching result is sent to described client, including: obtain user's belonging to described target image Information；Judge the image with described images match to be matched in the image of the user belonging to described target image Whether number is more than the threshold value pre-set；If with described in the image of the user belonging to described target image The image number of images match to be matched is more than described threshold value, then by the user's belonging to described target image Information is added in described matching result.
When the match-type of described image to be matched is place match-type, the mark of described matching algorithm Mark for matching algorithm corresponding to described place match-type.Described according to described image to be matched Join type and the mark of described matching algorithm, obtain the match parameter of image in image library, including: root According to match-type and the mark of described matching algorithm of described image to be matched, use and described ground Point matching The textural characteristics of the image in the algorithm described image library of extraction that type is corresponding.Described by described figure to be matched The match parameter of picture is mated with the match parameter of the image in described image library, by described image library The image that the match is successful as described target image, including: by the textural characteristics of described image to be matched Mate with the textural characteristics of the image in described image library, using the image that the match is successful as described mesh Logo image.
When the match-type of described image to be matched is place match-type, the mark of described matching algorithm Mark for matching algorithm corresponding to described place match-type.Described according to described image to be matched Join type and the mark of described matching algorithm, obtain the match parameter of image in image library, including: root According to match-type and the mark of described matching algorithm of described image to be matched, use described ground Point matching class The algorithm that type is corresponding extracts the textural characteristics of the image in described image library, and the figure from described image library The File header information of picture obtains the spot for photography information of image in described image library.Described treat described The match parameter of coupling image is mated with the match parameter of the image in described image library, coupling is become The image of merit as described target image, including: by the textural characteristics in described image to be matched and shooting Location information, mates with textural characteristics and the spot for photography information of the image in described image library, will The image that the match is successful is as described target image.
When the match-type of described image to be matched is author's match-type, the mark of described matching algorithm Mark for matching algorithm corresponding to described author's match-type.Described according to described image to be matched Join type and the mark of described matching algorithm, obtain the match parameter of image in image library, including: root According to match-type and the mark of described matching algorithm of described image to be matched, described author is used to mate class The algorithm that type is corresponding, obtains in described image library in the File header information of the image from described image library The information of the shooting author of image.By the match parameter of described image to be matched and the figure in described image library The match parameter of picture is mated, using the image that the match is successful as described target image, including: by institute The information of shooting author stating image to be matched is entered with the information of the shooting author of the image in described image library Row coupling, using the image that the match is successful as described target image.
The server of the present embodiment, the method that can be used for performing embodiment two, specific implementation and technology Effect is similar to, and repeats no more here.
One of ordinary skill in the art will appreciate that: realize all or part of step of above-mentioned each method embodiment Suddenly can be completed by the hardware that programmed instruction is relevant.Aforesaid program can be stored in a computer can Read in storage medium.This program upon execution, performs to include the step of above-mentioned each method embodiment；And Aforesaid storage medium includes: ROM, RAM, magnetic disc or CD etc. are various can store program code Medium.
Last it is noted that various embodiments above is only in order to illustrate technical scheme, rather than right It limits；Although the present invention being described in detail with reference to foregoing embodiments, this area common Skilled artisans appreciate that the technical scheme described in foregoing embodiments still can be modified by it, Or the most some or all of technical characteristic is carried out equivalent；And these amendments or replacement, and The essence not making appropriate technical solution departs from the scope of various embodiments of the present invention technical scheme.

Claims (26)

1. the method that an images match finds good friend, it is characterised in that including:
Client receives image to be matched and the match-type of described image to be matched of user's input；
It is described that described client uses the matching algorithm corresponding with the match-type of described image to be matched to obtain The match parameter of image to be matched；
Described client is by the match parameter of described image to be matched, the match-type of described image to be matched It is sent to server, so that described server is according to described image to be matched with the mark of described matching algorithm Match parameter, the match-type of described image to be matched and described matching algorithm mark and image library in Image mate, find and the target image of described images match to be matched；
Described client receives the matching result that described server returns, and described matching result includes coupling The information of the user belonging to described target image arrived.
Method the most according to claim 1, it is characterised in that when the coupling of described image to be matched When type is face match-type, described client uses corresponding with the match-type of described image to be matched Matching algorithm obtain described image to be matched match parameter, including:
Described client uses the matching algorithm corresponding with described face match-type, extracts described to be matched The face characteristic of image.
Method the most according to claim 1, it is characterised in that when the coupling of described image to be matched When type is Background matching type, described client uses corresponding with the match-type of described image to be matched Matching algorithm obtain described image to be matched match parameter, including:
Described client uses the algorithm corresponding with described Background matching type, extracts described image to be matched The color characteristic of background.
Method the most according to claim 1, it is characterised in that when the coupling of described image to be matched When type is place match-type, described client uses corresponding with the match-type of described image to be matched Matching algorithm obtain described image to be matched match parameter, including:
Described client uses the algorithm corresponding with described place match-type, extracts described image to be matched Textural characteristics.
Method the most according to claim 4, it is characterised in that the coupling ginseng of described image to be matched Number also includes the spot for photography information of described image to be matched, and the most described method also includes:
Described client obtains the bat of described image to be matched from the File header information of described image to be matched Take the photograph location information.
Method the most according to claim 1, it is characterised in that when the coupling of described image to be matched When type is author's match-type, described client uses corresponding with the match-type of described image to be matched Matching algorithm obtain described image to be matched match parameter, including:
Described client uses the algorithm corresponding with described author's match-type, from described image to be matched File header information obtains the information of the shooting author of described image to be matched.
7. the method that an images match finds good friend, it is characterised in that including:
Server receive client send the match parameter of image to be matched, described image to be matched Join the mark of the type matching algorithm corresponding with the match-type of described image to be matched；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image；
Described server obtains the information of the user belonging to described target image, according to described target image institute The information of the user belonged to generates matching result, and described matching result is sent to described client.
Method the most according to claim 7, it is characterised in that when the coupling of described image to be matched When type is face match-type, described matching algorithm be designated described face match-type corresponding Join the mark of algorithm；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The face characteristic of the image in described image library is extracted with the matching algorithm that described face match-type is corresponding；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the face of the face characteristic of described image to be matched with the image in described image library Feature is mated, using the image that the match is successful as described target image.
Method the most according to claim 7, it is characterised in that when the coupling of described image to be matched When type is Background matching type, described matching algorithm be designated described Background matching type corresponding Join the mark of algorithm；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The color of the background of image in described image library is extracted with matching algorithm corresponding to described Background matching type Feature；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the color characteristic of the background of described image to be matched and the image in described image library The color characteristic of background mate, using the image that the match is successful as described target image.
Method the most according to claim 9, it is characterised in that described server obtains described mesh The information of the user belonging to logo image, generates coupling knot according to the information of the user belonging to described target image Really, described matching result is sent to described client, including:
Described server obtains the information of the user belonging to described target image；
Described server judge in the image of the user belonging to described target image with described image to be matched Whether the image number joined is more than the threshold value pre-set；
If with the image number of described images match to be matched in the image of the user belonging to described target image More than described threshold value, the information of the user belonging to described target image is added to described by the most described server In matching result.
11. methods according to claim 7, it is characterised in that when described image to be matched Join type when being place match-type, described matching algorithm to be designated described place match-type corresponding The mark of matching algorithm；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted Textural characteristics with the image in the algorithm described image library of extraction corresponding with described place match-type；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that in described image library, the match is successful as described target image, including:
Described server is by the texture of the textural characteristics of described image to be matched with the image in described image library Feature is mated, using the image that the match is successful as described target image.
12. methods according to claim 7, it is characterised in that when described image to be matched Join type when being place match-type, described matching algorithm to be designated described place match-type corresponding The mark of matching algorithm；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted The textural characteristics of image in described image library is extracted with algorithm corresponding to described place match-type, and from The spot for photography obtaining the image in described image library in the File header information of the image in described image library is believed Breath；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the textural characteristics in described image to be matched and spot for photography information, with described figure As textural characteristics and the spot for photography information of the image in storehouse are mated, using the image that the match is successful as Described target image.
13. methods according to claim 7, it is characterised in that when described image to be matched Join type when being author's match-type, described matching algorithm to be designated described author's match-type corresponding The mark of matching algorithm；
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, obtains Take the match parameter of image in image library, including:
Described server, according to the match-type of described image to be matched and the mark of described matching algorithm, is adopted With algorithm corresponding to described author's match-type, the File header information of the image from described image library obtains Take the information of the shooting author of image in described image library；
Described server is by the match parameter of described image to be matched and mating of the image in described image library Parameter is mated, using the image that the match is successful as described target image, including:
Described server is by the information of the shooting author of described image to be matched and the image in described image library The information of shooting author mate, using the image that the match is successful as described target image.
14. 1 kinds of clients, it is characterised in that including:
Receiver module, for receiving image to be matched and the coupling class of described image to be matched of user's input Type；
Acquisition module, for using the matching algorithm corresponding with the match-type of described image to be matched to obtain The match parameter of described image to be matched；
Sending module, for by the match parameter of described image to be matched, the coupling of described image to be matched The mark of type and described matching algorithm is sent to server, so that described server is according to described to be matched The match parameter of image, the match-type of described image to be matched and the mark of described matching algorithm and image Image in storehouse mates, and finds and the target image of described images match to be matched；
Described receiver module, is additionally operable to receive the matching result that described server returns, described matching result Include the information of the user belonging to described target image matched.
15. clients according to claim 14, it is characterised in that when described image to be matched When match-type is face match-type, described acquisition module specifically for: use mate with described face The matching algorithm that type is corresponding, extracts the face characteristic of described image to be matched.
16. clients according to claim 14, it is characterised in that when described image to be matched When match-type is Background matching type, described acquisition module specifically for: use with described Background matching The algorithm that type is corresponding, extracts the color characteristic of the background of described image to be matched.
17. clients according to claim 14, it is characterised in that when described image to be matched When match-type is place match-type, described acquisition module specifically for: use with described ground Point matching The algorithm that type is corresponding, extracts the textural characteristics of described image to be matched.
18. clients according to claim 17, it is characterised in that described image to be matched Joining parameter and also include the spot for photography information of described image to be matched, described acquisition module is additionally operable to: from institute State the spot for photography information obtaining described image to be matched in the File header information of image to be matched.
19. clients according to claim 14, it is characterised in that when described image to be matched When match-type is author's match-type, described acquisition module specifically for: use mate with described author The algorithm that type is corresponding, obtains described image to be matched from the File header information of described image to be matched The information of shooting author.
20. 1 kinds of servers, it is characterised in that including:
Receiver module, for receiving the match parameter, described to be matched of the image to be matched that client sends The mark of the matching algorithm that the match-type of image is corresponding with the match-type of described image to be matched；
Acquisition module, for the match-type according to described image to be matched and the mark of described matching algorithm, The match parameter of the image in acquisition image library；
Matching module, for by the match parameter of described image to be matched and the image in described image library Match parameter is mated, using the image that the match is successful as described target image；
Sending module, for obtaining the information of the user belonging to described target image, according to described target figure As the information of affiliated user generates matching result, and described matching result is sent to described client.
21. servers according to claim 20, it is characterised in that when described image to be matched When match-type is face match-type, described matching algorithm to be designated described face match-type corresponding The mark of matching algorithm, described acquisition module specifically for: according to the coupling class of described image to be matched Type and the mark of described matching algorithm, use the matching algorithm that described face match-type is corresponding to extract described The face characteristic of the image in image library；
Described matching module specifically for: by the face characteristic of described image to be matched and described image library The face characteristic of image mate, using the image that the match is successful as described target image.
22. servers according to claim 20, it is characterised in that when described image to be matched When match-type is Background matching type, described matching algorithm to be designated described Background matching type corresponding The mark of matching algorithm, described acquisition module specifically for: according to the coupling class of described image to be matched Type and the mark of described matching algorithm, the matching algorithm using described Background matching type corresponding extracts described The color characteristic of the background of the image in image library；
Described matching module specifically for: by the color characteristic of the background of described image to be matched and described figure As the color characteristic of the background of the image in storehouse mates, using the image that the match is successful as described target Image.
23. servers according to claim 22, it is characterised in that described sending module is specifically used In:
Obtain the information of user belonging to described target image；
Judge the image with described images match to be matched in the image of the user belonging to described target image Whether number is more than the threshold value pre-set；
If with the image number of described images match to be matched in the image of the user belonging to described target image More than described threshold value, then the information of the user belonging to described target image is added in described matching result.
24. servers according to claim 20, it is characterised in that when described image to be matched When match-type is place match-type, described matching algorithm to be designated described place match-type corresponding The mark of matching algorithm, described acquisition module specifically for: according to the coupling class of described image to be matched Type and the mark of described matching algorithm, use the algorithm corresponding with described place match-type to extract described figure Textural characteristics as the image in storehouse；
Described matching module specifically for: by the textural characteristics of described image to be matched and described image library The textural characteristics of image mate, using the image that the match is successful as described target image.
25. servers according to claim 20, it is characterised in that when described image to be matched When match-type is place match-type, described matching algorithm to be designated described place match-type corresponding The mark of matching algorithm, described acquisition module specifically for: according to the coupling class of described image to be matched Type and the mark of described matching algorithm, use the algorithm that described place match-type is corresponding to extract described image The textural characteristics of the image in storehouse, and the File header information of the image from described image library obtain described The spot for photography information of the image in image library；
Described matching module specifically for: the textural characteristics in described image to be matched and spot for photography are believed Breath, mates with textural characteristics and the spot for photography information of the image in described image library, coupling is become The image of merit is as described target image.
26. servers according to claim 20, it is characterised in that when described image to be matched When match-type is author's match-type, described matching algorithm to be designated described author's match-type corresponding The mark of matching algorithm, described acquisition module specifically for: according to the coupling class of described image to be matched Type and the mark of described matching algorithm, use the algorithm that described author's match-type is corresponding, from described image The File header information of the image in storehouse obtains the information of the shooting author of image in described image library；
Described matching module specifically for: by information and the described figure of the shooting author of described image to be matched As the information of the shooting author of the image in storehouse is mated, using the image that the match is successful as described target Image.
CN201510090499.4A 2015-02-28 2015-02-28 Method and device for discovering friends by image matching Pending CN105989345A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201510090499.4A CN105989345A (en) 2015-02-28 2015-02-28 Method and device for discovering friends by image matching
PCT/CN2016/074650 WO2016134666A1 (en) 2015-02-28 2016-02-26 Method and device for discovering friend through image matching

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510090499.4A CN105989345A (en) 2015-02-28 2015-02-28 Method and device for discovering friends by image matching

Publications (1)

Publication Number Publication Date
CN105989345A true CN105989345A (en) 2016-10-05

Family

ID=56787879

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510090499.4A Pending CN105989345A (en) 2015-02-28 2015-02-28 Method and device for discovering friends by image matching

Country Status (2)

Country Link
CN (1) CN105989345A (en)
WO (1) WO2016134666A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106980688A (en) * 2017-03-31 2017-07-25 上海掌门科技有限公司 A kind of method, equipment and system for being used to provide friend-making object
CN108334522A (en) * 2017-01-20 2018-07-27 阿里巴巴集团控股有限公司 The method for determining customs's coding, and determine the method and system of type information
CN110139025A (en) * 2018-09-29 2019-08-16 广东小天才科技有限公司 A kind of social user's recommended method and wearable device based on the behavior of taking pictures
CN110955841A (en) * 2019-11-28 2020-04-03 维沃移动通信有限公司 Social relationship establishing method and electronic equipment
CN112163168A (en) * 2020-09-15 2021-01-01 济南雪景网络技术有限公司 Matching friend-making method based on computer vision
WO2023206576A1 (en) * 2022-04-29 2023-11-02 北京小米移动软件有限公司 Image processing method and apparatus

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111723767B (en) * 2020-06-29 2023-08-08 杭州海康威视数字技术股份有限公司 Image processing method, device and computer storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102831176A (en) * 2012-07-30 2012-12-19 东莞宇龙通信科技有限公司 Method and server for recommending friends
CN103514286A (en) * 2013-09-30 2014-01-15 鸿富锦精密工业（深圳）有限公司 Friend recommending system and method
US20140081960A1 (en) * 2006-09-22 2014-03-20 John Nicholas Gross Friend & Group Recommendations for Social Networks
CN104102656A (en) * 2013-04-09 2014-10-15 腾讯科技（深圳）有限公司 Friend recommending method, device, system and server
US20140365484A1 (en) * 2013-03-15 2014-12-11 Daniel Freeman Comprehensive user/event matching or recommendations based on awareness of entities, activities, interests, desires, location
CN104317932A (en) * 2014-10-31 2015-01-28 小米科技有限责任公司 Photo sharing method and device
CN104333506A (en) * 2014-10-24 2015-02-04 北京兴远永网络科技有限公司 Communication terminal contact method and device

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103327163B (en) * 2012-03-22 2015-11-25 宇龙计算机通信科技(深圳)有限公司 Search method and the communication terminal thereof of associated person information
US8843518B2 (en) * 2012-07-17 2014-09-23 Verizon Patent And Licensing Inc. Method and apparatus for establishing a connection with known individuals
CN103024118A (en) * 2012-11-23 2013-04-03 东莞宇龙通信科技有限公司 Contact searching method and communication terminal
CN103116751B (en) * 2013-01-24 2016-07-06 河海大学 A kind of Method of Automatic Recognition for Character of Lcecse Plate
CN103970771B (en) * 2013-01-29 2016-12-28 中国科学院计算技术研究所 The search method of a kind of human body and system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140081960A1 (en) * 2006-09-22 2014-03-20 John Nicholas Gross Friend & Group Recommendations for Social Networks
CN102831176A (en) * 2012-07-30 2012-12-19 东莞宇龙通信科技有限公司 Method and server for recommending friends
US20140365484A1 (en) * 2013-03-15 2014-12-11 Daniel Freeman Comprehensive user/event matching or recommendations based on awareness of entities, activities, interests, desires, location
CN104102656A (en) * 2013-04-09 2014-10-15 腾讯科技（深圳）有限公司 Friend recommending method, device, system and server
CN103514286A (en) * 2013-09-30 2014-01-15 鸿富锦精密工业（深圳）有限公司 Friend recommending system and method
CN104333506A (en) * 2014-10-24 2015-02-04 北京兴远永网络科技有限公司 Communication terminal contact method and device
CN104317932A (en) * 2014-10-31 2015-01-28 小米科技有限责任公司 Photo sharing method and device

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334522A (en) * 2017-01-20 2018-07-27 阿里巴巴集团控股有限公司 The method for determining customs's coding, and determine the method and system of type information
CN108334522B (en) * 2017-01-20 2021-12-14 阿里巴巴集团控股有限公司 Method for determining customs code, and method and system for determining type information
CN106980688A (en) * 2017-03-31 2017-07-25 上海掌门科技有限公司 A kind of method, equipment and system for being used to provide friend-making object
WO2018176954A1 (en) * 2017-03-31 2018-10-04 上海掌门科技有限公司 Method, device and system for providing friend-making objects
CN110139025A (en) * 2018-09-29 2019-08-16 广东小天才科技有限公司 A kind of social user's recommended method and wearable device based on the behavior of taking pictures
CN110955841A (en) * 2019-11-28 2020-04-03 维沃移动通信有限公司 Social relationship establishing method and electronic equipment
CN112163168A (en) * 2020-09-15 2021-01-01 济南雪景网络技术有限公司 Matching friend-making method based on computer vision
WO2023206576A1 (en) * 2022-04-29 2023-11-02 北京小米移动软件有限公司 Image processing method and apparatus

Also Published As

Publication number Publication date
WO2016134666A1 (en) 2016-09-01

Similar Documents

Publication Publication Date Title
CN105989345A (en) Method and device for discovering friends by image matching
CN107491701B (en) Automatically creating a group for sharing photos
RU2654168C2 (en) Photo sharing method and apparatus
AU2012295532B2 (en) Zero-click photo upload
CN105630977B (en) Application program recommended method, apparatus and system
CN112001274B (en) Crowd density determining method, device, storage medium and processor
CN105493501A (en) Virtual video camera
CN105760408A (en) Picture sharing method and apparatus and terminal device
CN107154890A (en) Method, the apparatus and system of contact person is added in social networks
CN105005599A (en) Photograph sharing method and mobile terminal
WO2014075413A1 (en) Method and device for determining terminal to be shared and system
CN106375193A (en) Remote group photographing method
CN111985281A (en) Image generation model generation method and device and image generation method and device
CN105516578B (en) Image processing method, device and electronic equipment
CN103401893B (en) A kind of method creating user communication group, device and equipment
WO2014082549A1 (en) Method, terminal, and system for multiple terminal file transmission
CN111414842B (en) Video comparison method and device, computer equipment and storage medium
CN106209584B (en) User information processing method, client and server Internet-based
US20180137661A1 (en) Picture processing method, sending method, processing apparatus and sending apparatus
CN105872857B (en) Video processing method and device and electronic equipment
CN103003837B (en) Image processing system, image processing terminal and image processing method
CN104202426B (en) Network account establishes the method and its network-termination device, cloud device of connection
CN111510746A (en) Media resource delivery method and device, storage medium and electronic device
CN206364929U (en) Image providing service system
KR20140061340A (en) Apparatus and method of managing game screenshot based on exif meta-data

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20161005