[go: up one dir, main page]

CN111199029B - Face recognition device and face recognition method - Google Patents

Face recognition device and face recognition method Download PDF

Info

Publication number
CN111199029B
CN111199029B CN201811367426.5A CN201811367426A CN111199029B CN 111199029 B CN111199029 B CN 111199029B CN 201811367426 A CN201811367426 A CN 201811367426A CN 111199029 B CN111199029 B CN 111199029B
Authority
CN
China
Prior art keywords
similarity
face
face image
image
current
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
Application number
CN201811367426.5A
Other languages
Chinese (zh)
Other versions
CN111199029A (en
Inventor
顾炯
周文
曹永刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ricoh Co Ltd
Original Assignee
Ricoh Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ricoh Co Ltd filed Critical Ricoh Co Ltd
Priority to CN201811367426.5A priority Critical patent/CN111199029B/en
Publication of CN111199029A publication Critical patent/CN111199029A/en
Application granted granted Critical
Publication of CN111199029B publication Critical patent/CN111199029B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • 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
    • G06V40/172—Classification, e.g. identification
    • 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
    • G06V40/50—Maintenance of biometric data or enrolment thereof

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Analysis (AREA)
  • Collating Specific Patterns (AREA)

Abstract

The invention relates to a face recognition device and a face recognition method, which are used for recognizing the face of a recognition object, wherein the face recognition device comprises: a reference image storage unit; an imaging unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a comprehensive similarity comparison judgment unit; and an identification object determination unit. The reference image storage part stores a plurality of groups of reference face images, the camera part acquires the face images to obtain a current face image, the similarity calculation part respectively compares the current face image with each group of reference face images in sequence, the angle deviation calculation part sequentially compares the current face image with each group of reference face images in angle, the comprehensive similarity calculation part calculates to obtain comprehensive similarity corresponding to a plurality of current face images, the comprehensive similarity comparison judgment part judges whether the highest comprehensive similarity is larger than or equal to a similarity threshold value, and the recognition object judgment part judges the identity information of the current recognition object.

Description

Face recognition device and face recognition method
Technical Field
The present invention relates to a face recognition apparatus and a face recognition method, and more particularly, to a face recognition apparatus and a face recognition method for recognizing a face of a recognition target.
Background
Biometric-based identification techniques have an increasingly important role and role in social life. In various biological authentication methods, recognition and authentication based on facial features of people are widely focused and valued because of the advantages of no invasiveness, low cost, good concealment, no need of special cooperation of testees and the like, and the method has wide application prospect.
In the existing face recognition device, one or a few face images of a recognition object need to be collected in advance as a reference face image, when the face recognition operation is carried out, the current face image of the recognition object in a current state is collected and compared with the reference face image to obtain similarity, and then the similarity is compared with a preset similarity threshold value to judge the identity information of the current recognition object. In practical application, the scheme is likely to generate misjudgment, and the reasons mainly include the following: 1. the number of the reference face images of a recognition object is small, and the related characteristics of the face are single; 2. the acquired current face image is likely to have lower definition; 3. the reference face image is not updated, however, the recognition object may change its hairstyle, make-up, or wear glasses to change its appearance.
Disclosure of Invention
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a face recognition device and a face recognition method capable of preparing a quick face recognition of a recognition target.
In order to achieve the above object, the present invention adopts the following structure:
< Structure one >
The invention provides a face recognition device for recognizing a face of a recognition object, which is characterized by comprising the following components: a reference image storage unit; an imaging unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a comprehensive similarity comparison judgment unit; and a recognition object determination unit in which a plurality of sets of reference face images corresponding to the plurality of recognition objects are stored in the reference image storage unit, each set of reference face images including a reference face image, a reference left face image, a reference right face image, a reference look-up face image, and a reference look-down face image of the recognition object, the image pickup unit acquires the current face image from the recognition object in the current state, the similarity calculation unit sequentially performs similarity comparison between the current face image and each of the reference face images, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image, to obtain a plurality of face similarities, left side similarities, right side similarities, look-up similarities, and look-down similarities corresponding to each set of reference face images, the angle deviation calculating part respectively performs angle comparison on the current face image and a reference front face image, a reference left face image, a reference right face image, a reference back face image and a reference top face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, back vision deviation angles and top view deviation angles which respectively correspond to each group of reference face images, the comprehensive similarity calculating part calculates the comprehensive similarity between the plurality of current face images and each group of reference face images according to the front face similarity, the left side similarity, the right side similarity, the back vision similarity and the top view similarity of each group and the corresponding front face deviation angles, the left side deviation angles, the right side deviation angles, the back vision deviation angles and the top view deviation angles according to a preset calculation rule, the comprehensive similarity comparison and judgment part compares the highest comprehensive similarity with a preset similarity threshold value to judge whether the highest comprehensive similarity is larger than or equal to the similarity threshold value, and when the highest comprehensive similarity is larger than or equal to the similarity threshold value, the identification object judgment part judges the corresponding identity information as the current identification object identity information according to the highest comprehensive similarity.
< Structure two >
The invention also provides a face recognition device for recognizing the face of the recognition object, which is characterized by comprising: a reference image storage unit; an imaging unit; a control unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a highest comprehensive similarity acquisition unit; an identity information acquisition unit; an information judgment unit; a definition calculating unit; a comprehensive definition calculation unit; and a recognition object judging part, wherein the reference image storing part stores a plurality of groups of reference face images corresponding to a plurality of recognition objects respectively, each group of reference face images comprises a reference face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image of the recognition object, the image pickup part acquires a plurality of face images of the recognition object in the current state to obtain a plurality of current face images, and the control part controls the similarity calculating part to sequentially compare the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images respectively to obtain a plurality of face similarities, left-side similarities, right-side similarities, look-up similarities and look-down similarities corresponding to each group of reference face images respectively; the control angle deviation calculating part is used for respectively comparing the current face image with a reference front face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which respectively correspond to each group of reference face images; the control comprehensive similarity calculation part calculates a plurality of comprehensive similarity according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity, the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group according to a preset calculation rule; the control highest comprehensive similarity obtaining part obtains the highest comprehensive similarity according to the plurality of comprehensive similarities as the highest comprehensive similarity; the control part further controls the identity information acquisition part to acquire corresponding identity information according to the highest comprehensive similarity, the control part further controls the information judgment part to judge whether the plurality of identity information respectively corresponding to the plurality of current face images and acquired by the identity information acquisition part correspond to different recognition objects, when judging that the plurality of identity information respectively corresponds to the different recognition objects, the definition calculation part respectively performs image analysis on the plurality of current face images to obtain a plurality of definition, the comprehensive definition calculation part respectively calculates corresponding comprehensive definition for the different recognition objects according to the plurality of definition, and the recognition object judgment part judges that the corresponding recognition object is the current recognition object according to the highest comprehensive definition.
The invention also provides a face recognition method for recognizing the face of the recognition object, which is characterized by comprising the following steps: a reference image storage step of acquiring and storing a plurality of groups of reference face images corresponding to a plurality of recognition objects, respectively, each group of reference face images including a reference front face image, a reference left face image, a reference right face image, a reference back face image, and a reference top face image of the recognition object; a shooting step, namely acquiring a face image of an identification object in a current state to obtain a current face image; a similarity calculation step, namely comparing the current face image with a reference front face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face similarity, left similarity, right similarity, look-up similarity and look-down similarity which correspond to each group of reference face images respectively; an angle deviation calculating step of comparing the current face image with the reference front face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which correspond to each group of reference face images respectively; a comprehensive similarity calculation step, namely calculating a plurality of comprehensive similarity between the current face images and each group of reference face images according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity, the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group according to a preset calculation rule; a comprehensive similarity comparison and judgment step, namely comparing the highest comprehensive similarity serving as the highest comprehensive similarity with a preset similarity threshold value, and judging whether the highest comprehensive similarity is larger than or equal to the similarity threshold value; and a step of identifying the object, in which when the highest comprehensive similarity is greater than or equal to a similarity threshold, corresponding identity information is identified as the current identity information of the identifying object according to the highest comprehensive similarity.
Effects and effects of the invention
According to the face recognition device and the face recognition method, the reference image storage part stores a plurality of groups of reference face images including the reference front face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image of the recognition object, the similarity calculation part compares the similarity of the current face image with that of each group of reference face images in sequence, the angle deviation calculation part compares the current face image with each group of reference face images in sequence, the comprehensive similarity calculation part obtains the comprehensive similarity between the current face image and each group of reference face images according to the preset calculation rule according to the comparison results of the similarity calculation part and the angle deviation calculation part, and the recognition object judgment part judges corresponding identity information as the identity information of the current recognition object according to the highest comprehensive similarity.
Drawings
Fig. 1 is a block diagram of a face recognition apparatus according to a first embodiment of the present invention;
FIG. 2 is a diagram showing an information display screen according to an embodiment of the present invention;
FIG. 3 is a diagram showing an information display screen according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating operations of the face recognition apparatus according to the first embodiment of the present invention;
fig. 5 is a diagram showing steps of updating processing of the face recognition device according to the first embodiment of the present invention;
fig. 6 is a block diagram of a face recognition device according to a second embodiment of the present invention; and
fig. 7 is a flowchart illustrating operations of the face recognition apparatus according to the second embodiment of the present invention.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement of the purposes and the effects of the present invention easy to understand, the following describes the face recognition device of the present invention with reference to the embodiments and the drawings.
As a first aspect, the present invention provides a face recognition apparatus for recognizing a face of a recognition target, comprising: a reference image storage unit; an imaging unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a comprehensive similarity comparison judgment unit; and a recognition object determination unit in which a plurality of sets of reference face images corresponding to the plurality of recognition objects are stored in the reference image storage unit, each set of reference face images including a reference face image, a reference left face image, a reference right face image, a reference look-up face image, and a reference look-down face image of the recognition object, the image pickup unit acquires the current face image from the recognition object in the current state, the similarity calculation unit sequentially performs similarity comparison between the current face image and each of the reference face images, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image, to obtain a plurality of face similarities, left side similarities, right side similarities, look-up similarities, and look-down similarities corresponding to each set of reference face images, the angle deviation calculating part respectively performs angle comparison on the current face image and a reference front face image, a reference left face image, a reference right face image, a reference back face image and a reference top face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, back vision deviation angles and top view deviation angles which respectively correspond to each group of reference face images, the comprehensive similarity calculating part calculates the comprehensive similarity between the plurality of current face images and each group of reference face images according to the front face similarity, the left side similarity, the right side similarity, the back vision similarity and the top view similarity of each group and the corresponding front face deviation angles, the left side deviation angles, the right side deviation angles, the back vision deviation angles and the top view deviation angles according to a preset calculation rule, the comprehensive similarity comparison and judgment part compares the highest comprehensive similarity with a preset similarity threshold value to judge whether the highest comprehensive similarity is larger than or equal to the similarity threshold value, and when the highest comprehensive similarity is larger than or equal to the similarity threshold value, the identification object judgment part judges the corresponding identity information as the current identification object identity information according to the highest comprehensive similarity.
In the first aspect, the present invention may further include: an update recording section storing update times of each set of reference face images, an update control section determining corresponding identity information based on a highest integrated similarity of the current face images, a similarity determination section controlling the similarity determination section to determine whether or not a highest integrated similarity between the current face image and the current reference face image set is between 80% and 90%, an update control section controlling the current time acquisition section to acquire a current time when it is determined that between 80% and 90%, and controlling the update interval determination section to determine whether or not an interval between a latest update time corresponding to the current reference face image set stored in the update recording section and the current time is a predetermined time interval, and when it is determined that the interval is greater than a predetermined time interval, the update control section controlling the similarity calculation section to calculate a corresponding set of reference face images as a current reference face image set, respectively, controlling the similarity calculation section to determine whether or not a highest integrated similarity between the current face image and the current reference face image set is between 80% and 90%, when it is determined that between 80% and 90% of the current reference face images, the update time is smaller than a value of the current reference face image stored in the update interval determination section, the update control section controlling the update time between the update time and the update time corresponding to the current reference face image set, and the update control section controlling the similarity calculation section to calculate a similarity between the current face image and the current reference face image set based on the current reference face image set, and controls the update recording section to record the update time of the set of reference face images correspondingly.
In the first aspect, the present invention may further include: a reference image deletion section in which each set of reference face images in the reference image storage section includes at least one reference front face image, at least one reference left face image, at least one reference right face image, at least one reference back face image, and at least one reference top face image, and the reference image update section includes an orientation determination unit that determines a face orientation of the current face image as a current face orientation based on a front face deviation angle, a left side deviation angle, a right side deviation angle, a back face deviation angle, and a top face deviation angle corresponding to the current reference face image set; the number judgment unit judges whether or not the number of reference face images in the same direction as the current face direction in the current reference face image group is smaller than a predetermined number, and when judged to be smaller, the update control unit sets the current face image as a new reference face image corresponding to the current face direction, sets the current time as the storage time of the new reference face image, controls the reference image storage unit to additionally store the new reference face image and the storage time thereof, and when judged not to be smaller, the update control unit sets the current face image as a new reference face image corresponding to the current face direction, sets the current time as the storage time of the new reference face image, controls the reference image storage unit to additionally store the new reference face image and the storage time thereof, and controls the reference image deletion unit to delete the reference face image corresponding to the current face direction and having the earliest storage time in the reference image storage unit.
In the first embodiment, the present invention may further have the feature that: the predetermined calculation rules comprise similarity coefficient giving rules and comprehensive similarity calculation rules, wherein the similarity coefficient giving rules are as follows: first, the positive face deviation angle, the left side deviation angle, the right side deviation angle, the upward vision deviation angle, and the downward vision deviation angle corresponding to each set of reference face images are expressed as (x) 0 ,y 0 )、(x 1 ,y 1 )、(x 2 ,y 2 )、(x 3 ,y 3 ) (x) 4 ,y 4 ) Thereby calculating a group of deviation degrees of the face deviation angle, the left deviation angle, the right deviation angle, the backstop deviation angle and the overlook deviation angle And +.>Then, a group of deviation degrees are respectively added with a fixed value of 1e-4 and are inverted to obtain a corresponding group of sequence values which are respectively +.> And +.>Finally, a group of similarity coefficients corresponding to the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity are calculated as +.>And +.>Wherein s=w 0 +w 1 +w 2 +w 3 +w 4 The comprehensive similarity calculation rule is to multiply the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity with the corresponding similarity coefficients respectively and then add the multiplied similarity coefficients to obtain the comprehensive similarity.
In the first embodiment, the present invention may further have the feature that: the camera shooting control unit controls the real face judging unit to judge whether the face image is a real face or not once the camera shoots a face image, and further controls the output unit to output the face image as a current face image when judging that the face image is the real face.
In the first embodiment, the present invention may further have the feature that: the face orientation of the reference front face image is the front, the face orientation of the reference left face image is the left side, the face orientation of the reference right face image is the right side, the face orientation of the reference bottom face image is the upper side, the face orientation of the reference top face image is the lower side, the angular deviation of the face orientation of the reference left face image from the face orientation of the reference front face image and the angular deviation of the face orientation of the reference right face image from the face orientation of the reference front face image are 10 DEG to 30 DEG, and the angular deviation of the face orientation of the reference bottom face image from the face orientation of the reference front face image and the angular deviation of the face orientation of the reference top face image from the face orientation of the reference front face image are 10 DEG to 20 deg.
In the first aspect, the present invention may further include: the system comprises a preset threshold storage part, a threshold updating control part, a threshold accuracy calculating part and a similarity threshold storage part, wherein the preset threshold storage part stores a plurality of preset similarity thresholds, the threshold updating control part controls the threshold accuracy calculating part to calculate the accuracy value of each preset similarity threshold according to the preset similarity threshold based on an accuracy value calculating rule and a plurality of groups of reference face images stored by the reference image storage part according to a preset threshold updating time point, and controls the similarity threshold storage part to store the preset similarity threshold with the highest accuracy value as the similarity threshold with the comprehensive similarity comparison judging part as a judgment reference.
In the first embodiment, the present invention may further have the feature that: the accurate value calculation rule is as follows: firstly, traversing all reference face images in a reference image storage part, pairing reference face images belonging to the same group in pairs to form a plurality of same group pairs, pairing reference face images not belonging to the same group in pairs to form a plurality of different group pairs, then calculating the similarity between the two reference face images of each same group pair and each different group pair, and finally calculating the accurate value a of each preset similarity threshold value as: Wherein a is 1 For similarity with a similarity higher than a preset similarity thresholdLogarithm of the same set of pairs of degrees, a 2 A is the logarithm of the heterogroup pair with similarity below a preset similarity threshold 3 Is the sum of the logarithm of the same group of pairs and the logarithm of different groups of pairs.
As a second aspect, the present invention provides a face recognition apparatus for recognizing a face of a recognition target, comprising: a reference image storage unit; an imaging unit; a control unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a highest comprehensive similarity acquisition unit; an identity information acquisition unit; an information judgment unit; a definition calculating unit; a comprehensive definition calculation unit; and a recognition object judging part, wherein the reference image storing part stores a plurality of groups of reference face images corresponding to a plurality of recognition objects respectively, each group of reference face images comprises a reference face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image of the recognition object, the image pickup part acquires a plurality of face images of the recognition object in the current state to obtain a plurality of current face images, and the control part controls the similarity calculating part to sequentially compare the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images respectively to obtain a plurality of face similarities, left-side similarities, right-side similarities, look-up similarities and look-down similarities corresponding to each group of reference face images respectively; the control angle deviation calculating part is used for respectively comparing the current face image with a reference front face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which respectively correspond to each group of reference face images; the control comprehensive similarity calculation part calculates a plurality of comprehensive similarity according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity, the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group according to a preset calculation rule; the control highest comprehensive similarity obtaining part obtains the highest comprehensive similarity according to the plurality of comprehensive similarities as the highest comprehensive similarity; the control part further controls the identity information acquisition part to acquire corresponding identity information according to the highest comprehensive similarity, the control part further controls the information judgment part to judge whether the plurality of identity information respectively corresponding to the plurality of current face images and acquired by the identity information acquisition part correspond to different recognition objects, when judging that the plurality of identity information respectively corresponds to the different recognition objects, the definition calculation part respectively performs image analysis on the plurality of current face images to obtain a plurality of definition, the comprehensive definition calculation part respectively calculates corresponding comprehensive definition for the different recognition objects according to the plurality of definition, and the recognition object judgment part judges that the corresponding recognition object is the current recognition object according to the highest comprehensive definition.
In a second embodiment, the present invention may further include the following features: wherein the integrated definition calculating section includes: a definition coefficient storage unit for storing a plurality of image definition and corresponding definition coefficients; the definition coefficient obtaining unit sequentially retrieves and obtains the corresponding definition coefficient from the definition coefficient storage unit according to the definition of each current face image; and the comprehensive accumulation unit is used for accumulating the definition coefficients of the current face image corresponding to each different recognition object in sequence to obtain the comprehensive definition of each recognition object.
As a third aspect, the present invention provides a face recognition method for recognizing a face of a recognition object, comprising the steps of: a reference image storage step of acquiring and storing a plurality of groups of reference face images corresponding to a plurality of recognition objects, respectively, each group of reference face images including a reference front face image, a reference left face image, a reference right face image, a reference back face image, and a reference top face image of the recognition object; a shooting step, namely acquiring a face image of an identification object in a current state to obtain a current face image; a similarity calculation step, namely comparing the current face image with a reference front face image, a reference left face image, a reference right face image, a reference look-up face image and a reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face similarity, left similarity, right similarity, look-up similarity and look-down similarity which correspond to each group of reference face images respectively; an angle deviation calculating step of comparing the current face image with the reference front face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which correspond to each group of reference face images respectively; a comprehensive similarity calculation step, namely calculating a plurality of comprehensive similarity between the current face images and each group of reference face images according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity, the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group according to a preset calculation rule; a comprehensive similarity comparison and judgment step, namely comparing the highest comprehensive similarity serving as the highest comprehensive similarity with a preset similarity threshold value, and judging whether the highest comprehensive similarity is larger than or equal to the similarity threshold value; and a step of identifying the object, in which when the highest comprehensive similarity is greater than or equal to a similarity threshold, corresponding identity information is identified as the current identity information of the identifying object according to the highest comprehensive similarity.
Example 1
In the first embodiment, the face recognition device is used for recognizing the face of the recognition object. Specifically, for example, the face recognition device may be an attendance checking device of an enterprise, and a set of reference face images of each employee of the enterprise are stored in advance. When a person checks his attendance and punches a card, the attendance checking device performs face recognition on the person, and when the attendance checking device recognizes the identity information of the person, the attendance checking operation is finished on the person. If the person is a non-enterprise employee, the attendance device displays prompt information of strangers.
Fig. 1 is a block diagram of a face recognition apparatus according to a first embodiment of the present invention.
As shown in fig. 1, the face recognition apparatus 100 includes a reference image storage unit 1, an imaging unit 2, a similarity calculation unit 3, an angle deviation calculation unit 4, a comprehensive similarity calculation unit 5, a comprehensive similarity comparison determination unit 6, a recognition object determination unit 7, a screen storage unit 8, an output display unit 9, an output display control unit 10, a similarity determination unit 11, a current time acquisition unit 12, an update recording unit 13, an update interval determination unit 14, a reference similarity calculation unit 15, a reference similarity determination unit 16, a reference image update unit 17, a reference image deletion unit 18, an update control unit 19, a preset threshold storage unit 20, a threshold accuracy calculation unit 21, a similarity threshold storage unit 22, a threshold update control unit 23, a communication unit 24, and a control unit 25.
The reference image storage unit 1 stores a plurality of sets of reference face images corresponding to a plurality of recognition objects, each set of reference face images including at least one reference front face image, at least one reference left face image, at least one reference right face image, at least one reference back face image, and at least one reference top face image of the recognition object. The face orientation of the reference front face image is the front, the face orientation of the reference left face image is the left side, the face orientation of the reference right face image is the right side, the face orientation of the reference back face image is the upper side, and the face orientation of the reference top face image is the lower side. In this embodiment, the number of reference front face images, reference left face images, reference right face images, reference back face images, and reference top face images of each group of reference face images is one, and of course, may be set to be plural according to the actual needs of the enterprise.
The face orientation of the reference front face image is taken as the reference face orientation, the angular deviation between the face orientation of the reference left face image and the face orientation of the reference right face image and the reference face orientation is 10-30 degrees, and the angular deviation between the face orientation of the reference back face image and the face orientation of the reference overlook face image and the reference face orientation is 10-20 degrees.
The image pickup part 2 collects face images of the recognition object in the current state to obtain the current face image, and comprises a camera, an image pickup control unit, a true face judging unit and an output unit. The camera is used for shooting and collecting face images. The image pickup control unit is used for controlling the real face judging unit to judge whether the face image is a real face or not, and when the real face judging unit judges that the face image is a real face, the image pickup control unit controls the output unit to output the face image as a current face image. For example, in an attendance card punching system of an enterprise, after a camera shoots an image, a real face judging unit judges whether the image is a real face image or not so as to prevent the camera from acquiring a photo image of a certain employee which is prepared in advance.
The similarity calculation unit 3 sequentially compares the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image in each of the reference face images, respectively, to obtain a plurality of face similarities, left-side similarities, right-side similarities, look-up similarities, and look-down similarities, which correspond to each of the reference face images.
The angle deviation calculating unit 4 sequentially compares the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image in each of the reference face images, respectively, to obtain a plurality of face deviation angles, left deviation angles, right deviation angles, look-up deviation angles, and look-down deviation angles, which correspond to each of the reference face images.
The comprehensive similarity calculation unit 5 calculates the comprehensive similarity between the plurality of current face images and each group of reference face images based on the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the look-down similarity, the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle, and the look-down deviation angle of each group according to a predetermined calculation rule. Wherein the predetermined calculation rule includes a similarity coefficient assigning rule and a comprehensive similarity calculation rule.
The similarity coefficient assignment rule is:
first, the positive face deviation angle, the left side deviation angle, the right side deviation angle, the upward vision deviation angle, and the downward vision deviation angle corresponding to each set of reference face images are expressed as (x) 0 ,y 0 )、(x 1 ,y 1 )、(x 2 ,y 2 )、(x 3 ,y 3 ) (x) 4 ,y 4 ) Thereby calculating the face deviation angle b 0 Left offset angle b 1 Right side deviation angle b 2 Angle b of upward visual deviation 3 Top-view deviation angle b 4 Is respectively as followsand Wherein x is 0 Is the deviation angle, x, of the right front of the face of the current face image and the right front of the face of the reference face image in the horizontal direction 1 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference left face image in the horizontal direction, x 2 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference right face image in the horizontal direction, x 3 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference look-up face image in the horizontal direction, x 4 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference overlook face image in the horizontal direction. y is 0 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference face image in the vertical direction, y 1 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference left face image in the vertical direction, y 2 Is the deviation angle of the right front of the face of the current face image and the right front of the face of the reference right face image in the vertical direction, y 3 Is the deviation angle of the front face of the current face image and the front face of the reference face-up image in the vertical direction, y 4 Is at presentThe deviation angle in the vertical direction between the front face of the face image and the front face of the reference planar face image.
Then, the deviation degree is added with a fixed value of 1e-4 and the inverse is calculated to obtain a corresponding group of sequence values respectively asAnd +.>Wherein w is 0 Is the sequence value corresponding to the face deviation angle, w 1 Is the sequence value corresponding to the left deviation angle, w 2 Is the sequence value corresponding to the right deviation angle, w 3 Is the sequence value corresponding to the upward error angle, w 4 The sequence value corresponds to the top-down deviation angle.
Finally, a group of similarity coefficients corresponding to the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity are calculated and obtained respectivelyAnd +.>Wherein w is 0 Is the similarity coefficient of the positive face similarity, w 1 Similarity coefficient, w, being left-hand similarity 2 Similarity coefficient, w, is the right-side similarity 3 To look up the similarity coefficient of similarity, w 4 For the similarity coefficient of the overlooking similarity, s is the corresponding sequence value w 0 、w 1 、w 2 、w 3 W 4 Sum, i.e. s=w 0 +w 1 +w 2 +w 3 +w 4 。
The comprehensive similarity calculation rule is as follows: and multiplying the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity by the corresponding similarity coefficients respectively, and then adding to obtain the comprehensive similarity.
The integrated similarity comparison and judgment unit 6 compares the highest integrated similarity with a predetermined similarity threshold value, and judges whether or not the highest integrated similarity is equal to or greater than the similarity threshold value.
The identification object determining unit 7 is configured to determine, when the highest overall similarity is equal to or greater than a similarity threshold, corresponding identity information as the current identification object identity information based on the highest overall similarity.
The screen storage unit 8 stores a stranger presenting screen 8a and a plurality of identification information display screens 8b corresponding to the plurality of identification objects.
Fig. 2 is a schematic diagram of an information display screen according to an embodiment of the present invention.
As shown in fig. 2, the stranger prompt screen 8a includes a stranger prompt box 81a, and the stranger prompt box 81a is used for displaying a prompt message of "the current recognition object is a stranger".
Fig. 3 is a schematic diagram of an information display screen according to an embodiment of the present invention.
As shown in fig. 3, the plurality of identity information display screens 8b each include an identity information display frame 81b for displaying the identity information of the corresponding recognition object. For example, the content displayed in the identity information display box corresponding to employee a of the enterprise is "employee a is the current identification object".
The output display unit 9 displays the screen.
The output display control unit 10 controls the output display unit 9 to display the screen according to the determination result of the recognition object determination unit 7.
The similarity determination section 11 is for determining whether the highest overall similarity is between 80% and 90%.
The current time acquisition unit 12 acquires a current time.
The update recording section 13 is for recording and storing the update time of each set of reference face images.
The update interval determination unit 14 determines whether or not the interval between the latest update time corresponding to the current reference face image group stored in the update recording unit 13 and the current time acquired by the current time acquisition unit 12 is within a predetermined time interval. In this embodiment, the predetermined time interval is 1 day, which may be set to other days according to the actual needs of the enterprise.
The reference similarity calculating section 15 is configured to calculate current reference similarity between the current face image and each reference face image in the current reference face image group when the interval between the latest update time and the current time is greater than a predetermined time interval, and calculate a plurality of reference similarities between each pair of reference face images in the current reference face image group.
The reference similarity determination section 16 is for determining whether or not the minimum value in the current reference similarity is smaller than the average value of the reference similarity.
The reference image updating unit 17 is configured to update the corresponding reference face image stored in the reference image storage unit 1 based on the current face image when the minimum value of the current reference similarity is smaller than the average value of the reference similarity. The reference image updating section 17 includes an orientation determination unit and a number determination unit.
The orientation determination unit is used for determining the face orientation of the current face image as the current face orientation according to the positive face deviation angle, the left side deviation angle, the right side deviation angle, the upward vision deviation angle and the overlooking deviation angle corresponding to the current reference face image group. For example, when the value of the left deviation angle is maximum, the face orientation of the reference left face image in the current reference face image group is the current face orientation.
The number judging unit is used for judging whether the number of the reference face images in the same direction as the current face direction in the current reference face image group is smaller than a preset number.
The reference image deletion unit 18 deletes the reference face image which corresponds to the current face orientation and has the earliest storage time in the reference image storage unit 1.
The update control unit 19 is for controlling operations of components related to image update, and includes: when the identification object determining section 7 determines the identity information of the current identification object, the control similarity determining section 11 determines whether or not the highest overall similarity between the current face image and the current reference face image set is between 80% and 90% using the corresponding one of the reference face images as the current reference face image set; when judging that the time is between 80% and 90%, controlling the current time acquisition part to acquire the current time, and controlling the update interval judgment part 14 to judge whether the interval between the latest update time corresponding to the current reference face image group and the current time is larger than a preset time interval or not; when it is determined to be greater than the current reference similarity and the plurality of reference similarities, the control reference similarity calculation section 15 calculates the current reference similarity and the plurality of reference similarities, respectively, and further controls the reference similarity determination section 16 to determine whether or not the minimum value of the current reference similarity is smaller than the average value of the plurality of reference similarities; when the number of the reference face images is smaller than the preset number, the control direction judging unit judges the face direction of the current face image; when the image is determined to be smaller than the predetermined value, the control reference image storage unit 1 additionally stores the current face image and the current time; when it is determined that the stored time is not smaller than the predetermined time, the reference image storage unit 1 is controlled to additionally store the current face image and the current time, the reference image deletion unit 18 is controlled to delete the reference face image having the earliest stored time, and the update recording unit 13 is controlled to record the update time in association with the current face image and the current time.
The preset threshold value storage section 20 stores a plurality of preset similarity threshold values set in advance,
the accuracy calculating section 21 is configured to calculate the accurate value of each preset similarity threshold for the preset similarity threshold based on the accurate value calculating rule and the plurality of sets of reference face images stored in the reference image storing section 1.
The accurate value calculation rule is as follows:
first, all reference face images in the reference image storage unit are traversed, reference face images belonging to the same group are paired two by two to form a plurality of identical group pairs, and reference face images not belonging to the same group are paired two by two to form a plurality of different group pairs.
Then, the similarity between the two reference face images of each same group pair and each different group pair is calculated.
Finally, each preset phase is calculatedThe exact value of the similarity threshold, a, is:wherein a is 1 A is the logarithm of the same group pair with the similarity higher than a preset similarity threshold value 2 A is the logarithm of the heterogroup pair with similarity below a preset similarity threshold 3 Is the sum of the logarithm of the same group of pairs and the logarithm of different groups of pairs.
The similarity threshold storage unit 22 is configured to store a preset similarity threshold with the highest accuracy value as a similarity threshold with the integrated similarity comparison and judgment unit 6 as a judgment reference.
The threshold value update control unit 23 is for controlling operations of the means for updating the similarity threshold value, and includes: according to the preset threshold updating time point, the control threshold accuracy calculating section 21 calculates the accurate value of each preset similarity threshold, and controls the similarity threshold storing section 22 to store the similarity threshold having the integrated similarity comparison judging section 6 as the judgment reference. In this embodiment, the preset threshold updating time point is 24 points of each day, and of course, other times may be set according to the actual needs of the customer.
The communication unit 24 is used for data exchange between the respective constituent parts of the face recognition device 100.
The control unit 25 is configured to control operations of the respective constituent elements of the face recognition device 100.
The face recognition device of the embodiment collects the current face image of the recognition object in the current state, then analyzes the similarity between the current face image and each group of reference face images, calculates the comprehensive similarity, judges the identity information of the recognition object through the highest comprehensive similarity, and performs image updating operation on the group of reference face images corresponding to the recognition object.
Fig. 4 is a flowchart illustrating operations of the face recognition apparatus according to the first embodiment of the present invention.
As shown in fig. 4, in the first embodiment, the operation flow of the face recognition device 100 includes the following steps:
step S1-1-1, the camera shoots the face image, and then step S1-1-2 is carried out.
In the step S1-1-2, the image pickup control unit controls the real face judging unit to judge whether the face image shot by the camera is a real face or not, and the step S1-1-3 is entered when the judgment is yes, and the step S1-1-1 is entered when the judgment is no.
And step S1-1-3, the image pickup control unit controls the output unit to output the face image as the current face image, and then the step S1-1-4 is carried out.
In step S1-1-4, the similarity calculating unit 3 compares the current face image with the reference front face image, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image in each group of reference face images in order to obtain a plurality of front face similarities, left side similarities, right side similarities, look-up similarities, and look-down similarities, which correspond to each group of reference face images, respectively, and then proceeds to step S1-1-5.
In the step S1-1-5, the angular deviation calculating unit 4 performs angular comparison on the current face image and the reference front face image, the reference left face image, the reference right face image, the reference back view face image, and the reference top view face image in each group of reference face images in order, respectively, to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, back view deviation angles, and top view deviation angles, which correspond to each group of reference face images, respectively, and then proceeds to the step S1-1-6.
Step S1-1-6, the comprehensive similarity calculation part 5 calculates the comprehensive similarity between the plurality of current face images and each group of reference face images according to a preset calculation rule, and then the step S1-1-7 is performed.
In the step S1-1-7, the comprehensive similarity comparison and judgment unit 6 compares the highest comprehensive similarity with a predetermined similarity threshold value as the highest comprehensive similarity, judges whether the highest comprehensive similarity is equal to or greater than the predetermined threshold value, and proceeds to the step S1-1-8 when judging yes, and proceeds to the step S1-1-10 when judging no.
In step S1-1-8, the identification object determining unit 7 determines the corresponding identity information as the current identification object identity information based on the highest overall similarity, and then proceeds to step S1-1-9.
In step S1-1-9, the output display control unit 10 controls the output display unit 9 to display the identity information display screen 8b corresponding to the identity information of the current recognition object, and then proceeds to step S1-2 to update a set of reference face images corresponding to the identity information of the current recognition object.
In step S1-1-10, the recognition object determining unit 7 determines that the current recognition object is a stranger, and then proceeds to step S1-1-11.
In steps S1-1-11, the output display control unit 10 controls the output display unit 9 to display the stranger presenting screen 8a, and then enters the end state.
In the above-mentioned operation flow, the number of the reference front face image, the reference left face image, the reference right face image, the reference bottom face image, and the reference top face image of each group of reference face images is one, and when the number of the reference front face image, the reference left face image, the reference right face image, the reference bottom face image, and the reference top face image of each group of reference face images is plural, the above-mentioned steps S1-1-4 may be repeated plural times to obtain plural front face similarities, plural left side similarities, plural right side similarities, plural bottom view similarities, and plural top view similarities of each group of reference face images, and averaging the plurality of front face similarities, the plurality of left side similarities, the plurality of right side similarities, the plurality of look-up similarities and the plurality of overlook similarities to obtain an average front face similarity, an average left side similarity, an average right side similarity, an average look-up similarity and an average overlook similarity, which are used as the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the overlook similarity corresponding to each group of reference face images, repeating the steps S1-1-5 repeatedly, and performing the same for a plurality of times to obtain an average front face deviation angle, an average left side deviation angle, an average right side deviation angle, an average look-up deviation angle and an average overlook deviation angle which are used as the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group of reference face images, and then performing the steps S1-1-6.
Fig. 5 is a diagram showing steps of updating processing of the face recognition device according to the first embodiment of the present invention.
As shown in fig. 5, after the face recognition operation is completed, the face recognition device 100 of the present embodiment further performs update processing on the current reference image group in the reference image storage unit with the current reference face image as the face image to be updated, and the update processing S1-2 includes the steps of:
in step S1-2-1, the update control unit 19 sets a set of reference face images corresponding to the current identification object identity information as a current reference face image set, and the control similarity determination unit 11 determines whether the highest overall similarity between the current face image and the current reference face image set is 80% -90%, and if yes, the process proceeds to step S1-2-2, and if no, the process proceeds to an end state.
In step S1-2-2, the update control section 19 controls the current time acquisition section 12 to acquire the current time, and then proceeds to step S1-2-3.
In step S1-2-3, the update control unit 19 controls the update interval determination unit 14 to determine whether or not the interval between the latest update time corresponding to the current reference face image group stored in the update recording unit 13 and the current time is greater than a predetermined time interval, and when the determination is yes, the process proceeds to step S1-2-4, and when the determination is no, the process proceeds to an end state.
In step S1-2-4, the update control section 19 controls the reference similarity calculation section 15 to calculate the current reference similarity between the current face image and each of the reference face images in the current reference face image group and calculate the plurality of reference similarities between each pair of reference face images in the current reference face image group, respectively, and then proceeds to step S1-2-5.
In step S1-2-5, the update control unit 19 controls the reference similarity determination unit 16 to determine whether or not the minimum value in the current reference similarity is smaller than the average value of the reference similarity, and proceeds to step S1-2-6 when the determination is yes, and proceeds to the end state when the determination is no.
In step S1-2-6, the update control unit 19 controls the orientation determination unit to determine the face orientation of the current face image as the current face orientation based on the positive face deviation angle, the left side deviation angle, the right side deviation angle, the bottom view deviation angle, and the top view deviation angle corresponding to the current reference face image group, and then proceeds to step S1-2-7.
In step S1-2-7, the update control unit 19 controls the number judgment unit to judge whether or not the number of reference face images in the same orientation as the current face orientation in the current reference face image group is smaller than a predetermined number, and proceeds to step S1-2-8 when judged yes, and proceeds to step S1-2-9 when judged no.
In step S1-2-8, the update control unit 19 sets the current face image as a new reference face image corresponding to the current face orientation, sets the current time as the storage time of the new reference face image, controls the reference image storage unit 1 to additionally store the new reference face image and the storage time thereof, and then proceeds to step S1-2-11.
In step S1-2-9, the update control unit 19 sets the current face image as a new reference face image corresponding to the current face orientation, sets the current time as the storage time of the new reference face image, controls the reference image storage unit 1 to additionally store the new reference face image and the storage time thereof, and then proceeds to step S1-2-10.
In step S1-2-10, the update control section 19 controls the reference image deletion section to delete the reference face image corresponding to the current face orientation and stored for the earliest time in the reference image storage section, and then proceeds to step S1-2-11.
In step S1-2-11, the update control unit 19 controls the update recording unit 13 to record the current time as the update time of the set of reference face images, and then enters the end state.
Operation and Effect of embodiment one
According to the face recognition device and the face recognition method according to the present embodiment, since the reference image storage section stores a plurality of sets of reference face images including the recognition target, the reference left face image, the reference right face image, the reference look-up face image, and the reference face images of the reference look-down face image, the similarity calculation section sequentially compares the current face image with each set of reference face images in similarity, the angle deviation calculation section sequentially compares the current face image with each set of reference face images in angle, the comprehensive similarity calculation section obtains the comprehensive similarity between the current face image and each set of reference face images according to a predetermined calculation rule, and the recognition target determination section determines the corresponding identity information as the identity information of the current recognition target according to the highest comprehensive similarity, so that the face recognition device of the present embodiment can perform similarity calculation for different reference images and calculate the highest comprehensive similarity, and perform information determination through the highest comprehensive similarity and the similarity threshold, thereby reducing the influence of the face orientation of the current face image on the recognition result with high accuracy.
In addition, the update control part controls the reference image update part to update the corresponding reference face image according to the current face image, so that the face recognition device of the embodiment can update the reference face image regularly, and the situation that the face recognition cannot be performed due to the face change of the face of the recognition object is avoided.
And the updating control part controls the reference image updating part to update the reference face image when the similarity judging part judges that the highest comprehensive similarity of the current face image is 80-90%, the reference similarity judging part judges that the minimum value in the current reference similarity is smaller than the average value of the reference similarity and the updating interval judging part judges that the interval between the latest updating time of the current reference image group and the current time is not in a preset time interval, thereby ensuring that the current reference face image group has higher image quality and ensuring that the reference face image can be updated regularly.
In addition, the face direction determination unit can determine the face direction of the current face image as the current face direction; the update control part can take the current face image as a new reference face image corresponding to the current face orientation, and control the reference image storage part to additionally store the new reference face image and the storage time thereof, so that the embodiment can judge the face orientation of the current face image and correspondingly add the face orientation to the corresponding current reference face image group, thereby ensuring that the image characteristics of the five original different face orientations in the current reference face image group are still maintained.
In addition, the face orientation of the reference front face image is the front, the face orientation in the reference left face image is the left side, the face orientation in the reference right face image is the right side, the face orientation in the reference bottom face image is the upper side, and the face orientation in the reference top face image is the lower side, so each group of reference face images of the present embodiment has the face orientations of five directions, so that when the current face image is acquired, the identification object can be identified without particularly limiting the face orientation of the identification object, and the identification accuracy is high.
In addition, the threshold updating control unit controls the threshold accuracy calculating unit to calculate the accuracy value of each preset similarity threshold according to the preset threshold updating time point, and controls the similarity threshold storing unit to store the preset similarity threshold with the highest accuracy value as the similarity threshold with the comprehensive similarity comparison judging unit as the judging reference, so that the similarity threshold can be updated regularly, and the face recognition device of the embodiment has a higher preparation rate when performing face recognition.
< example two >
In comparison with the first embodiment, the second embodiment is provided with the same reference numerals for the constituent elements having the same configuration as the first embodiment, and the corresponding description is omitted.
Fig. 6 is a block diagram of a face recognition device according to a second embodiment of the present invention.
As shown in fig. 6, in the present embodiment, the face recognition apparatus 200 is an apparatus for recognizing a face based on a plurality of captured face images, and specifically includes a reference image storage unit 1, an imaging unit 2, a similarity calculation unit 3, an angle deviation calculation unit 4, a comprehensive similarity calculation unit 5, a highest comprehensive similarity acquisition unit 26, an identity information acquisition unit 27, an information judgment unit 28, a sharpness calculation unit 29, a comprehensive sharpness calculation unit 30, a comprehensive similarity comparison judgment unit 6, a recognition object judgment unit 7, a screen storage unit 8, an output display unit 9, an output display control unit 10, a similarity judgment unit 11, a current time acquisition unit 12, an update recording unit 13, an update interval judgment unit 14, a reference similarity calculation unit 15, a reference similarity judgment unit 16, a reference image update unit 17, a reference image deletion unit 18, an update control unit 19, a preset threshold storage unit 20, a threshold accuracy calculation unit 21, a similarity threshold storage unit 22, a threshold update control unit 23, a communication unit 24, and a control unit 25.
The image capturing portion 2 is configured to acquire face images of the recognition object in the current state for multiple times to obtain multiple current face images, and an action flow of each face image acquisition is the same as that of the first embodiment, and is not described herein again. In the present embodiment, the image pickup section 2 performs five face collection image collection of the recognition object in the current state to obtain 5 current face images.
For each collected current face image, the similarity calculating part 3 is used for calculating the similarity between the current face image and each group of reference images, and the angle deviation calculating part 4 is used for calculating the angle deviation between the current face image and each group, and the specific process is the same as that of the first embodiment, and is not repeated here.
The highest integrated similarity obtaining unit 26 is configured to obtain the highest integrated similarity according to the plurality of integrated similarities calculated by the integrated similarity calculating unit 5, as the highest integrated similarity corresponding to each current face image.
The identity information acquiring unit 27 is configured to acquire corresponding identity information according to the highest comprehensive similarity corresponding to each current face image.
The information judgment section 28 is for judging whether or not the plurality of pieces of identity information respectively corresponding to the plurality of current face images and acquired by the identity information acquisition section 27 correspond to different recognition objects.
The sharpness calculation unit 29 is configured to perform image analysis on a plurality of current face images to obtain a plurality of sharpness when the plurality of pieces of identity information acquired by the identity information acquisition unit 27 correspond to different recognition objects.
The integrated definition calculating unit 30 is configured to calculate corresponding integrated definition for different recognition objects, and includes a definition coefficient storage unit, a definition coefficient acquisition unit, and an integrated integration unit.
The definition coefficient storage unit stores a plurality of image definition and corresponding definition coefficients.
The definition coefficient obtaining unit sequentially retrieves and obtains the corresponding definition coefficient from the definition coefficient storage unit according to the definition of each current face image.
The comprehensive accumulation unit sequentially accumulates the definition coefficients of the current face image corresponding to each different recognition object to obtain the comprehensive definition of each recognition object. For example, the number of the current face images is 5, wherein two pieces of identity information correspond to staff a, the definition coefficients are respectively 0.1 and 0.3, the other three pieces of identity information correspond to staff B, the definition is respectively 0.1, 0.3 and 0.2, the comprehensive definition corresponding to staff a is 0.4, and the comprehensive definition corresponding to staff B is 0.6.
The recognition object determining unit 7 determines that the corresponding recognition object is the current recognition object based on the highest overall sharpness.
The screen storage section 8 stores a plurality of identity information display screens 8b and stranger presenting screens 8a, and the configuration of the identity information display screens 8b and the stranger presenting screens 8a is the same as that of the first embodiment, and will not be described again here.
The output display unit 9 displays the screen.
The output display control unit 10 is configured to control the output display unit 9 to display the above-mentioned picture, and the specific implementation process is the same as that of the first embodiment, and will not be described herein.
The communication unit 24 is used for data exchange between the respective constituent parts of the face recognition device 200.
The control unit 25 is configured to control operations of the respective constituent elements of the face recognition device 200.
After a plurality of current face images are collected, the face recognition device 200 of this embodiment performs similarity analysis on each current face image and each group of reference face images, calculates the corresponding highest comprehensive similarity, determines whether the plurality of highest comprehensive similarities correspond to different identity information, and when the plurality of highest comprehensive similarities correspond to the different identity information, calculates the comprehensive definition of the current face image corresponding to the same identity information, and uses the current face image corresponding to the final comprehensive definition as the current face image corresponding to the current recognition object.
Fig. 7 is a flowchart illustrating operations of the face recognition apparatus according to the second embodiment of the present invention.
As shown in fig. 7, in the second embodiment, the operation flow of the face recognition device 200 includes the following steps:
In step S2-1, the image pickup unit 2 acquires face images of the recognition object in the current state a plurality of times to obtain a plurality of current face images, and then the process proceeds to step S2-2.
In step S2-2, the control unit 25 controls the similarity calculation unit 3, the angle deviation calculation unit 4, and the integrated similarity calculation unit 5 to perform corresponding operations on the plurality of current face images to obtain integrated similarity between each current face image and each group of reference face images, and then proceeds to step S2-3.
In step S2-3, the control unit 25 controls the highest integrated similarity obtaining unit 26 to obtain the highest integrated similarity for each current face image based on the plurality of integrated similarities for each current face image as the highest integrated similarity corresponding to each current face image, and then proceeds to step S2-4.
In step S2-4, the control unit 25 controls the identity information acquiring unit 27 to acquire corresponding identity information for each of the plurality of current face images based on the highest overall similarity corresponding to each of the plurality of current face images, and then proceeds to step S2-5.
In step S2-5, the control unit 25 controls the information judgment unit 28 to judge whether or not the plurality of pieces of identification information respectively corresponding to the plurality of pieces of current face images and acquired by the identification information acquisition unit 27 correspond to different identification objects, and proceeds to step S2-6 when judging yes, and proceeds to step S2-10 when judging no.
In step S2-6, the sharpness calculation unit 29 performs image analysis on each of the plurality of current face images to obtain a plurality of sharpness, and then proceeds to step S2-7.
And step S2-7, the definition coefficient obtaining unit sequentially retrieves and obtains the corresponding definition coefficient from the definition coefficient storage unit according to the definition of each current face image, and then the step S2-8 is carried out.
And S2-8, accumulating the definition coefficients of the current face image corresponding to each different recognition object by the comprehensive accumulation unit in sequence to obtain the comprehensive definition of each recognition object, and then entering into the step S2-9.
In step S2-9, the recognition object determining unit 7 determines that the corresponding recognition object is the current recognition object based on the highest overall definition, and then proceeds to step S2-10.
In step S2-10, the integrated similarity comparison/judgment unit 6 compares the highest value of the highest integrated similarity corresponding to the current recognition object with a predetermined similarity threshold value as the current highest integrated similarity, judges whether or not the current highest integrated similarity is equal to or greater than the similarity threshold value, and proceeds to step S2-11 when yes, and proceeds to step S2-13 when no.
S2-11, the identification object judging part 7 judges the corresponding identity information as the current identification object identity information according to the highest comprehensive similarity, and then the step S2-12 is carried out.
In steps S2 to 12, the output display control unit 10 controls the output display unit 9 to display the identity information display screen 8b corresponding to the current identity information of the identification target, and then enters the end state.
In step S2-13, the recognition object determining unit 7 determines that the current recognition object is a stranger, and then proceeds to step S2-14.
In step S2-14, the output display control unit 10 controls the output display unit 9 to display the stranger presenting screen 8a, and then enters the end state.
In the above process, the specific process of acquiring a plurality of current face images each time of face image acquisition in step S2-1 is the same as steps S1-1-1 to S1-1-2 of the first embodiment; the specific process of the comprehensive similarity between the current face image and each group of reference face images in step S2-2 is the same as that of steps S1-1-3 to S1-1-5 in the first embodiment, and will not be described here again.
In this embodiment, the update control unit 19 controls the similarity determination unit 11, the current time acquisition unit 12, the update recording unit 13, the update interval determination unit 14, the reference similarity calculation unit 15, the reference similarity determination unit 16, the reference image update unit 17, the reference image storage unit 1, and the reference image deletion unit 18 to update the current reference face image group in the reference image storage unit by using the current face image having the highest overall similarity as the face image to be updated, and the procedure is the same as in the first embodiment and will not be repeated here.
The actions and effects of the second embodiment
According to the face recognition device and the face recognition method according to the second embodiment, since the image capturing portion captures face images multiple times to obtain multiple current face images, the control portion controls the similarity calculating portion, the angle deviation calculating portion, the comprehensive similarity calculating portion, and the highest comprehensive similarity obtaining portion to perform corresponding operations for each current face image, so as to obtain the highest comprehensive similarity of each current face image as the corresponding highest comprehensive similarity; the control information judging part judges whether the plurality of identity information corresponds to different recognition objects, when judging that the plurality of identity information corresponds to different recognition objects, the definition calculating part can respectively conduct image analysis on a plurality of current face images to obtain a plurality of definitions, the comprehensive definition calculating part calculates corresponding comprehensive definition of different recognition objects according to the plurality of definitions, and the recognition object judging part judges that the recognition object corresponding to the highest comprehensive definition is the current recognition object, so that the second embodiment has the effect of being the same as the first embodiment, can avoid the situation that recognition errors occur due to the fact that the acquired current face images are not clear, improves the accuracy and efficiency of face recognition, and reduces the time spent by the recognition objects in face recognition.
The above embodiments are preferred examples of the present invention, and are not intended to limit the scope of the present invention.

Claims (11)

1. A face recognition apparatus for recognizing a face of a recognition object, comprising:
a reference image storage unit; an imaging unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a comprehensive similarity comparison judgment unit; a recognition object judging unit for judging whether the recognition object is a recognition object,
wherein the reference image storage unit stores a plurality of sets of reference face images corresponding to a plurality of recognition objects, each set of reference face images including a reference front face image, a reference left face image, a reference right face image, a reference bottom face image, and a reference top face image of the recognition object,
the image pick-up part acquires the face image of the identification object in the current state to obtain the current face image,
the similarity calculation unit compares the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image in order to obtain a plurality of face similarities, left-side similarities, right-side similarities, look-up similarities, and look-down similarities corresponding to each group of the reference face images,
The angle deviation calculating unit compares the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image, and the reference look-down face image in order to obtain a plurality of face deviation angles, left deviation angles, right deviation angles, look-up deviation angles, and look-down deviation angles corresponding to the reference face images,
the comprehensive similarity calculation unit calculates a plurality of comprehensive similarities between the current face image and each of the reference face images based on the forward face similarity, the left side similarity, the right side similarity, the look-up similarity, the top view similarity, and the corresponding forward face deviation angle, left side deviation angle, right side deviation angle, look-up deviation angle, and top view deviation angle of each of the groups according to a predetermined calculation rule,
the integrated similarity comparison and judgment section compares a highest integrated similarity, which is the highest integrated similarity, with a predetermined similarity threshold value, judges whether or not the highest integrated similarity is equal to or greater than the similarity threshold value,
When the highest integrated similarity is equal to or greater than the similarity threshold, the recognition object judging part judges corresponding identity information as the current identity information of the recognition object according to the highest integrated similarity,
the predetermined calculation rule includes a similarity coefficient assigning rule and a comprehensive similarity calculation rule,
the similarity coefficient assignment rule is as follows: calculating a set of deviation degrees of the front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the look-down deviation angle, calculating a corresponding set of sequence values according to the set of deviation degrees, further calculating a set of similarity coefficients corresponding to the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity according to the set of sequence values,
the comprehensive similarity calculation rule is to multiply the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity with the corresponding similarity coefficients respectively and then add the multiplied similarity coefficients to obtain the comprehensive similarity.
2. The face recognition device of claim 1, further comprising:
An update recording unit, an update control unit, a similarity determination unit, a current time acquisition unit, an update interval determination unit, a reference similarity calculation unit, a reference similarity determination unit, and a reference image update unit,
wherein the update recording section stores an update time of each set of the reference face image,
once the identification object determining section determines corresponding identity information based on the highest integrated similarity of the current face image, the update control section uses a corresponding set of the reference face images as a current reference face image set, controls the similarity determining section to determine whether the highest integrated similarity between the current face image and the current reference face image set is between 80% and 90%,
when it is determined to be between 80% and 90%, the update control section controls the current time acquisition section to acquire a current time, and controls the update interval determination section to determine whether or not an interval between a latest update time corresponding to the current reference face image group stored in the update recording section and the current time is a predetermined time interval,
when it is determined to be greater than, the update control section controls the reference similarity calculation section to calculate the current reference similarity between the current face image and each of the reference face images in the current reference face image group and calculate a plurality of reference similarities between each pair of the reference face images in the current reference face image group, respectively, further controls the reference similarity determination section to determine whether or not a minimum value of the current reference similarities is smaller than an average value of the reference similarities,
When the update control unit determines that the current face image is smaller than the reference face image, the update control unit controls the reference image update unit to perform an update operation on the corresponding reference face image stored in the reference image storage unit according to the current face image, and controls the update recording unit to record update times of the group of reference face images correspondingly.
3. The face recognition device of claim 2, further comprising:
a reference image deletion section for deleting the reference image,
wherein each set of the reference face images in the reference image storage section includes at least one reference front face image, at least one reference left side face image, at least one reference right side face image, at least one reference bottom face image, and at least one reference top face image,
the reference image updating section includes an orientation determination unit and a number determination unit,
the orientation determination unit determines a face orientation of the current face image as a current face orientation according to the positive face deviation angle, the left side deviation angle, the right side deviation angle, the upward vision deviation angle, and the downward view deviation angle corresponding to the current reference face image group;
the number judging unit judges whether the number of the reference face images of the same orientation as the current face orientation in the current reference face image group is smaller than a predetermined number,
When the update control unit determines that the current face image is smaller than the predetermined threshold value, the update control unit sets the current face image as a new reference face image corresponding to the current face orientation, sets the current time as a storage time of the new reference face image, controls the reference image storage unit to additionally store the new reference face image and the storage time thereof,
when it is determined that the current face image is not smaller than the current face image, the update control unit sets the current face image as a new reference face image corresponding to the current face orientation, sets the current time as a storage time of the new reference face image, controls the reference image storage unit to additionally store the new reference face image and the storage time thereof, and controls the reference image deletion unit to delete the reference face image which corresponds to the current face orientation and has the earliest storage time in the reference image storage unit.
4. The face recognition apparatus according to claim 1, wherein:
the similarity coefficient assigning rule assigning process comprises the following steps:
first, the positive face deviation angle, the left side deviation angle, the right side deviation angle corresponding to each group of the reference face images are calculated in the form of coordinates The side deviation angle, the bottom deviation angle, and the top deviation angle are respectively expressed as (x) 0 ,y 0 )、(x 1 ,y 1 )、(x 2 ,y 2 )、(x 3 ,y 3 ) (x) 4 ,y 4 ) Thereby calculating the group of deviation degrees of the face deviation angle, the left deviation angle, the right deviation angle, the bottom deviation angle and the overlook deviation angle respectively as followsAnd +.>Wherein x is 0 、x 1 、x 2 、x 3 、x 4 A deviation angle, y, in the horizontal direction of the front face of the current face image and the front face of the reference face image, the reference left face image, the reference right face image, the reference look-up face image, and the front face of the reference look-down face image, respectively 0 、y 1 、y 2 、y 3 、y 4 Respectively the deviation angles of the right front of the face of the current face image and the right front of the face of the reference face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in the vertical direction,
then, adding a fixed value of 1e-4 to the deviation degree and obtaining a corresponding group of sequence values respectively as And +.>
Finally, calculating to obtain the corresponding positionThe group of similarity coefficients corresponding to the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity are respectively And +.>Wherein s=w 0 +w 1 +w 2 +w 3 +w 4 。
5. The face recognition apparatus according to claim 1, wherein:
wherein the image pick-up part comprises a camera, an image pick-up control unit, a true face judging unit and an output unit,
once the camera shoots a face image, the camera shooting control unit controls the face judging unit to judge whether the face image is a face or not, and further controls the output unit to output the face image as the current face image when judging that the face image is a face.
6. The face recognition apparatus according to claim 1, wherein:
wherein the face orientation of the reference frontal face image is the frontal direction, the face orientation of the reference left face image is the left side, the face orientation of the reference right face image is the right side, the face orientation of the reference look-up face image is the upper direction, the face orientation of the reference look-down face image is the lower direction,
the angular deviation of the face orientation in the reference left face image and the face orientation of the reference front face image and the angular deviation of the face orientation of the reference right face image and the face orientation of the reference front face image are 10 degrees to 30 degrees,
The angular deviation of the face orientation in the reference upward face image and the face orientation of the reference forward face image and the angular deviation of the face orientation of the reference upward face image and the face orientation of the reference forward face image are 10-20 degrees.
7. The face recognition device of claim 1, further comprising:
a preset threshold value storage part, a threshold value updating control part, a threshold value accuracy calculating part and a similarity threshold value storage part,
wherein the preset threshold storage part stores a plurality of preset similarity thresholds,
the threshold updating control part controls the threshold accuracy calculating part to calculate an accurate value of each preset similarity threshold according to a preset threshold updating time point based on an accurate value calculating rule for the preset similarity threshold and the plurality of groups of reference face images stored by the reference image storing part, and controls the similarity threshold storing part to store the preset similarity threshold with the highest accurate value as the similarity threshold with the comprehensive similarity comparison judging part as a judging reference.
8. The face recognition apparatus according to claim 7, wherein:
Wherein, the accurate value calculation rule is:
firstly, traversing all the reference face images in the reference image storage part, pairing the reference face images belonging to the same group in pairs to form a plurality of same group pairs, pairing the reference face images not belonging to the same group in pairs to form a plurality of different group pairs,
then, the similarity between the two reference face images of each same group pair and each different group pair is calculated,
finally, calculating the accurate value a of each preset similarity threshold value as follows:wherein a is 1 A being the logarithm of the same group of pairs having a similarity above the preset similarity threshold 2 A being the logarithm of the outlier pair having a similarity below the preset similarity threshold 3 Is the sum of the logarithm of the same group of pairs and the logarithm of the different group of pairs.
9. A face recognition apparatus for recognizing a face of a recognition object, comprising:
a reference image storage unit; an imaging unit; a control unit; a similarity calculation unit; an angle deviation calculation unit; a comprehensive similarity calculation unit; a highest comprehensive similarity acquisition unit; an identity information acquisition unit; an information judgment unit; a definition calculating unit; a comprehensive definition calculation unit; a recognition object judging unit for judging whether the recognition object is a recognition object,
Wherein the reference image storage unit stores a plurality of sets of reference face images corresponding to a plurality of recognition objects, each set of reference face images including a reference front face image, a reference left face image, a reference right face image, a reference bottom face image, and a reference top face image of the recognition object,
the image pick-up part acquires face images of the identification object in the current state for a plurality of times to obtain a plurality of current face images,
the control part controls the similarity calculation part to sequentially compare the current face image with the reference front face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of the reference face images according to each current face image to obtain a plurality of front face similarity, left side similarity, right side similarity, look-up similarity and look-down similarity which correspond to each group of the reference face images respectively; controlling the angle deviation calculating part to respectively perform angle comparison on the current face image and the reference positive face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images in sequence to obtain a plurality of positive face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which respectively correspond to each group of reference face images; controlling the comprehensive similarity calculation part to calculate a plurality of comprehensive similarity according to a preset calculation rule according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity and the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group; controlling the highest comprehensive similarity obtaining part to obtain the highest comprehensive similarity as the highest comprehensive similarity according to the plurality of comprehensive similarities; further controlling the identity information acquisition part to acquire corresponding identity information according to the highest comprehensive similarity,
The control section further controls the information judging section to judge whether or not the plurality of pieces of the identification information respectively corresponding to the plurality of current face images and acquired by the identification information acquiring section correspond to different recognition objects,
when the image is judged to correspond to different recognition objects, the definition calculating part performs image analysis on the plurality of current face images respectively to obtain a plurality of resolutions,
the integrated definition calculating section calculates respective integrated definitions for the different recognition objects based on the plurality of definitions,
the identification object determination unit determines that the identification object corresponding to the highest integrated sharpness is the current identification object.
10. The face recognition apparatus according to claim 9, wherein:
wherein the integrated sharpness calculation section includes:
a definition coefficient storage unit for storing a plurality of image definition and corresponding definition coefficients;
a definition coefficient obtaining unit, configured to retrieve and obtain a corresponding definition coefficient from the definition coefficient storage unit according to the definition of each current face image in sequence;
and the comprehensive accumulation unit is used for accumulating the definition coefficients of the current face image corresponding to each different recognition object in sequence to obtain the comprehensive definition of each recognition object.
11. A face recognition method for recognizing a face of a recognition object, comprising the steps of:
a reference image storage step of acquiring and storing a plurality of groups of reference face images respectively corresponding to a plurality of recognition objects, wherein each group of reference face images comprises a reference front face image, a reference left face image, a reference right face image, a reference look-up face image and a reference overlook face image of the recognition object;
a camera shooting step, namely acquiring a face image of the identification object in the current state to obtain a current face image;
a similarity calculation step, namely comparing the current face image with the reference face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images in sequence to obtain a plurality of face similarity, left similarity, right similarity, look-down similarity and look-down similarity which correspond to each group of reference face images respectively;
an angle deviation calculating step of comparing the current face image with the reference front face image, the reference left face image, the reference right face image, the reference look-up face image and the reference look-down face image in each group of reference face images in sequence to obtain a plurality of front face deviation angles, left side deviation angles, right side deviation angles, look-up deviation angles and look-down deviation angles which correspond to each group of reference face images;
A comprehensive similarity calculation step, according to a preset calculation rule, calculating a plurality of comprehensive similarity between the current face image and each group of reference face image according to the front face similarity, the left side similarity, the right side similarity, the look-up similarity, the overlook similarity and the corresponding front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the overlook deviation angle of each group;
a comprehensive similarity comparison and judgment step, wherein the highest comprehensive similarity is used as the highest comprehensive similarity to be compared with a preset similarity threshold value, and whether the highest comprehensive similarity is larger than or equal to the similarity threshold value is judged;
a recognition object judging step of judging corresponding identity information as the current identity information of the recognition object according to the highest comprehensive similarity when the highest comprehensive similarity is greater than or equal to the similarity threshold,
the predetermined calculation rule includes a similarity coefficient assigning rule and a comprehensive similarity calculation rule,
the similarity coefficient assignment rule is as follows: calculating a set of deviation degrees of the front face deviation angle, the left side deviation angle, the right side deviation angle, the look-up deviation angle and the look-down deviation angle, calculating a corresponding set of sequence values according to the set of deviation degrees, further calculating a set of similarity coefficients corresponding to the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity according to the set of sequence values,
The comprehensive similarity calculation rule is to multiply the front face similarity, the left side similarity, the right side similarity, the look-up similarity and the look-down similarity with the corresponding similarity coefficients respectively and then add the multiplied similarity coefficients to obtain the comprehensive similarity.
CN201811367426.5A 2018-11-16 2018-11-16 Face recognition device and face recognition method Active CN111199029B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811367426.5A CN111199029B (en) 2018-11-16 2018-11-16 Face recognition device and face recognition method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811367426.5A CN111199029B (en) 2018-11-16 2018-11-16 Face recognition device and face recognition method

Publications (2)

Publication Number Publication Date
CN111199029A CN111199029A (en) 2020-05-26
CN111199029B true CN111199029B (en) 2023-07-18

Family

ID=70746096

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811367426.5A Active CN111199029B (en) 2018-11-16 2018-11-16 Face recognition device and face recognition method

Country Status (1)

Country Link
CN (1) CN111199029B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112001932B (en) * 2020-09-01 2023-10-31 腾讯科技（深圳）有限公司 Face recognition method, device, computer equipment and storage medium
CN112199998B (en) * 2020-09-09 2023-06-20 浙江大华技术股份有限公司 Face recognition method, device, equipment and medium
CN113344132A (en) * 2021-06-30 2021-09-03 成都商汤科技有限公司 Identity recognition method, system, device, computer equipment and storage medium
CN113255631B (en) * 2021-07-15 2021-10-15 浙江大华技术股份有限公司 Similarity threshold updating method, face recognition method and related device
CN114579786B (en) * 2021-12-31 2025-06-27 飞诺门阵(北京)科技有限公司 Face image comparison method, device, electronic device and storage medium
CN115937950A (en) * 2022-12-16 2023-04-07 阿波罗智联(北京)科技有限公司 Multi-angle face data acquisition method, device, equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102013011A (en) * 2010-12-16 2011-04-13 重庆大学 Front-face-compensation-operator-based multi-pose human face recognition method
CN102799877A (en) * 2012-09-11 2012-11-28 上海中原电子技术工程有限公司 Method and system for screening face images
CN107545252A (en) * 2017-08-31 2018-01-05 北京图铭视界科技有限公司 Face identification method and device in video based on multi-pose Face model
CN108647651A (en) * 2018-05-14 2018-10-12 深圳市科发智能技术有限公司 A kind of face identification method, system and device improving the rate that is identified by

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5031641B2 (en) * 2008-03-31 2012-09-19 富士通株式会社 Pattern alignment method, verification method, and verification device
JP6351243B2 (en) * 2013-11-28 2018-07-04 キヤノン株式会社 Image processing apparatus and image processing method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102013011A (en) * 2010-12-16 2011-04-13 重庆大学 Front-face-compensation-operator-based multi-pose human face recognition method
CN102799877A (en) * 2012-09-11 2012-11-28 上海中原电子技术工程有限公司 Method and system for screening face images
CN107545252A (en) * 2017-08-31 2018-01-05 北京图铭视界科技有限公司 Face identification method and device in video based on multi-pose Face model
CN108647651A (en) * 2018-05-14 2018-10-12 深圳市科发智能技术有限公司 A kind of face identification method, system and device improving the rate that is identified by

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
王亚南 ; 苏剑波 ; .基于图像合成的多姿态人脸图像识别方法.模式识别与人工智能.2015,(09),第82-90页. *
王磊 ; 胡超 ; 吴捷 ; 贺庆 ; 刘伟 ; .基于多相机的人脸姿态识别.计算机应用.2010,(12),第173-176页. *

Also Published As

Publication number Publication date
CN111199029A (en) 2020-05-26

Similar Documents

Publication Publication Date Title
CN111199029B (en) Face recognition device and face recognition method
CN111862296B (en) Three-dimensional reconstruction method, three-dimensional reconstruction device, three-dimensional reconstruction system, model training method and storage medium
CN105874474B (en) System and method for face representation
CN104221054B (en) Character attribute deduction system and study data generating device
CN107944378A (en) The personal identification method and self-help serving system of a kind of Self-Service
EP2620896A2 (en) System And Method For Face Capture And Matching
CN109635620B (en) Self-service terminal (SST) face authentication process
CN106780814A (en) A kind of whole-process automatic attendance checking system used in classroom
JP6822482B2 (en) Line-of-sight estimation device, line-of-sight estimation method, and program recording medium
CN108564673A (en) A kind of check class attendance method and system based on Global Face identification
US20190286798A1 (en) User authentication method using face recognition and device therefor
WO2022217784A1 (en) Data processing methods and apparatus, device, and medium
CN108596135A (en) Personal identification method and system
CN111325051A (en) A face recognition method and device based on face image ROI selection
WO2020172870A1 (en) Method and apparatus for determining motion trajectory of target object
TW201719477A (en) Method, apparatus, and system for detecting living body based on eyeball tracking
CN113792662B (en) Image detection method, device, electronic equipment and storage medium
CN104978734B (en) The extracting method and device of foreground image
CN110516649A (en) Alumnus&#39;s authentication method and system based on recognition of face
US10891561B2 (en) Image processing for item recognition
CN105913014A (en) Identity card chip photograph and video-based identity authenticating method and system
CN107256388A (en) A kind of method and device for obtaining front face image
CN114511915B (en) A mobile client-based trusted certificate photo collection system and method
CN111767845B (en) Certificate identification method and device
CN114462646B (en) Pole number plate identification method and system based on contact network safety inspection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant