CN105069426B - Similar pictures judgment method and device - Google Patents
Similar pictures judgment method and device Download PDFInfo
- Publication number
- CN105069426B CN105069426B CN201510463518.3A CN201510463518A CN105069426B CN 105069426 B CN105069426 B CN 105069426B CN 201510463518 A CN201510463518 A CN 201510463518A CN 105069426 B CN105069426 B CN 105069426B
- Authority
- CN
- China
- Prior art keywords
- pictures
- face
- main body
- picture
- threshold value
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
Present disclose provides a kind of similar pictures judgment method and devices, belong to picture process field.Method includes:Two pictures for treating judgement carry out Feature Points Matching;When Feature Points Matching degree is more than the first predetermined threshold value, Face datection is carried out respectively to two pictures;If including face in two pictures, recognition of face is carried out to two pictures respectively, obtains the first face characteristic and the second face characteristic;First face characteristic and the second face characteristic are matched, when face characteristic matching degree is more than the second predetermined threshold value, determine that two pictures are similar pictures.The disclosure carries out Feature Points Matching by treating two pictures of judgement first, when Feature Points Matching degree is more than the first predetermined threshold value, in conjunction with Face datection and the process of recognition of face, it is whether similar to face included in picture further to be judged, substantially increase the accuracy and fineness of similar pictures judgement.
Description
Technical field
This disclosure relates to picture process field more particularly to a kind of similar pictures judgment method and device.
Background technology
With the fast development of information technology, the similar pictures search based on " scheme to search figure " is answered by people are increasingly wider
Used in search field.Being searched for similar pictures becomes a kind of new search trend, major network application platform, such as millet,
Baidu and Tencent are proposed similar pictures function of search.Similar pictures search based on " scheme to search figure " is the figure to be searched for
Piece is matched with existing picture in database, judges whether picture is similar, to obtain figure similar with the picture to be searched for
Piece.
The method judged at present similar pictures is Feature Points Matching to be carried out to two pictures, when the matching of two pictures
Degree is more than predetermined threshold value, it is determined that this two pictures is similar.
Above-mentioned similar pictures judgment method only considered the whole similitude of picture, and there is no in view of main body in picture
Difference, this is difficult to meet requirement of the people to similar pictures accuracy of identification.For example, two pictures are the same backgrounds, but it is main
Body is different people, this two pictures is after overmatching, it is more likely that is judged as similar pictures, leads to the standard of picture similitude
True property is relatively low.
Invention content
To overcome the problems in correlation technique, a kind of similar pictures judgment method of disclosure offer and device.
According to the first aspect of the embodiments of the present disclosure, a kind of similar pictures judgment method is provided, including:
Two pictures for treating judgement carry out Feature Points Matching;
When Feature Points Matching degree is more than the first predetermined threshold value, Face datection is carried out respectively to two pictures;
If including face in two pictures, recognition of face is carried out to two pictures respectively, obtains the
One face characteristic and the second face characteristic;
First face characteristic and the second face characteristic are matched, preset when face characteristic matching degree is more than second
When threshold value, determine that two pictures are similar pictures.
In the first possible realization method of first aspect, two pictures for treating judgement carry out Feature Points Matching packet
It includes:
Using scale invariant feature SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, treats judgement
Two pictures carry out Feature Points Matching.
In second of possible realization method of first aspect, recognition of face is carried out to two pictures respectively, is obtained
First face characteristic and the second face characteristic include:
The main body of taking pictures of two pictures is determined respectively;
Feature extraction is carried out to the main body of taking pictures of two pictures respectively, the first face characteristic is obtained and the second face is special
Sign.
In the third possible realization method of first aspect, determine that the main body of taking pictures of two pictures includes respectively:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes that at least two to be smaller than the 4th pre-
If when the face of threshold value, at least two face is determined as main body of taking pictures.
In the 4th kind of possible realization method of first aspect, for any pictures in two pictures, work as institute
When to state picture include the face that at least two are smaller than the 4th predetermined threshold value, at least two face is determined as taking pictures
After main body, the method further includes:
Judge whether the number taken pictures included by main body in two pictures is identical;
When number is identical, the step of recognition of face is carried out to two pictures respectively is executed.
In the 5th kind of possible realization method of first aspect, the method further includes:
For any pictures in two pictures, when the face in the picture is located at image edge region,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture is less than third predetermined threshold value,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture
When, it is main body of taking pictures to determine the face not.
In the 6th kind of possible realization method of first aspect, the method further includes:
When the time interval between the shooting time information of two pictures is less than preset duration, described two are determined
Picture is similar pictures.
According to the second aspect of the embodiment of the present disclosure, a kind of similar pictures judgment means are provided, including:
Feature Points Matching module, two pictures for treating judgement carry out Feature Points Matching;
Face detection module, for when Feature Points Matching degree is more than the first predetermined threshold value, distinguishing two pictures
Carry out Face datection;
Face recognition module, if for including face in two pictures, respectively to two pictures into
Row recognition of face obtains the first face characteristic and the second face characteristic;
Matching module, for being matched to first face characteristic and the second face characteristic;
Similar pictures determining module, for when face characteristic matching degree is more than the second predetermined threshold value, determining described two
Picture is similar pictures.
In the first possible realization method of second aspect, the Feature Points Matching module is used for special using Scale invariant
It levies SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, two pictures for treating judgement carry out characteristic point
Match.
In second of possible realization method of second aspect, the face recognition module for determining described two respectively
The main body of taking pictures of picture;Feature extraction is carried out to the main body of taking pictures of two pictures respectively, obtains the first face characteristic and the
Two face characteristics.
In the third possible realization method of second aspect, the face recognition module is used for:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes that at least two to be smaller than the 4th pre-
If when the face of threshold value, at least two face is determined as main body of taking pictures.
In the 4th kind of possible realization method of second aspect, described device further includes:
Whether number judgment module, the number for judging to take pictures included by main body in two pictures are identical;
The face recognition module, for when number is identical, executing and carrying out recognition of face to two pictures respectively
The step of.
In the 5th kind of possible realization method of second aspect, the face recognition module is used for:
For any pictures in two pictures, when the face in the picture is located at image edge region,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture is less than third predetermined threshold value,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture
When, it is main body of taking pictures to determine the face not.
In the 6th kind of possible realization method of second aspect, similar pictures determining module is additionally operable to work as two pictures
Shooting time information between time interval be less than preset duration when, determine two pictures be similar pictures.
According to the third aspect of the embodiment of the present disclosure, a kind of similar pictures judgment means are provided, including:
Processor;
Memory for storing the executable instruction of processor;
Wherein, the processor is configured as:
Two pictures for treating judgement carry out Feature Points Matching;
When Feature Points Matching degree is more than the first predetermined threshold value, Face datection is carried out respectively to two pictures;
If including face in two pictures, recognition of face is carried out to two pictures respectively, obtains the
One face characteristic and the second face characteristic;
First face characteristic and the second face characteristic are matched, preset when face characteristic matching degree is more than second
When threshold value, determine that two pictures are similar pictures.
The technical scheme provided by this disclosed embodiment can include the following benefits:
Two pictures by treating judgement first carry out Feature Points Matching, are more than the first default threshold in Feature Points Matching degree
It is whether similar to face included in picture to carry out further in conjunction with Face datection and the process of recognition of face when value
Judge, avoid because of situations such as overall background is similar and main body of actually taking pictures is different and caused by error in judgement, substantially increase
The accuracy and fineness that similar pictures judge.
It should be understood that above general description and following detailed description is only exemplary and explanatory, not
The disclosure can be limited.
Description of the drawings
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is a kind of flow chart of similar pictures judgment method shown according to an exemplary embodiment.
Fig. 2 is a kind of flow chart of similar pictures judgment method shown according to an exemplary embodiment.
Fig. 3 is a kind of similar pictures judgment means block diagram shown according to an exemplary embodiment.
Fig. 4 is a kind of block diagram of similar pictures judgment means 400 shown according to an exemplary embodiment.
Fig. 5 is a kind of block diagram of similar pictures judgment means 500 shown according to an exemplary embodiment.
Specific implementation mode
To keep the purpose, technical scheme and advantage of the disclosure clearer, below in conjunction with attached drawing to disclosure embodiment party
Formula is described in further detail.
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all implementations consistent with this disclosure.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the disclosure.
Fig. 1 is a kind of flow chart of similar pictures judgment method shown according to an exemplary embodiment, as shown in Figure 1,
Similar pictures judgment method is used in terminal, is included the following steps.
In a step 101, two pictures for treating judgement carry out Feature Points Matching.
In a step 102, when Feature Points Matching degree is more than the first predetermined threshold value, to two pictures respectively into pedestrian
Face detects.
In step 103, if including face in two pictures, face is carried out to two pictures respectively
Identification, obtains the first face characteristic and the second face characteristic.
At step 104, first face characteristic and the second face characteristic are matched, when face characteristic matching degree
When more than the second predetermined threshold value, determine that two pictures are similar pictures.
The method that the embodiment of the present disclosure provides, two pictures by treating judgement first carry out Feature Points Matching, in spy
When sign point matching degree is more than the first predetermined threshold value, in conjunction with Face datection and the process of recognition of face, to included in picture
Face it is whether similar further judged, avoid due to overall background is similar and main body of actually taking pictures is different and caused by sentence
Situations such as disconnected error, substantially increases the accuracy and fineness of similar pictures judgement.
In one embodiment that the disclosure provides, the two pictures progress Feature Points Matching for treating judgement includes:
Using scale invariant feature SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, treats judgement
Two pictures carry out Feature Points Matching.
In one embodiment that the disclosure provides, recognition of face is carried out to two pictures respectively, is obtained the first
Face feature and the second face characteristic include:
The main body of taking pictures of two pictures is determined respectively;
Feature extraction is carried out to the main body of taking pictures of two pictures respectively, the first face characteristic is obtained and the second face is special
Sign.
In one embodiment that the disclosure provides, determine that the main body of taking pictures of two pictures includes respectively:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes that at least two to be smaller than the 4th pre-
If when the face of threshold value, at least two face is determined as main body of taking pictures.
In one embodiment that the disclosure provides, for any pictures in two pictures, when the picture
When including the face that at least two are smaller than the 4th predetermined threshold value, by least two face be determined as taking pictures main body it
Afterwards, the method further includes:
Judge whether the number taken pictures included by main body in two pictures is identical;
When number is identical, the step of recognition of face is carried out to two pictures respectively is executed.
In one embodiment that the disclosure provides, the method further includes:
For any pictures in two pictures, when the face in the picture is located at image edge region,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture is less than third predetermined threshold value,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture
When, it is main body of taking pictures to determine the face not.
In one embodiment that the disclosure provides, the method further includes:
When the time interval between the shooting time information of two pictures is less than preset duration, described two are determined
Picture is similar pictures.
The alternative embodiment that any combination forms the disclosure may be used, herein no longer in above-mentioned all optional technical solutions
It repeats one by one.
Fig. 2 is a kind of flow chart of similar pictures judgment method shown according to an exemplary embodiment, as shown in Fig. 2,
Similar pictures judgment method is used in terminal, is included the following steps.
In step 201, two pictures for treating judgement carry out Feature Points Matching.
In the embodiments of the present disclosure, two pictures to be judged can be arbitrary in carrying out similar pictures identification process
Two pictures, the identification object of the similar pictures identification process can be in all pictures or certain several photograph album in a photograph album
All pictures be not specifically limited herein it is, of course, also possible to be the certain pictures specified by user.
The Feature Points Matching process of the embodiment of the present disclosure may include treating two pictures of judgement to carry out characteristic point respectively
Extraction, pairing processing is carried out to the characteristic point.It should be noted that two pictures for treating judgement carry out Feature Points Matching packet
It includes:SIFT (Scale Invariant Feature Transform, using scale invariant feature), SURF (Speeded Up
Robust Features accelerate robust feature), color characteristic or structural edge feature, treat judgement two pictures carry out it is special
Sign point matching.
In step 202, when Feature Points Matching degree is more than the first predetermined threshold value, to two pictures respectively into pedestrian
Face detects.
When carrying out Feature Points Matching, the Feature Points Matching degree between two pictures, the matching of this feature point can be got
Similarity degree of the degree for weighing two pictures, when Feature Points Matching degree is more than the first predetermined threshold value, it is believed that this two
Picture is likely to similar pictures, in order to improve the fineness of detection, it is also necessary to further execute step 203, and work as characteristic point
When matching degree is less than or equal to the first predetermined threshold value, it is believed that two pictures are not similar pictures, terminate matching process.
It should be noted that the first predetermined threshold value can be configured by server, if to improve the inspection of similar pictures
Survey fineness, can by the first predetermined threshold value be arranged it is larger, and if to detection fineness require it is relatively low, can also be by first
Predetermined threshold value is arranged smaller.
In the embodiments of the present disclosure, whether Face datection refers to being wrapped in detection picture per pictures in two pictures
Face is included, if including face in per pictures, obtains the face information in every pictures, which may include
Location information, size information and the posture information of face.
In the embodiments of the present disclosure, only after two pictures judged carry out Face datection respectively and to be detected to described
It is illustrated for face.
Further, in the case where not detecting face to two pictures to be judged, can directly by this two
Pictures are determined as similar pictures, however, in order to improve detection fineness, can also separately be arranged one and be more than the first predetermined threshold value
Threshold value, i.e. the 5th predetermined threshold value, judging characteristic point matching degree whether be more than the 5th predetermined threshold value, if Feature Points Matching degree is big
In the 5th predetermined threshold value, then two pictures are determined as similar pictures, if Feature Points Matching degree is less than the 5th predetermined threshold value,
Then two pictures are determined as not being similar pictures, in actual scene, any of the above-described kind of judgment mode, this public affairs may be used
Embodiment is opened to be not construed as limiting this.
In step 203, after carrying out Face datection respectively to two pictures, detect that two pictures wrap
When containing at least one face, the main body of taking pictures of two pictures is determined respectively.
It should be noted that generally when being taken pictures, picture may each comprise main body of taking pictures, especially for figure map
Piece, main body of taking pictures can be one or more people, therefore, in order to avoid the interference that non-main body of taking pictures judges similar pictures,
Need first to determine the main body of taking pictures of each picture.
In the embodiments of the present disclosure, determine that the process of the main body of taking pictures of each picture may include following at least one realization
Mode:
The first realization method, for any pictures in two pictures, when the face in the picture is located at
When picture intermediate region, the face is determined as main body of taking pictures.
In general, when being shot to personage, can be shot based on personage, therefore, the people as main body of taking pictures
Therefore face, which can be located at picture intermediate region, when it includes face to determine picture, can get the location information of face.Wherein,
Location information can be coordinate information of the face in picture.
During main body is taken pictures in determination, for each face in picture, picture middle area can be determined first
Domain, the picture intermediate region can be a specified ranges on the basis of picture midpoint, then, be believed according to the position of face
Breath, judges whether the face can be determined as the face if face is in picture intermediate region in picture intermediate region
Main body of taking pictures can ignore the face if face is not in picture intermediate region.
Second of realization method, for any pictures in two pictures, when the face in the picture is more than
When third predetermined threshold value, the face is determined as main body of taking pictures.
In general, when being shot to personage, can be shot based on personage, therefore, the people as main body of taking pictures
Face can occupy a part of area of picture, therefore, when it includes face to determine picture, can get the size information of face.
Wherein, size information can be face area information occupied in picture.
During main body is taken pictures in determination, for each face in picture, the area of picture can be determined first,
Also, the threshold value for the minimum area that one can tell face can be set according to the area of the picture, referred to as third presets threshold
Then value according to the size information of face, judges whether the face is more than third predetermined threshold value, if face is pre- more than third
If when threshold value, can the face be determined as main body of taking pictures, if face is less than third predetermined threshold value, the face can be ignored.
In the embodiments of the present disclosure, the first above-mentioned realization method and second of realization method be can be combined with to be clapped
The determination of license-master's body, for example, the face having detected that is both in the intermediate region of picture to be judged, the face is again default more than third
The face can be then determined as main body of taking pictures by threshold value.
The third realization method, for any pictures in two pictures, when the picture includes at least two
When a face for being smaller than the 4th predetermined threshold value, at least two face is determined as main body of taking pictures.
In general, when being shot to personage, is often shot as main body for multiple personages, therefore, work as bat
When the personage taken the photograph is at least two people, certain spacing is had as between the different faces for main body of taking pictures, therefore, is schemed determining
When piece includes at least two face, the spacing of each two face can be got according to the location information of each face.
Determine take pictures main body during, for each two face in picture, can determine first the length of picture with
Then according to the length and width of the picture, the 4th predetermined threshold value is arranged by server in width, prescribed threshold each two people
Then maximum spacing between face compares the spacing for obtaining each two face and the 4th predetermined threshold value, if the spacing is small
In the 4th predetermined threshold value, which can be determined as main body of taking pictures, if the spacing is more than the 4th predetermined threshold value, can ignored
The face.
It in the embodiments of the present disclosure, can also be by the first above-mentioned realization method, second of realization method and the third reality
Existing mode is two-by-two in conjunction with come the determination for the main body that take pictures.For example, in conjunction with the first above-mentioned realization method and the third realization side
Formula take pictures the determination of main body, and each face for having detected that at least two faces, and having detected that both was waiting for judgement figure
The spacing of the intermediate region of piece, each two face is less than the 4th predetermined threshold value again, can the face be determined as main body of taking pictures.
In conjunction with above-mentioned second of realization method and the third realization method take pictures the determination of main body, has detected that at least
Two faces, and the area for each face having detected that, both more than third predetermined threshold value, the spacing of each two face is
Less than the 4th predetermined threshold value, which can be determined as main body of taking pictures.
It is, of course, also possible in conjunction with the first above-mentioned realization method, second of realization method and the third realization method come into
The determination of capable main body of taking pictures, has detected that at least two faces, and each face having detected that is in picture to be judged
Between region, the area of each face having detected that is more than third predetermined threshold value, each two face to be smaller than the 4th default
The face can be determined as main body of taking pictures by threshold value.
When the embodiment of the present disclosure is applied in actual scene, for example, collective note based at least two faces into
Row shooting, if picture to be judged is detected as group picture, the face as main body of taking pictures in group picture is respectively positioned on picture
Intermediate region, and the size of each face will be greater than third predetermined threshold value, in addition, the spacing of each two face will be less than
4th predetermined threshold value.
It should be noted that generally when being taken pictures, the main body of taking pictures of picture may not be personage but landscape or
Include the personage of the personage and non-main body of taking pictures as main body of taking pictures, i.e. passerby in some pictures, therefore, in order to avoid non-
The interference that main body of taking pictures judges similar pictures needs while main body is taken pictures in the determination of each picture, to further discriminate between out non-
It takes pictures main body.
In the embodiments of the present disclosure, determine the personage of each picture be not take pictures main body process may include it is following at least
A kind of realization method:
The first realization method, for any pictures in two pictures, when the face in the picture is located at
When image edge region, it is main body of taking pictures to determine the face not.
In general, when being shot to landscape, it will not be shot based on personage, if having detected that picture includes
Face then may be located at image edge region as the face of non-main body of taking pictures, therefore, when it includes face to determine picture,
The location information of face can be got.Wherein, location information can be coordinate information of the face in picture.
During it is to take pictures main body to determine face not, for each face in picture, figure can be determined first
Piece intermediate region, which can be a specified range on the basis of picture midpoint, then, according to face
Location information, judges whether the face is in picture intermediate region, if face is in image edge region, it may be determined that the people
Face is not main body of taking pictures.At this time, it may be determined that two pictures to be detected are similar pictures.
Second of realization method, for any pictures in two pictures, when the face in the picture is less than
When third predetermined threshold value, it is main body of taking pictures to determine the face not.
In general, when being shot to landscape, it will not be shot based on personage, if having detected that picture includes
Face then can still occupy a part of area of picture as the face of non-main body of taking pictures, and therefore, determine that picture includes face
When, the size information of face can be got.Wherein, size information can be face area information occupied in picture.
Determine face be it is non-take pictures main body during, for each face in picture, can determine figure first
The area of piece, also, the threshold value for the minimum area that one can tell face can be set according to the area of the picture, referred to as
Then three predetermined threshold values according to the size information of face, judge whether the face is less than third predetermined threshold value, if face is small
When third predetermined threshold value, it may be determined that the face is not main body of taking pictures.At this time, it may be determined that two pictures to be detected are
Similar pictures.
In the embodiments of the present disclosure, the first above-mentioned realization method and second of realization method be can be combined with to be clapped
The determination of license-master's body has detected that picture includes face, and both in the fringe region of picture, the size of the face is less than the face again
Third predetermined threshold value, then it is main body of taking pictures that can determine the face not.At this time, it may be determined that two pictures to be detected are phase
Like picture.
The third realization method, for any pictures in two pictures, when the face in the picture is not inconsistent
When closing default human face posture, it is main body of taking pictures to determine the face not.
In general, when being shot to landscape, it will not be shot based on personage, if having detected that picture includes
Then certain angle may be presented as the human face posture of non-main body of taking pictures in face with picture level face, therefore, scheme determining
When piece includes face, the posture information of face can be got.Wherein, posture information can be that human face posture and the picture are flat
The angle that face is presented.
During it is to take pictures main body to determine face not, for each face in picture, figure can be determined first
The horizontal position of piece, also, set one according to the horizontal position of the picture and remain to tell people in maximum angle with the position
The angular interval of face is then default human face posture according to the posture information of face, judges whether the face meets default face
Posture, if the posture information of face does not meet default human face posture, it may be determined that the face is not main body of taking pictures.At this point,
It can determine that two pictures to be detected are similar pictures.
It in the embodiments of the present disclosure, can also be by the first above-mentioned realization method, second of realization method and the third reality
Existing mode is not two-by-two in conjunction with being the determination of main body of taking pictures the face that is had detected that.
It is not main body of taking pictures in conjunction with the face that the first above-mentioned realization method and the third realization method are had detected that
Determination, detected picture include face, both in the fringe region of picture, the posture information of the face is not met the face again
Default human face posture, then it is main body of taking pictures that can determine the face not.At this time, it may be determined that two pictures to be detected are phase
Like picture.
It is not main body of taking pictures in conjunction with the face that above-mentioned second of realization method and the third realization method are had detected that
Determination, detected that picture includes face, the size of the face is both less than third predetermined threshold value, and the posture information of the face is again
Default human face posture is not met, then it is main body of taking pictures that can determine the face not.At this time, it may be determined that two figures to be detected
Piece is similar pictures.
It is, of course, also possible in conjunction with the first above-mentioned realization method, second of realization method and the third realization method come into
The face that has detected that of row is not the determination of main body of taking pictures, and has detected that picture includes face, the face is in the marginal zone of picture
Domain, the size of the face is less than third predetermined threshold value, and the posture information of the face does not meet default human face posture, then can be with
It is main body of taking pictures to determine the face not.At this time, it may be determined that two pictures to be detected are similar pictures.
In step 204, feature extraction is carried out to the main body of taking pictures of two pictures respectively, obtains the first face characteristic
With the second face characteristic.
Since the main body of taking pictures of picture may include that a face or multiple faces distinguish both of these case
It introduces:
The first situation, for each pictures in two pictures, when only being examined in the main body of taking pictures of the picture
When measuring a face, feature extraction is carried out to the main body of taking pictures of two pictures respectively, obtains the first face characteristic and the
Two face characteristics.
In the main body of taking pictures of every pictures when one-man's face, then face included in main body of taking pictures can be based on
Feature extraction is carried out, further to differentiate that two pictures are personage's picture of same person.
The second situation, for any pictures in two pictures, when in the main body of taking pictures of the picture at least
When detecting two faces, judge whether the number taken pictures included by main body in two pictures is identical, if number is identical,
Feature extraction then is carried out to the main body of taking pictures of two pictures respectively, obtains the first face characteristic and the second face characteristic, such as
Fruit number is different, it is determined that two pictures are not similar pictures.
If include multiple faces in the main body of taking pictures of two pictures, it can determine that two pictures are collective's figures
Piece just will be considered that two pictures are for the same collective only when the face in collective's picture is identical and number is identical
Collective's picture, therefore, it is also desirable to be compared to the number included by main body of taking pictures.
It should be noted that this is taken pictures, the number included by main body can be obtained according to during Face datection, and be passed through
Determination to main body of taking pictures is adjusted the number, for example, summarizing carrying out Face datection process, is examined to picture A
It surveys, knows in picture A to include 8 faces, but by the determination to main body of taking pictures, know that wherein 5 faces are to take pictures
Main body, 3 are non-main bodys of taking pictures, it is thus possible to which it is 5 to obtain the main body face number of taking pictures after being adjusted.
In this second situation, the first face characteristic and the second face characteristic extracted each means multiple faces
Face characteristic.
Above-mentioned steps 202~204 are after carrying out Face datection to two pictures respectively, to carry out feature extraction, to
Obtain the process of the first face characteristic and the second face characteristic.In the embodiments of the present disclosure, it is only referred to the first face characteristic logical
It crosses obtained as a result, the second face characteristic is referred to by two pictures to a progress feature extraction in two pictures
Another Zhang Jinhang feature extractions it is obtained as a result, and above-mentioned first face characteristic and the second face characteristic may each be to figure
Multiple faces carry out obtained by feature extraction in piece.
In step 205, first face characteristic and the second face characteristic are matched, when face characteristic matching degree
When more than the second predetermined threshold value, determine that two pictures are similar pictures.
When carrying out face characteristic matching, the face characteristic matching degree between two groups of face characteristics, the people can be got
Face characteristic matching degree is used to weigh the similarity degree of face in two pictures, when face characteristic matching degree is more than the second predetermined threshold value
When, it is believed that two pictures include identical face, which is likely to similar pictures, and works as face characteristic
When matching degree is less than or equal to the second predetermined threshold value, it is believed that two pictures do not include identical face, and two pictures are not
It is similar pictures, terminates matching process.For example, two pictures to be detected are to be judged as group picture, then, in order to judge
Whether two group pictures are similar, then need to carrying out fisrt feature face as two groups of faces of main body of taking pictures in two group pictures
With the extraction of second feature face, if face characteristic matching degree is more than the second predetermined threshold value, then it is assumed that two group pictures are
Similar pictures, otherwise it is assumed that two group pictures are dissimilar pictures.
It should be noted that the second predetermined threshold value can be configured by server, if to improve the inspection of similar pictures
Survey fineness, can by the second predetermined threshold value be arranged it is larger, and if to detection fineness require it is relatively low, can also be by second
Predetermined threshold value is arranged smaller.
In the above-mentioned steps 201-205 of the embodiment of the present disclosure, it is combined with picture similarity degree and passes through recognition of face
Judge whether two pictures are similar pictures come the face similarity degree determined, and due in shooting process, camera terminal
Can also records photographing temporal information, and hence it is also possible in conjunction with picture shooting time information carry out similar pictures judgement.Example
Such as, under general same background, when different people takes pictures, the regular hour is required for switch over, if picture to be judged or
It is only differed several seconds between the shooting time information for being the picture that Feature Points Matching degree is more than the first predetermined threshold value, such as continuous shooting scene
Deng then can directly determining that two pictures are similar pictures, without carrying out step 202-205 again.It that is to say, the disclosure is real
Example is applied after performing step 201, following steps can also be performed:When between the shooting time information of two pictures
When time interval is less than preset duration, determine that two pictures are similar pictures.
The method that the embodiment of the present disclosure provides, two pictures by treating judgement first carry out Feature Points Matching, in spy
When sign point matching degree is more than the first predetermined threshold value, in conjunction with Face datection and the process of recognition of face, to included in picture
Face it is whether similar further judged, avoid due to overall background is similar and main body of actually taking pictures is different and caused by sentence
Situations such as disconnected error, substantially increases the accuracy and fineness of similar pictures judgement.
Fig. 3 is a kind of similar pictures judgment means block diagram shown according to an exemplary embodiment.With reference to Fig. 3, the device
Including Feature Points Matching module 301, face detection module 302, face recognition module 303, matching module 304 and similar pictures are true
Cover half block 305.
Feature Points Matching module 301, two pictures for treating judgement carry out Feature Points Matching;
Face detection module 302, for when Feature Points Matching degree is more than the first predetermined threshold value, dividing two pictures
Face datection is not carried out;
Face recognition module 303, if for including face in two pictures, respectively to two pictures
Recognition of face is carried out, the first face characteristic and the second face characteristic are obtained;
Matching module 304, for being matched to first face characteristic and the second face characteristic;
Similar pictures determining module 305, for when face characteristic matching degree is more than the second predetermined threshold value, determining described two
Pictures are similar pictures.
In one embodiment of the disclosure, the Feature Points Matching module 304301 is used to use scale invariant feature
SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, two pictures for treating judgement carry out characteristic point
Match.
In one embodiment of the disclosure, the face recognition module 303 for determining two pictures respectively
It takes pictures main body;Feature extraction is carried out to the main body of taking pictures of two pictures respectively, obtains the first face characteristic and the second face
Feature.
In one embodiment of the disclosure, the face recognition module 303 is used for:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value,
The face is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes that at least two to be smaller than the 4th pre-
If when the face of threshold value, at least two face is determined as main body of taking pictures.
In one embodiment of the disclosure, described device further includes:
Whether number judgment module, the number for judging to take pictures included by main body in two pictures are identical;
The face recognition module 303, for when number is identical, executing and carrying out face knowledge to two pictures respectively
Other step.
In one embodiment of the disclosure, the face recognition module 303 is used for:
For any pictures in two pictures, when the face in the picture is located at image edge region,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture is less than third predetermined threshold value,
It is main body of taking pictures to determine the face not;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture
When, it is main body of taking pictures to determine the face not.
In one embodiment of the disclosure, similar pictures determining module 305 is additionally operable to the shooting when two pictures
When time interval between temporal information is less than preset duration, determine that two pictures are similar pictures.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Fig. 4 is a kind of block diagram of similar pictures judgment means 400 shown according to an exemplary embodiment.For example, device
400 can be mobile phone, computer, digital broadcast terminal, messaging devices, game console, tablet device, and medical treatment is set
It is standby, body-building equipment, personal digital assistant etc..
With reference to Fig. 4, device 400 may include following one or more components:Processing component 402, memory 404, power supply
Component 406, multimedia component 408, audio component 410, the interface 412 of input/output (I/O), sensor module 414, and
Communication component 416.
The integrated operation of 402 usual control device 400 of processing component, such as with display, call, data communication, phase
Machine operates and record operates associated operation.Processing element 402 may include that one or more processors 420 refer to execute
It enables, to perform all or part of the steps of the methods described above.In addition, processing component 402 may include one or more modules, just
Interaction between processing component 402 and other assemblies.For example, processing component 402 may include multi-media module, it is more to facilitate
Interaction between media component 408 and processing component 402.
Memory 404 is configured as storing various types of data to support the operation in device 400.These data are shown
Example includes instruction for any application program or method that are operated on device 400, contact data, and telephone book data disappears
Breath, picture, video etc..Memory 404 can be by any kind of volatibility or non-volatile memory device or their group
It closes and realizes, such as static RAM (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable to compile
Journey read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash
Device, disk or CD.
Electric power assembly 406 provides electric power for the various assemblies of device 400.Electric power assembly 406 may include power management system
System, one or more power supplys and other generated with for device 400, management and the associated component of distribution electric power.
Multimedia component 408 is included in the screen of one output interface of offer between described device 400 and user.One
In a little embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touch sensings
Device is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding action
Boundary, but also detect duration and pressure associated with the touch or slide operation.In some embodiments, more matchmakers
Body component 408 includes a front camera and/or rear camera.When device 400 is in operation mode, such as screening-mode or
When video mode, front camera and/or rear camera can receive external multi-medium data.Each front camera and
Rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 410 is configured as output and/or input audio signal.For example, audio component 410 includes a Mike
Wind (MIC), when device 400 is in operation mode, when such as call model, logging mode and speech recognition mode, microphone by with
It is set to reception external audio signal.The received audio signal can be further stored in memory 404 or via communication set
Part 416 is sent.In some embodiments, audio component 410 further includes a loud speaker, is used for exports audio signal.
I/O interfaces 412 provide interface between processing component 402 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include but be not limited to:Home button, volume button, start button and lock
Determine button.
Sensor module 414 includes one or more sensors, and the state for providing various aspects for device 400 is commented
Estimate.For example, sensor module 414 can detect the state that opens/closes of device 400, and the relative positioning of component, for example, it is described
Component is the display and keypad of device 400, and sensor module 414 can be with 400 1 components of detection device 400 or device
Position change, the existence or non-existence that user contacts with device 400,400 orientation of device or acceleration/deceleration and device 400
Temperature change.Sensor module 414 may include proximity sensor, be configured to detect without any physical contact
Presence of nearby objects.Sensor module 414 can also include optical sensor, such as CMOS or ccd image sensor, at
As being used in application.In some embodiments, which can also include acceleration transducer, gyro sensors
Device, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 416 is configured to facilitate the communication of wired or wireless way between device 400 and other equipment.Device
400 can access the wireless network based on communication standard, such as WiFi, 2G or 3G or combination thereof.In an exemplary implementation
In example, communication component 416 receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.
In one exemplary embodiment, the communication component 416 further includes near-field communication (NFC) module, to promote short range communication.Example
Such as, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology,
Bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 400 can be believed by one or more application application-specific integrated circuit (ASIC), number
Number processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing above-mentioned similar pictures judgement side
Method.
In the exemplary embodiment, it includes the non-transitorycomputer readable storage medium instructed, example to additionally provide a kind of
Such as include the memory 404 of instruction, above-metioned instruction can be executed by the processor 420 of device 400 to complete the above method.For example,
The non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk
With optical data storage devices etc..
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium is additionally provided, when the storage is situated between
When instruction in matter is executed by the processor of mobile terminal so that mobile terminal is able to carry out a kind of similar pictures judgment method.
Fig. 5 is a kind of block diagram of similar pictures judgment means 500 shown according to an exemplary embodiment.For example, device
500 may be provided as a server.With reference to Fig. 5, device 500 includes processing component 522, further comprises one or more
Processor, and by the memory resource representated by memory 532, for store can by the instruction of the execution of processing component 522,
Such as application program.The application program stored in memory 532 may include it is one or more each correspond to one
The module of group instruction.In addition, processing component 522 is configured as executing instruction, to execute above-mentioned similar pictures judgment method.
Device 500 can also include the power management that a power supply module 526 is configured as executive device 500, and one has
Line or radio network interface 550 are configured as device 500 being connected to network and input and output (I/O) interface 558.Dress
Setting 500 can operate based on the operating system for being stored in memory 532, such as Windows ServerTM, Mac OS XTM,
UnixTM,LinuxTM, FreeBSDTMOr it is similar.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Person's adaptive change follows the general principles of this disclosure and includes the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by following
Claim is pointed out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by the accompanying claims.
Claims (13)
1. a kind of similar pictures judgment method, which is characterized in that the method includes:
Two pictures for treating judgement carry out Feature Points Matching;
When Feature Points Matching degree is more than the first predetermined threshold value, Face datection is carried out respectively to two pictures;
If including face in two pictures, recognition of face is carried out to two pictures respectively, is obtained the first
Face feature and the second face characteristic, including:The main body of taking pictures of two pictures is determined respectively;Respectively to two pictures
Main body of taking pictures carries out feature extraction, obtains first face characteristic and second face characteristic;
First face characteristic and the second face characteristic are matched, when face characteristic matching degree is more than the second predetermined threshold value
When, determine that two pictures are similar pictures.
2. according to the method described in claim 1, it is characterized in that, two pictures for treating judgement carry out Feature Points Matching packet
It includes:
Using scale invariant feature SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, treats the two of judgement
Pictures carry out Feature Points Matching.
3. according to the method described in claim 1, it is characterized in that, determining that the main body of taking pictures of two pictures includes respectively:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region, by institute
It states face and is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value, by institute
It states face and is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes at least two being smaller than the 4th default threshold
When the face of value, at least two face is determined as main body of taking pictures.
4. according to the method described in claim 3, it is characterized in that, for any pictures in two pictures, work as institute
When to state picture include the face that at least two are smaller than the 4th predetermined threshold value, at least two face is determined as taking pictures
After main body, the method further includes:
Judge whether the number taken pictures included by main body in two pictures is identical;
When number is identical, the step of recognition of face is carried out to two pictures respectively is executed.
5. according to the method described in claim 1, it is characterized in that, the method further includes:
For any pictures in two pictures, when the face in the picture is located at image edge region, determine
The face is not main body of taking pictures;Or,
Any pictures in two pictures are determined when the face in the picture is less than third predetermined threshold value
The face is not main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture, really
The fixed face is not main body of taking pictures.
6. according to the method described in claim 1, it is characterized in that, the method further includes:
When the time interval between the shooting time information of two pictures is less than preset duration, two pictures are determined
For similar pictures.
7. a kind of similar pictures judgment means, which is characterized in that described device includes:
Feature Points Matching module, two pictures for treating judgement carry out Feature Points Matching;
Face detection module, for when Feature Points Matching degree is more than the first predetermined threshold value, being carried out respectively to two pictures
Face datection;
Face recognition module, if for including face in two pictures, respectively to two pictures into pedestrian
Face identifies, obtains the first face characteristic and the second face characteristic, including:The main body of taking pictures of two pictures is determined respectively;Point
The other main body of taking pictures to two pictures carries out feature extraction, obtains first face characteristic and second face is special
Sign;
Matching module, for being matched to first face characteristic and the second face characteristic;
Similar pictures determining module, for when face characteristic matching degree is more than the second predetermined threshold value, determining two pictures
For similar pictures.
8. device according to claim 7, which is characterized in that the Feature Points Matching module is used for special using Scale invariant
It levies SIFT, accelerate robust feature SURF, color characteristic or structural edge feature, two pictures for treating judgement carry out characteristic point
Match.
9. device according to claim 7, which is characterized in that the face recognition module is used for:
For any pictures in two pictures, when the face in the picture is located at picture intermediate region, by institute
It states face and is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture is more than third predetermined threshold value, by institute
It states face and is determined as main body of taking pictures;Or,
For any pictures in two pictures, when the picture includes at least two being smaller than the 4th default threshold
When the face of value, at least two face is determined as main body of taking pictures.
10. device according to claim 9, which is characterized in that described device further includes:
Whether number judgment module, the number for judging to take pictures included by main body in two pictures are identical;
The face recognition module, for when number is identical, executing the step that respectively two pictures are carried out with recognition of face
Suddenly.
11. device according to claim 7, which is characterized in that the face recognition module is used for:
For any pictures in two pictures, when the face in the picture is located at image edge region, determine
The face is not main body of taking pictures;Or,
Any pictures in two pictures are determined when the face in the picture is less than third predetermined threshold value
The face is not main body of taking pictures;Or,
For any pictures in two pictures, when the face in the picture does not meet default human face posture, really
The fixed face is not main body of taking pictures.
12. device according to claim 7, which is characterized in that similar pictures determining module is additionally operable to when described two figures
When time interval between the shooting time information of piece is less than preset duration, determine that two pictures are similar pictures.
13. a kind of similar pictures judgment means, which is characterized in that including:
Processor;
Memory for storing the executable instruction of processor;
Wherein, the processor is configured as:
Two pictures for treating judgement carry out Feature Points Matching;
When Feature Points Matching degree is more than the first predetermined threshold value, Face datection is carried out respectively to two pictures;
If including face in two pictures, recognition of face is carried out to two pictures respectively, is obtained the first
Face feature and the second face characteristic, including:The main body of taking pictures of two pictures is determined respectively;Respectively to two pictures
Main body of taking pictures carries out feature extraction, obtains first face characteristic and second face characteristic;
First face characteristic and the second face characteristic are matched, when face characteristic matching degree is more than the second predetermined threshold value
When, determine that two pictures are similar pictures.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510463518.3A CN105069426B (en) | 2015-07-31 | 2015-07-31 | Similar pictures judgment method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510463518.3A CN105069426B (en) | 2015-07-31 | 2015-07-31 | Similar pictures judgment method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105069426A CN105069426A (en) | 2015-11-18 |
CN105069426B true CN105069426B (en) | 2018-09-04 |
Family
ID=54498788
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510463518.3A Active CN105069426B (en) | 2015-07-31 | 2015-07-31 | Similar pictures judgment method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105069426B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105487774B (en) * | 2015-11-27 | 2019-04-19 | 小米科技有限责任公司 | Image group technology and device |
CN105574512A (en) * | 2015-12-21 | 2016-05-11 | 小米科技有限责任公司 | Method and device for processing image |
CN106919604A (en) * | 2015-12-25 | 2017-07-04 | 北京奇虎科技有限公司 | Electron album management method and device |
CN106021542A (en) * | 2016-05-26 | 2016-10-12 | 珠海市魅族科技有限公司 | Image display method and image server, terminal |
CN106339695B (en) * | 2016-09-20 | 2019-11-15 | 北京小米移动软件有限公司 | Face similarity detection method, device and terminal |
CN107330859B (en) * | 2017-06-30 | 2021-06-15 | Oppo广东移动通信有限公司 | Image processing method and device, storage medium and terminal |
CN107516105B (en) * | 2017-07-20 | 2020-06-16 | 阿里巴巴集团控股有限公司 | Image processing method and device |
CN108022274B (en) * | 2017-11-29 | 2021-10-01 | Oppo广东移动通信有限公司 | Image processing method, image processing device, computer equipment and computer readable storage medium |
CN108908377B (en) * | 2018-07-06 | 2020-06-23 | 达闼科技(北京)有限公司 | Speaker recognition method and device and robot |
CN110084298B (en) * | 2019-04-23 | 2021-09-28 | 北京百度网讯科技有限公司 | Method and device for detecting image similarity |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101510257A (en) * | 2009-03-31 | 2009-08-19 | 华为技术有限公司 | Human face similarity degree matching method and device |
KR20110103223A (en) * | 2010-03-12 | 2011-09-20 | 경북대학교 산학협력단 | Keypoint matching system and method using sift algorithm for the face recognition |
CN103631889A (en) * | 2013-11-15 | 2014-03-12 | 北京奇虎科技有限公司 | Image recognizing method and device |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013145496A1 (en) * | 2012-03-27 | 2013-10-03 | 日本電気株式会社 | Information processing device, information processing method, and program |
-
2015
- 2015-07-31 CN CN201510463518.3A patent/CN105069426B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101510257A (en) * | 2009-03-31 | 2009-08-19 | 华为技术有限公司 | Human face similarity degree matching method and device |
KR20110103223A (en) * | 2010-03-12 | 2011-09-20 | 경북대학교 산학협력단 | Keypoint matching system and method using sift algorithm for the face recognition |
CN103631889A (en) * | 2013-11-15 | 2014-03-12 | 北京奇虎科技有限公司 | Image recognizing method and device |
Also Published As
Publication number | Publication date |
---|---|
CN105069426A (en) | 2015-11-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105069426B (en) | Similar pictures judgment method and device | |
CN104850828B (en) | Character recognition method and device | |
CN105491289B (en) | Prevent from taking pictures the method and device blocked | |
CN106454336B (en) | The method and device and terminal that detection terminal camera is blocked | |
CN106951884A (en) | Gather method, device and the electronic equipment of fingerprint | |
CN105205494B (en) | Similar pictures recognition methods and device | |
CN105160320B (en) | Fingerprint identification method, device and mobile terminal | |
CN109446994A (en) | Gesture critical point detection method, apparatus, electronic equipment and storage medium | |
CN106709399A (en) | Fingerprint identification method and device | |
CN105653085B (en) | Touch-responsive method and apparatus | |
CN107992814A (en) | Object finding method and device | |
CN107688781A (en) | Face identification method and device | |
CN109493852A (en) | A kind of evaluating method and device of speech recognition | |
CN106375782A (en) | Video playing method and device | |
CN109557999A (en) | Bright screen control method, device and storage medium | |
CN108320756B (en) | Method and device for detecting whether audio is pure music audio | |
CN105335714B (en) | Photo processing method, device and equipment | |
CN104408404A (en) | Face identification method and apparatus | |
CN104933700B (en) | A kind of method and apparatus carrying out picture material identification | |
CN106503537B (en) | Verification method and device based on flexible display screen | |
CN108108671A (en) | Description of product information acquisition method and device | |
CN108154466A (en) | Image processing method and device | |
CN105630239B (en) | Operate detection method and device | |
CN108717542A (en) | Identify the method, apparatus and computer readable storage medium of character area | |
CN109034150A (en) | Image processing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |