CN102750519A - Method and device for identifying and filtering images - Google Patents
Method and device for identifying and filtering images Download PDFInfo
- Publication number
- CN102750519A CN102750519A CN201210175229XA CN201210175229A CN102750519A CN 102750519 A CN102750519 A CN 102750519A CN 201210175229X A CN201210175229X A CN 201210175229XA CN 201210175229 A CN201210175229 A CN 201210175229A CN 102750519 A CN102750519 A CN 102750519A
- Authority
- CN
- China
- Prior art keywords
- image
- attitude
- user
- real
- meets pre
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention provides a method for identifying and filtering images. The method includes receiving an image provided by a user; determining whether postures of objects in the image conform to preset regulations; accepting the provided image if postures of objects in the image conform to preset regulations, otherwise, refusing to accept the image. Correspondingly, the invention further provides a device for identifying and filtering the images. The method and the device can improve the authenticity of the images authenticated by websites, and enhance the security of website information correspondingly.
Description
Technical field
The present invention relates to Flame Image Process, network application technology, relate in particular to the method and apparatus of a kind of image recognition, filtration.
Background technology
The use of image on the website is very general, especially various authentication websites as make friends, the love and marriage website, the user judges another user's characteristics such as appearance and makings through image such as half body according to, living photo, thus decision whether with another customer interaction information.If image information is untrue, can make the user that the experience of website is received very ill effect.
At present; Prior art is uploaded image according to the user usually and whether is met the certain size object images and examine roughly; And the source of object images is lacked examination; Thereby it is my true picture that the careless image of downloading or user obtain through shooting from the internet that the user be difficult to distinguish object images that the user uploads.If the user is any download images or utilize drawing tool to revise image arbitrarily from network, and with its authentication information as self, will make the authenticity that influences site information.
Summary of the invention
In order to solve above-mentioned prior art problems, at this identification and filter method and device that a kind of image is provided, be used to improve website institute authentication image authenticity, prevent that the user from will the online image of downloading arbitrarily pretending to be this deceptive practices of image of himself.
According to an aspect of the present invention, a kind of identification and filter method of image is provided, this method may further comprise the steps:
Receive the image that the user uploads;
Whether the attitude of judging the object in the said image meets pre-defined rule;
If the attitude of the object in the said image meets pre-defined rule, then accept the image that this is uploaded, otherwise refusal is accepted this image.
According to another aspect of the present invention, a kind of identification and filtration unit of image is provided also, comprised:
Receiver module is used to receive the image that the user uploads;
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule;
Whether respond module meets pre-defined rule according to the attitude of the object in the said image and makes response, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
Compared with prior art, the present invention has the following advantages:
1) improves the authenticity of website institute authentication image, prevent that the user from will the online image of downloading arbitrarily pretending to be this deceptive practices of image of himself, improved the reliability of the registered image information of website;
2) screen out the non-image in person that the user uploads, the guiding user uploads its real image;
3) the final user images that obtains of method of the present invention has unitarity, promotes user experience.
Description of drawings
Through reading the detailed description of doing with reference to following accompanying drawing that non-limiting example is done, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 is the method flow diagram of the identification and the filtration of image in accordance with a preferred embodiment of the present invention;
Fig. 2 is the synoptic diagram of image in accordance with a preferred embodiment of the present invention;
Fig. 3 a and 3b are the synoptic diagram of image according to a further advantageous embodiment of the invention;
The method flow diagram of Fig. 4 for carrying out image recognition and filtration in a kind of login system according to another preferred embodiment of the present invention; And
Fig. 5 a and 5b make the synoptic diagram that uses gesture and select for user in a kind of selective system according to a further advantageous embodiment of the invention.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
A kind of identification and filter method of image are provided according to an aspect of the present invention.It should be noted that the mentioned object of this paper refers generally to the personage, hereinafter will specify the method for identification provided by the present invention and filtering image.
Please refer to Fig. 1, Fig. 1 is the method flow diagram of the identification and the filtration of image in accordance with a preferred embodiment of the present invention.As shown in Figure 1, this method comprises:
Step S101 receives the image that the user uploads.
Particularly, the user uploads images onto on the website through webpage, supplies ownly to watch with other people or supply the website to put on record, as uploads images onto on the webpage as own QQ head portrait, MSN head portrait, uploads images onto the QQ space and makes photograph album and supply own and other people watch.Particularly existing marriage and making friend website, the user need check that other people image makes friends, therefore, on the website, uploading image becomes a kind of general phenomenon.These images can be the original images that the user takes, and also can be the images that user's modification is crossed, and upload as with photoshop, thin softhearted part etc. original image being made amendment again; Certainly; The image that some website requires the user to upload is the image of unmodified, and the image that some website requires the user to upload can pass through modification, but can only make amendment to background, the color of image; Can not upload through the image handled of splicing etc., promptly different web sites is uploaded the requirement difference of image to the user.The angle of these image takings also can be arbitrarily angled, from the height angle, is divided into the flat bat of clapping, take a crane shot, face upward; From taking direction, be divided into positive angle, side angle, back side angle etc.; Object from taking can be divided into personage, animal, plant, landscape etc.
Step S 102, judge whether the attitude of the object in the said image meets pre-defined rule.
Particularly; Alleged in the present invention " attitude " is not limited to posture; But has wider implication; For example, the attitude of reference object in image can comprise the form of expression of reference objects such as motion characteristic of direction, the object of angle, the shooting of jewelry that object is entrained or sign, shooting.
Wherein, jewelry that object is entrained or sign comprise any visual sign.During photographic images; Jewelry or sign can be put into various angles, be placed on the optional position that can carry out IMAQ with object simultaneously; As this jewelry of the calligraphy and painting that will write " smile " or sign are put in positions such as object front, upper left side, upper right side; During shooting, the jewelry or the sign of " smile " are taken with object, form image.
Wherein, the height angle of shooting comprises and flat claps, takes a crane shot, faces upward bat, during shooting, image collecting device is put into certain angle, as from flatly clap, face upward bat, all angles reference object such as take a crane shot, the image of formation different angles.
Wherein, the direction of shooting comprises positive angle, side angle etc.During shooting, image collecting device can be from all directions reference object, like the reference object from the front, the front of reference object or reverse side and surrounding; The reference object from the left side, the left side profile and the surrounding of reference object; The reference object from the right side, the right lateral contours of reference object and surrounding etc., thereby the image of formation different directions.
Wherein, the motion characteristic of object comprises expression that object does, gesture etc., is example with the gesture, and like forefinger and the middle finger that object reaches simultaneously, the pendulum forming V-shape also is placed on front, left side or right side etc.
The requirement of uploading image with the website is that the image that the user uploads is not through any example that is revised as; Before whether the attitude of the object in judging said image meets pre-defined rule; Whether the image that judges is uploaded is the image of revising arbitrarily through drawing tool; The method of these judgements can realize through method commonly used in the prior art, judges like the continuity of the pixel through check image whether image has the vestige of modification.
Through judging,, then judge automatically through system's (as accepting the client or the server of the image of uploading) whether the attitude of the object in the image meets predetermined rule if the image that the user uploads does not pass through any modification.Wherein, predetermined rule is meant at aspects such as the captured angle of object, direction, action, contents sets unified standard, and certainly, predetermined rule can be set unified standard to one of them aspect or many aspects.With the character image is example, and predetermined rule comprises:
Require the personage at image entrained jewelry or sign specialization, as require that perhaps
this jewelry of pattern or sign are put in personage's upper left side with writing " smile " literal;
Require the height angle of shooting specific, as the angle that requires to take is clapped for flat;
Require the direction of shooting specific, as the direction that requires to take is positive angle;
Require personage's motion characteristic specific, must finger put forming V-shape and be put in the front of object like the personage.
When requiring the user to upload image, can inform the above-mentioned rule of user through literal or illustrated mode.
In above-mentioned, image collecting device comprise in the prior art arbitrarily can images acquired equipment, like camera, camera etc.
Judge that method that whether image meets pre-defined rule can adopt existing Flame Image Process and image matching method to carry out machine and judge automatically; For example; Extract the image outline characteristic, whether the shape of user's gesture part can the judges gesture meet pre-defined rule in the recognition image profile.
Step S103 if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
Particularly, process is made and whether is accepted the response that this uploads image after judging whether the attitude of the object in the image meets predetermined rule.For example, the predetermined rule of taking pictures is that the motion characteristic of object is for putting forming V-shape with the forefinger of finger and middle finger and being put in the left side side of object.If the user puts forming V-shape with the forefinger or the middle finger of finger, and is placed on the left side side of object, then accept this image; The II shape (like the gesture among Fig. 2 2) that if the motion characteristic of object is fist shape (like the gesture among Fig. 2 1), thumb and forefinger to be put forming V-shape or forefinger and middle finger and becomes, these gestures all do not meet predetermined rule, therefore refuse to accept this image; If the user puts forming V-shape (like the gesture among Fig. 2 3) with the forefinger or the middle finger of finger, but be placed on the right side side of object, then same refusal is accepted this image, and can require to provide again image.
Preferably, upload the authenticity of image, can optimize as follows such scheme for judges more accurately:
1) opens image collecting device through client, gather user's image in real time;
Particularly, usually, portable terminals such as mobile phone, computer all carry the first-class image collecting device of shooting, and when user's request was registered on certain website or released news, client can be opened image collecting device automatically, gathers user's image in real time.Described real-time, refer to upload onto the server through the image synchronization of image collecting device collection.
2) receive the image that the user uploads, judge whether this image is the image of said real-time collection,, then further judge whether the attitude of the object in the said image meets pre-defined rule if this image is the image of said real-time collection; If this image is not the image of said real-time collection, then refusal is accepted this image.
Particularly, after the image uploading, whether be that the image of said real-time collection is judged to this image, if this image is the image of said real-time collection, then further judge whether the attitude of the object in the said image meets pre-defined rule; If this image is not the image of said real-time collection, be the image of shooting before the user transfers from the memory storage of portable terminal or the image that the user arbitrarily downloads from network like this image, then refusal is accepted this image.Judge whether that real-time images acquired can have several different methods, the sign and the shooting time of wherein a kind of method embedded images collecting device in the image of taking in real time can be judged what whether this image was taken in real time according to image capture device sign and shooting time.
Can filter out the image that a collection of non real-time is taken in advance through above-mentioned method for optimizing, increase the authenticity of image.
In another embodiment, the step of the image that above-mentioned reception user uploads also further comprises the steps:
1) opens the image collecting device that photo frame is provided through client, gather user's image and collection user's image is increased specific photo frame;
Particularly, image collecting device carries the specific photo frame of multiclass, when gathering user's image; The user can select a certain group of photo frame, is used for continuous shooting, also can select some photo frames; Carry out individual shooting, each of shooting image all is embedded in the photo frame of user's selection;
When 2) receiving the image that the user uploads; Client or server judge whether said image has the photo frame that said image collecting device provides; Provide if the photo frame in the said image is said image collecting device, then further judge whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
Particularly, the photo frame in the image that the user uploads possibly derive from specific image collecting device, also possibly be synthetic from the take up an official post photo frame that in the mind carries of network, also might draw and forms through repairing figure software etc.Therefore; Need in the image that judges uploads; Whether its photo frame derives from that specific image collecting device provided, and provides if the photo frame in the said image is said image collecting device, then further judges whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
Judge that the photo frame whether photo frame provides for the predetermined image harvester also can adopt existing Flame Image Process and image matching method to realize automatically.
Preferably, the method for the identification of image provided by the invention and filtration also comprises image is reappeared processing.
Particularly; Through clipping image, revise image and become unified format and show, usurp the unjustified image that other users upload and be used to prevent the user; Destroy the reliability that the present invention discerns and filters user images; The original image of other users being uploaded like the user is usurped as the image of oneself and is uploaded, and misleads other users, influences user's experience.
Reappear and handle, comprising:
1) attitude of the object in the manuscript picture.
Particularly, as indicated above, be example with the object images, the attitude of object comprises jewelry that object is entrained or sign, the angle of shooting, the direction of shooting, the motion characteristic of object in the image.Shown in Fig. 3 a and 3b, Fig. 3 a and 3b are the synoptic diagram of image according to a further advantageous embodiment of the invention.According to Fig. 3, establish the right side side that the predetermined rule request object of suppose object carries material object and is put in object, among Fig. 3 a; Object is put in the right side with fork, meets predetermined rule, then manuscript as the time; Extract the entrained in-kind portion of object in the image; And this part given amputation, the mode of amputation can be manual for user oneself, also can be automatic for equipment.Manually so-called, refer to system's (as accepting the client or the server of the image of uploading) and give the user certain modification authority, the part of appointment in the image is carried out amputation.Automatically so-called, the system of referring to automatically identifies the part that comprises predetermined attitude or gesture in the image, and this part is given amputation usurp this image to avoid other people.Automatically during amputation, client or server can be adjusted according to the size and the content of image automatically.
2) image of demonstration after processing.
Particularly, image as the gesture in the object images is partly deleted amputation, then shows the image after processing, shown in Fig. 3 b after processing.Show the content except that the gesture part.Certainly, no matter the deletion amputation is automatic amputation or manual amputation, the form of final images displayed is unified.For manual amputation, can give the user photo frame of a unified size, arbitrarily move on the image that the user can pass photo frame above that and can amplify this image, the parts of images in obtaining the photo frame that the user wants to keep, and the outer part of amputation photo frame.
Compared with prior art, the method for the identification of image provided by the present invention and filtration has the following advantages:
1) improves the authenticity of website institute authentication image, and then improved the Information Security of website;
2) image that finally appears has unitarity, has promoted user experience.
According to a further advantageous embodiment of the invention, the method for carrying out image recognition and filtration in a kind of login system is provided.With reference to Fig. 4, the method flow diagram of Fig. 4 for carrying out image recognition and filtration in a kind of login system according to another preferred embodiment of the present invention.Login system is common has a booting computer head portrait Verification System, and system is through image collecting device identification user's identity, thereby makes corresponding response, as getting into or stop the computer login system.According to Fig. 4, this method comprises:
Step S401 receives the image that the user uploads.
Like above-mentioned content, the user uploads images onto the website, and particular content is the same, for for simplicity, no longer details.
Step S402 judges whether the attitude of the object in the said image meets pre-defined rule.
Like above-mentioned content, after the user uploads images onto the website, whether the attitude of object in this image is met predetermined rule, judge, for for simplicity, no longer detail.
Step S403 if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
Like above-mentioned content, if being the attitude of the object in the said image, the result who judges meets pre-defined rule, then accept the image that this is uploaded, otherwise refusal is accepted this image.
Step S404, the image that storage is accepted to upload forms the IMAQ storehouse.
Particularly, the image that all acceptance are uploaded stores in the corresponding IMAQ storehouse of this user, certainly, in the corresponding IMAQ storehouse of this user, in the image to as if same.The image of why all acceptance being uploaded stores the IMAQ storehouse into; Be for the characteristic in the image that extracts the same target all angles; For use in doing comparison, judge whether the image in image to be certified and the IMAQ storehouse is same user with image to be certified.Certainly, also be fine if only upload an image.A plurality of users and corresponding a plurality of IMAQs storehouse can be arranged, the different images in the corresponding respectively IMAQ storehouse of perhaps a plurality of users.
Step S405 starts login system, opens the real-time images acquired of image collecting device, and the image of said real-time collection and the image in the said IMAQ storehouse are compared, and confirms the entering result of login system according to comparative result.
Particularly, when starting login system, login system is opened image collecting device automatically; Head portrait to the user is gathered in real time; After the collection, the image of said real-time collection and the image in the said IMAQ storehouse are compared, confirm the entering result of login system according to result relatively.If through judging that at least one image in the image of said real-time collection and the IMAQ storehouse is complementary, then confirm to get into the account of matching image respective user in the IMAQ storehouse; Otherwise termination login system.
According to a further advantageous embodiment of the invention, also provide, having comprised through image is discerned the method for carrying out selection operation:
1) receives the image that has attitude that the user uploads.
Whether the attitude of 2) judging the object in the said image meets pre-defined rule.
3) if the attitude of the object in the said image meets pre-defined rule, then accept the image that this is uploaded, otherwise refusal is accepted this image.
Above-mentioned three steps are same as above, for for simplicity, no longer detail.
4) be used for extracting the posture feature of institute's acceptance pattern picture object,, make corresponding selection according to the corresponding option of predetermined posture feature.
Particularly, attitude described here can be a gesture, defines the corresponding different option of different gesture in advance, like the gesture of frequent used expression 1-9 in the life, representes 1 with forefinger, representes 2 with forefinger and middle finger, representes 6 etc. with thumb and little finger of toe.If define the corresponding different options of different gestures in advance, then system can automatically carry out the selection of option through the gesture of recognition image, and as representing 1 with forefinger, then system automatically confirms option one.This image can be a dynamic image; Variation along with user's gesture; System can judge automatically selects different options, this for user distance equipment far can not the situation of direct control equipment under, can change the action that comes complete operation equipment through the gesture of taking the user.
With reference to Fig. 5 a and 5b, Fig. 5 a and 5b make the synoptic diagram that uses gesture and select for user in a kind of selective system according to a further advantageous embodiment of the invention.Suppose that two options are arranged in the selective system, then shown in Fig. 5 a, the user stretches out forefinger, representes that then the user selects wherein first option; Shown in Fig. 5 b, the user stretches out forefinger and middle finger, representes that then the user selects second option.Need to prove; In the selective system; When the user is known user's selection through image collecting device; Be not limited to the gesture of object in the image, can also comprise the entrained in kind or sign of object, the angle of taking, the direction of shooting, the various attitudes such as motion characteristic of object, and other can obtain the method that the user selects through image collecting device and all can use in the present invention.
Can find out that by preceding text the method for carrying out image recognition and filtration in the selective system that present embodiment provides can be so that to contactless operated from a distance of system equipment and control.
A kind of identification and filtration unit of image also are provided according to another aspect of the present invention.This device comprises:
Receiver module is used to receive the image that the user uploads;
Particularly, the user uploads images onto on the website through webpage, supplies ownly to watch with other people, as uploads images onto on the webpage as own QQ head portrait, MSN head portrait, uploads images onto the QQ space and makes photograph album and supply own and other people watch.Particularly existing marriage and making friend website, the user need check that other people image makes friends, therefore, on the website, uploading image becomes a kind of general phenomenon.These images can be the original images that the user takes, and also can be the images that user's modification is crossed, and upload as with photoshop software original image being made amendment again; The angle of these image takings can be arbitrarily angled, from the height angle, is divided into the flat bat of clapping, take a crane shot, face upward; From taking direction, be divided into positive angle, side angle, back side angle etc.; The object of these image takings can be personage, animal, plant, landscape etc.
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule;
Particularly, the preferred who object of embodiments of the invention is as the main contents of image, and the attitude of object in image can comprise jewelry that object is entrained or sign, the angle of shooting, the direction of shooting, the motion characteristic of object.
Wherein, jewelry that object is entrained or sign comprise any visual sign; During photographic images; Jewelry or sign can be put into various angles, be placed on the optional position that can carry out IMAQ with object simultaneously, are put in positions such as object front, upper left side, upper right side like this jewelry of the calligraphy and painting that will write " smile " or sign, during shooting; The jewelry or the sign of " smile " are taken with object, form image.
Wherein, the height angle of shooting comprises the flat bat of clapping, take a crane shot, face upward.During shooting, image collecting device is put into certain angle, as from flat clap, face upward bats, all angles reference object such as take a crane shot, the image of formation different angles.
Wherein, the direction of shooting comprises positive angle, side angle, back side angle etc.During shooting, image collecting device can be from all directions reference object, like the reference object from the front, the front of reference object or reverse side and surrounding; The reference object from the left side, the left side profile and the surrounding of reference object; The reference object from the right side, the right lateral contours of reference object and surrounding etc., thereby the image of formation different directions.
Wherein, the motion characteristic of object comprises expression that object does, gesture etc., is example with the gesture, and like forefinger and the middle finger that object reaches simultaneously, the pendulum forming V-shape also is placed on front, left side or right side etc.
The requirement of uploading image with the website is that the image that the user uploads is not through any example that is revised as; Before whether the attitude of the object in judging said image meets pre-defined rule; Whether the image that judges is uploaded is the image of revising arbitrarily through drawing tool; The method of these judgements can realize through method commonly used in the prior art, as judging through the continuity of check image pixel whether image has the vestige of modification.
Through judging,, then judge automatically through system's (as accepting the client or the server of the image of uploading) whether the attitude of the object in the image meets predetermined rule if the image that the user uploads does not pass through any modification.Wherein, predetermined rule is meant at aspects such as the captured angle of object, direction, action, contents sets unified standard, and certainly, predetermined rule can be set unified standard to one of them aspect or many aspects.With the object images is example, and predetermined rule comprises:
Require object at image entrained jewelry or sign specialization, as require that perhaps
this jewelry of image or sign are put in the upper left side of object with writing " smile " literal;
Require the height angle of shooting specific, as the angle that requires to take is clapped for flat;
Require the direction of shooting specific, as the direction that requires to take is positive angle;
Require the motion characteristic of object specific, must finger put forming V-shape and be put in the front of object like object.
In above-mentioned, image collecting device comprise in the prior art arbitrarily can images acquired equipment, like camera, camera etc.
Judge that method that whether image meets pre-defined rule can adopt existing Flame Image Process and image matching method to carry out machine and judge automatically; For example; Extract the image outline characteristic, whether the shape of user's gesture part can the judges gesture meet pre-defined rule in the recognition image profile.
Whether respond module meets pre-defined rule according to the attitude of the object in the said image and makes response, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
Particularly, process is made and whether is accepted the response that this uploads image after judging whether the attitude of the object in the image meets predetermined rule.For example, the predetermined rule of taking pictures is that the motion characteristic of object is for putting forming V-shape with the forefinger of finger and middle finger and being put in the left side side of object.If the user puts forming V-shape with the forefinger or the middle finger of finger, and is placed on the left side side of object, then accept this image; The I I shape (like the gesture among Fig. 2 2) that if the motion characteristic of object is fist shape (like the gesture among Fig. 2 1), thumb and forefinger to be put forming V-shape or forefinger and middle finger and becomes, these gestures all do not meet predetermined rule, therefore refuse to accept this image; If the user puts forming V-shape (like the gesture among Fig. 2 3) with the forefinger or the middle finger of finger, but be placed on the right side side of object, then same refusal is accepted this image, and can require to provide again image.
Preferably, upload the authenticity of image, can optimize as follows such scheme for judges more accurately:
1) opens image collecting device through client, gather user's image in real time;
Particularly, usually, portable terminals such as mobile phone, computer all carry the first-class image collecting device of shooting, and when user's request was registered on certain website or released news, client can be opened image collecting device automatically, gathers user's image in real time.Described real-time, refer to upload onto the server through the image synchronization of image collecting device collection.
2) receive the image that the user uploads, judge whether this image is the image of said real-time collection,, then further judge whether the attitude of the object in the said image meets pre-defined rule if this image is the image of said real-time collection; If this image is not the image of said real-time collection, then refusal is accepted this image.
Particularly, after the image uploading, whether be that the image of said real-time collection is judged to this image, if this image is the image of said real-time collection, then further judge whether the attitude of the object in the said image meets pre-defined rule; If this image is not the image of said real-time collection, be the image of shooting before the user transfers from the memory storage of portable terminal or the image that the user arbitrarily downloads from network like this image, then refusal is accepted this image.
Can filter out the image that a collection of non real-time is taken in advance through above-mentioned method for optimizing, increase the authenticity of image.
In another embodiment, above-mentioned receiver module is also carried out following steps:
1) opens the image collecting device that photo frame is provided through client, gather user's image;
Particularly, image collecting device carries the specific photo frame of multiclass, when gathering user's image; The user can select a certain group of photo frame, is used for continuous shooting, also can select some photo frames; Carry out individual shooting, each of shooting image all is embedded in the photo frame of user's selection;
When 2) receiving the image that the user uploads; Client or server judge whether said image has the photo frame that said image collecting device provides; Provide if the photo frame in the said image is said image collecting device, then further judge whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
Particularly, the photo frame in the image that the user uploads possibly derive from specific image collecting device, also possibly be synthetic from the take up an official post photo frame that in the mind carries of network, also might draw and forms through repairing figure software etc.Therefore; Need in the image that judges uploads; Whether its photo frame derives from that specific image collecting device provided, and provides if the photo frame in the said image is said image collecting device, then further judges whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
Judge that the photo frame whether photo frame provides for the predetermined image harvester also can adopt existing Flame Image Process and image matching method to realize automatically.
Preferably, the method for the identification of image provided by the invention and filtration also comprises image is reappeared processing.
Particularly; Through clipping image, revise image and become unified format and show, usurp the unjustified image that other users upload and be used to prevent the user; Destroy the reliability that the present invention discerns and filters user images; The original image of other users being uploaded like the user is usurped as the image of oneself and is uploaded, and misleads other users, influences user's experience.
This reappears module, comprising:
1) processing module is used for the attitude of the object of manuscript picture.
Particularly, as indicated above, be example with the object images, the attitude of object comprises jewelry that object is entrained or sign, the angle of shooting, the direction of shooting, the motion characteristic of object in the image.Shown in Fig. 3 a and 3b, Fig. 3 a and 3b are the synoptic diagram of image according to a further advantageous embodiment of the invention.According to Fig. 3, establish the right side side that the predetermined rule request object of suppose object carries material object and is put in object, among Fig. 3 a; Object is put in the right side with fork, meets predetermined rule, then manuscript as the time; Extract the entrained in-kind portion of object in the image; And this part given amputation, the mode of amputation can be manual for user oneself, also can be automatic for equipment.Manually so-called, refer to system's (as accepting the client or the server of the image of uploading) and give the user certain modification authority, the part of appointment in the image is carried out amputation.What is called refers to the particular pose or the gesture part of the automatic recognition image of system automatically, and this particular pose or gesture are partly given amputation, usurps to avoid other people.Automatically during amputation, client or server can be adjusted according to the size and the content of image automatically.
2) display module is used to show the image after processing.
Particularly, image as the gesture in the object images is partly deleted amputation, then shows the image after processing, shown in Fig. 3 b after processing.Show the content except that the gesture part.Certainly, no matter the deletion amputation is automatic amputation or manual amputation, the form of final images displayed is unified.For manual amputation, can give the user photo frame of a unified size, arbitrarily move on the image that the user can pass photo frame above that and can amplify this image, the parts of images in obtaining the photo frame that the user wants to keep, and the outer part of amputation photo frame.
The identification and the filtration unit of image provided by the present invention have the following advantages:
1) improves the authenticity of website institute authentication image, and then improved the Information Security of website;
2) image that finally appears has unitarity, has promoted user experience.
According to a further advantageous embodiment of the invention, the device that carries out image recognition and filtration in a kind of login system is provided.Login system is common has a booting computer head portrait Verification System, and system is through image collecting device identification user's identity, thereby makes corresponding response, as getting into or stop the computer login system.This device comprises:
Receiver module is used to receive the image that the user uploads.
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule.
Respond module is made corresponding judgment according to the content of said image, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
Particularly, more than the function again of three modules, for for simplicity, no longer detail.
The image memory module is used to store the image accepting to upload, forms the IMAQ storehouse.
Particularly, the image that all acceptance are uploaded stores in the corresponding IMAQ storehouse of this user, certainly, in the corresponding IMAQ storehouse of this user, in the image to as if same.The image of why all acceptance being uploaded stores the IMAQ storehouse into; Be for the characteristic in the image that extracts the same target all angles; For use in doing comparison, judge whether the image in image to be certified and the IMAQ storehouse is same user with image to be certified.Certainly, also be fine if only upload an image.Can there be the corresponding pictures different of a plurality of users to gather the storehouse, the different images in the corresponding respectively unified IMAQ of the perhaps a plurality of users storehouse.
Login module is used to start login system, opens the real-time images acquired of image collecting device, and the image of said real-time collection and the image in the said IMAQ storehouse are compared, and confirms the entering result of login system according to comparative result.
Particularly, when starting login system, login system is opened image collecting device automatically; Head portrait to the user is gathered in real time; After the collection, the image of said real-time collection and the image in the said IMAQ storehouse are compared, confirm the entering result of login system according to result relatively.If through judging that at least one image in the image of said real-time collection and the IMAQ storehouse is complementary, the user's that the image that then definite login gets into this coupling is corresponding account; Otherwise confirm to stop login system.
According to a further advantageous embodiment of the invention, the device that carries out image recognition and filtration in a kind of selective system is provided, has comprised:
Receiver module is used to receive the image that has gesture that the user uploads.
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule.
Respond module is made corresponding judgment according to the content of said image, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this is uploaded, otherwise refusal is accepted this image.
The function again of above-mentioned three modules for for simplicity, no longer details.
Select module, be used for extracting the posture feature of institute's acceptance pattern picture object,, make corresponding selection according to the corresponding option of predetermined posture feature.
Particularly, this posture feature is a gesture feature, and gesture is defined different options in advance, like the gesture of frequent used expression 1-9 in the life, representes 1 with forefinger, representes 2 with forefinger and middle finger, representes 6 etc. with thumb and little finger of toe.If the gesture content of object conforms to the gesture implication that system is scheduled in institute's acceptance pattern picture, then selective system is automatically carried out corresponding option operations, and as representing 1 with forefinger, then system automatically confirms option one.
With reference to Fig. 5 a and 5b, Fig. 5 a and 5b make the synoptic diagram that uses gesture and select for user in a kind of selective system according to a further advantageous embodiment of the invention.Suppose that two options are arranged in the selective system, then shown in Fig. 5 a, the user stretches out forefinger, representes that then the user selects wherein first option; Shown in Fig. 5 b, the user stretches out forefinger and middle finger, representes that then the user selects second option.Need to prove; In the selective system; When the user is known user's selection through image collecting device; Be not limited to the gesture of object in the image, can also comprise the entrained in kind or sign of object, the angle of taking, the direction of shooting, the various attitudes such as motion characteristic of object, and other can obtain the method that the user selects through image collecting device and all can use in the present invention.
The method of carrying out image recognition and filtration in the selective system that present embodiment provides can be so that the selection of system be more intelligent, more convenient.
Above disclosedly be merely a kind of preferred embodiment of the present invention, can not limit the present invention's interest field certainly with this, the equivalent variations of therefore doing according to claim of the present invention still belongs to the scope that the present invention is contained.
Claims (22)
1. the identification of an image and filter method comprise:
Receive the image that the user provides;
Whether the attitude of judging the object in the said image meets pre-defined rule;
If the attitude of the object in the said image meets pre-defined rule, then accept the image that this provides, otherwise refusal is accepted this image.
2. method according to claim 1, wherein, the step of the image that described reception user provides also comprises the steps:
Open image collecting device, gather user's image in real time;
Receive the image that the user provides, judge whether this image is the image of said real-time collection,, then further judge whether the attitude of the object in the said image meets pre-defined rule if this image is the image of said real-time collection;
If this image is not the image of said real-time collection, then refusal is accepted this image.
3. method according to claim 1, wherein, the step of the image that described reception user provides also comprises the steps:
Unlatching provides the image collecting device of photo frame, gathers user's image and collection user's image is increased specific photo frame;
When receiving the image that the user provides; Judge whether said image has the photo frame that said image collecting device provides; Provide if the photo frame in the said image is said image collecting device, then further judge whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
4. according to each described method of claim 1-3, wherein, the attitude of the object in the said image comprises one or more in the motion characteristic of direction, object of angle, the shooting of jewelry that object is entrained or sign, shooting at least.
5. method according to claim 4, wherein, the motion characteristic of described object comprises the gesture of object.
6. method according to claim 1 and 2 also comprises image is reappeared processed steps, and it comprises:
The attitude of the object in the manuscript picture;
Show the image after processing.
7. method according to claim 6, wherein, described processing steps comprises:
When object comprises gesture in the described image, the gesture of said image is partly discerned and amputation automatically.
8. the method for carrying out image recognition and filtration in the login system comprises:
Receive the image that the user provides;
Whether the attitude of judging the object in the said image meets pre-defined rule;
If the attitude of the object in the said image meets pre-defined rule, then accept the image that this provides, otherwise refusal is accepted this image;
The image that storage is accepted to provide forms the IMAQ storehouse;
Start login system, open the real-time images acquired of image collecting device, the image of said real-time collection and the image in the said IMAQ storehouse are compared, confirm the entering result of login system according to comparative result.
9. method according to claim 8, wherein, the image in the image of described real-time collection and the said IMAQ storehouse compares, and specifically comprises:
The image of judging said real-time collection whether with the IMAQ storehouse at least one image be complementary;
If the wherein image in the image of said real-time collection and the IMAQ storehouse is complementary, then confirm to get into the account of matching image respective user in the IMAQ storehouse;
Otherwise termination login system.
10. one kind through discerning the method for carrying out selection operation to image, comprising:
Receive the image that has attitude that the user provides;
Whether the attitude of judging the object in the said image meets pre-defined rule;
If the attitude of the object in the said image meets pre-defined rule, then accept the image that this provides, otherwise refusal is accepted this image;
Extract the posture feature of object in institute's acceptance pattern picture,, make corresponding selection according to the corresponding option of predetermined posture feature.
11. method according to claim 10, wherein, described posture feature is a gesture, according to the corresponding option of predetermined gesture, makes corresponding selection.
12. the identification of an image and filtration unit comprise:
Receiver module is used to receive the image that the user provides;
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule;
Whether respond module meets pre-defined rule according to the attitude of the object in the said image and makes response, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this provides, otherwise refusal is accepted this image.
13. device according to claim 12 wherein also comprises:
Image collecting device is used for gathering in real time user's image;
Said rule judgment module judges whether this image is the image of said real-time collection earlier, if this image is the image of said real-time collection, then further judges whether the attitude of the object in the said image meets pre-defined rule; If this image is not the image of said real-time collection, then refusal is accepted this image.
14. device according to claim 12 wherein also comprises:
The image collecting device of photo frame is provided, is used to gather user's image and collection user's image is increased specific photo frame;
Said rule judgment module judges whether said image has the photo frame that said image collecting device provides earlier; Provide if the photo frame in the said image is said image collecting device, then further judge whether the attitude of the object in the said image meets pre-defined rule; If do not have photo frame or the not said image collecting device of said photo frame to provide in the said image, then refusal is accepted this image.
15. according to each described device of claim 12-14, wherein, the attitude of the object in the said image comprises one or more in the motion characteristic of direction, object of angle, the shooting of jewelry that object is entrained or sign, shooting at least.
16. device according to claim 15, wherein, the motion characteristic of described object comprises the gesture of object.
17. according to claim 14 or 15 described devices, also comprise the reproduction module, specifically comprise:
Processing module is used for the attitude of the object of manuscript picture;
Display module is used to show the image after processing.
18. device according to claim 17, wherein, when object comprised gesture in the described image, said processing module partly carried out automatic amputation to the gesture of said image.
19. the device that carries out image recognition and filtration in the login system comprises:
Receiver module is used to receive the image that the user provides;
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule;
Respond module is made corresponding judgment according to the content of said image, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this provides, otherwise refusal is accepted this image;
The image memory module is used to store the image accepting to provide, forms the IMAQ storehouse;
Image collecting device is used for gathering in real time user's image;
Login module is used to start login system, opens the real-time images acquired of said image collecting device, and the image of said real-time collection and the image in the said IMAQ storehouse are compared, and confirms the entering result of login system according to comparative result.
20. device according to claim 19, wherein, said login module is also carried out following operation:
The image of judging said real-time collection whether with the IMAQ storehouse at least one image be complementary;
If the wherein image in the image of said real-time collection and the IMAQ storehouse is complementary, then confirm to get into the account of matching image respective user in the IMAQ storehouse;
Otherwise termination login system.
21. one kind through discerning the device that carries out selection operation to image, comprising:
Receiver module is used to receive the image that has attitude that the user provides;
The rule judgment module is used for judging whether the attitude of the object of said image meets pre-defined rule;
Respond module is made corresponding judgment according to the content of said image, if the attitude of the object in the said image meets pre-defined rule, then accepts the image that this provides, otherwise refusal is accepted this image;
Select module, be used for extracting the posture feature of institute's acceptance pattern picture object,, make corresponding selection according to the corresponding option of predetermined posture feature.
22. device according to claim 21, wherein, described posture feature is a gesture, according to the corresponding option of predetermined gesture, makes corresponding selection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210175229XA CN102750519A (en) | 2012-05-30 | 2012-05-30 | Method and device for identifying and filtering images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210175229XA CN102750519A (en) | 2012-05-30 | 2012-05-30 | Method and device for identifying and filtering images |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102750519A true CN102750519A (en) | 2012-10-24 |
Family
ID=47030685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210175229XA Pending CN102750519A (en) | 2012-05-30 | 2012-05-30 | Method and device for identifying and filtering images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102750519A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104156058A (en) * | 2013-05-14 | 2014-11-19 | 研祥智能科技股份有限公司 | Method and system for generating control orders |
| CN105898140A (en) * | 2016-03-30 | 2016-08-24 | 联想(北京)有限公司 | Information processing method and device |
| CN107292140A (en) * | 2016-04-02 | 2017-10-24 | 李辉 | A kind of social software and its attraction user, the method and device of the authentication image true and false |
| CN110781716A (en) * | 2018-07-31 | 2020-02-11 | 美凯利 | Information processing method and apparatus, computer-readable non-transitory storage medium |
| CN114005004A (en) * | 2021-12-30 | 2022-02-01 | 成都无糖信息技术有限公司 | Fraud website identification method and system based on picture instance level characteristics |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1444753A (en) * | 2000-05-26 | 2003-09-24 | 萨里大学 | Personal identity authentication process and system |
| US6823078B1 (en) * | 1999-10-26 | 2004-11-23 | Koninklijke Philips Electronics N.V. | Image processing method, system and apparatus for noise reduction in an image sequence representing a threadlike structure |
| CN101131609A (en) * | 2006-08-25 | 2008-02-27 | 株式会社东芝 | Interface apparatus and interface method |
| CN101174948A (en) * | 2006-11-02 | 2008-05-07 | 上海银晨智能识别科技有限公司 | Network login system and method with face authentication |
| CN102385703A (en) * | 2010-08-27 | 2012-03-21 | 北京中星微电子有限公司 | Identity authentication method and identity authentication system based on human face |
-
2012
- 2012-05-30 CN CN201210175229XA patent/CN102750519A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6823078B1 (en) * | 1999-10-26 | 2004-11-23 | Koninklijke Philips Electronics N.V. | Image processing method, system and apparatus for noise reduction in an image sequence representing a threadlike structure |
| CN1444753A (en) * | 2000-05-26 | 2003-09-24 | 萨里大学 | Personal identity authentication process and system |
| CN101131609A (en) * | 2006-08-25 | 2008-02-27 | 株式会社东芝 | Interface apparatus and interface method |
| CN101174948A (en) * | 2006-11-02 | 2008-05-07 | 上海银晨智能识别科技有限公司 | Network login system and method with face authentication |
| CN102385703A (en) * | 2010-08-27 | 2012-03-21 | 北京中星微电子有限公司 | Identity authentication method and identity authentication system based on human face |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104156058A (en) * | 2013-05-14 | 2014-11-19 | 研祥智能科技股份有限公司 | Method and system for generating control orders |
| CN104156058B (en) * | 2013-05-14 | 2018-08-07 | 研祥智能科技股份有限公司 | Generate the method and system of control instruction |
| CN105898140A (en) * | 2016-03-30 | 2016-08-24 | 联想(北京)有限公司 | Information processing method and device |
| CN105898140B (en) * | 2016-03-30 | 2019-11-26 | 联想(北京)有限公司 | A kind of information processing method and device |
| CN107292140A (en) * | 2016-04-02 | 2017-10-24 | 李辉 | A kind of social software and its attraction user, the method and device of the authentication image true and false |
| CN110781716A (en) * | 2018-07-31 | 2020-02-11 | 美凯利 | Information processing method and apparatus, computer-readable non-transitory storage medium |
| CN114005004A (en) * | 2021-12-30 | 2022-02-01 | 成都无糖信息技术有限公司 | Fraud website identification method and system based on picture instance level characteristics |
| CN114005004B (en) * | 2021-12-30 | 2022-04-01 | 成都无糖信息技术有限公司 | Fraud website identification method and system based on picture instance level characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11605389B1 (en) | User identification using voice characteristics | |
| US20150277686A1 (en) | Systems and Methods for the Real-Time Modification of Videos and Images Within a Social Network Format | |
| CN108108012B (en) | Information interaction method and device | |
| CN105095873A (en) | Picture sharing method and apparatus | |
| WO2017087568A1 (en) | A digital image capturing device system and method | |
| CN105144156A (en) | Associating metadata with images in a personal image collection | |
| JP2017531950A (en) | Method and apparatus for constructing a shooting template database and providing shooting recommendation information | |
| CN104239416A (en) | User identification method and system | |
| CN102750519A (en) | Method and device for identifying and filtering images | |
| EP3055793A1 (en) | Systems and methods for adding descriptive metadata to digital content | |
| EP2544126A1 (en) | Recognition system based on augmented reality and remote computing and related method | |
| US20180225505A1 (en) | Processing images from an electronic mirror | |
| CN104123485A (en) | Real-name system information binding method and device | |
| JP2014222439A (en) | Information processing apparatus, part generating and using method, and program | |
| US9538072B2 (en) | Gesture invoked image capture | |
| CN105654033A (en) | Face image verification method and device | |
| KR20170002485A (en) | Connecting current user activities with related stored media collections | |
| CN103823858A (en) | Information processing method and information processing device | |
| CN111488477A (en) | Album processing method, apparatus, electronic device and storage medium | |
| US11503110B2 (en) | Method for presenting schedule reminder information, terminal device, and cloud server | |
| CN105872252A (en) | Image processing method and device | |
| CN113852619A (en) | Interactive verification method and device and electronic equipment | |
| CN110086991A (en) | Image pickup method, device, portable terminal and the storage medium of terminal detection image | |
| CN105808231B (en) | System and method for recording and playing script | |
| CN107656959B (en) | Message leaving method and device and message leaving equipment |
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: 20121024 |