CN112801013A - Face recognition method, system and device based on key point recognition and verification - Google Patents

Face recognition method, system and device based on key point recognition and verification Download PDF

Info

Publication number
CN112801013A
CN112801013A CN202110170397.9A CN202110170397A CN112801013A CN 112801013 A CN112801013 A CN 112801013A CN 202110170397 A CN202110170397 A CN 202110170397A CN 112801013 A CN112801013 A CN 112801013A
Authority
CN
China
Prior art keywords
image
face
recognized
recognition
attribute characteristic
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.)
Granted
Application number
CN202110170397.9A
Other languages
Chinese (zh)
Other versions
CN112801013B (en
Inventor
王涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dilu Technology Co Ltd
Original Assignee
Dilu Technology 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 Dilu Technology Co Ltd filed Critical Dilu Technology Co Ltd
Priority to CN202110170397.9A priority Critical patent/CN112801013B/en
Publication of CN112801013A publication Critical patent/CN112801013A/en
Application granted granted Critical
Publication of CN112801013B publication Critical patent/CN112801013B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161—Detection; Localisation; Normalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation

Abstract

The invention discloses a face recognition method, a system and a device based on key point recognition check, when an image to be recognized of a target object is collected, the image to be recognized is obtained by judging whether the target object is a living body or not based on a face sample image, the image to be recognized and corresponding attribute characteristic values thereof are stored in the face sample image, the face contour is recognized, the attribute characteristic values of the image to be recognized are extracted according to the face contour, the attribute characteristic values of the image to be recognized and the corresponding attribute characteristic values in each face sample image are compared, a face attribute characteristic value comparison result is obtained, further evaluation indexes of the image to be recognized are obtained, and the face recognition of the image to be recognized is completed. By the technical scheme, the living body identification efficiency is improved, the human intervention input is reduced, the face identification method is optimized, and the quality efficiency of the living body identification in the prior art is met.

Description

Face recognition method, system and device based on key point recognition and verification
Technical Field
The invention relates to the field of face recognition, in particular to a face recognition method, a face recognition system and a face recognition device based on key point recognition verification.
Background
Face recognition, it is a biological identification technology based on the face characteristic information of people to carry on the identification, gather the picture or video stream containing human face with the camera or lens, and detect and track the human face in the picture automatically, and then carry on a series of relevant technologies of face recognition to the human face detected, traditional face recognition technology is mainly based on the face recognition of the visible light image, this is a familiar identification mode, there has been more than 30 years of development history, but this kind of mode has the defect that is difficult to overcome, especially when the ambient light changes, the recognition effect can drop sharply, can't meet the needs of the actual system, the existing face recognition technology generally includes four flows: the method comprises the following steps of face acquisition, image preprocessing, image feature extraction and face matching identification, but the following defects exist in the existing face identification technology:
1. in the existing face recognition method, when face recognition comparison is carried out, the comparison workload is huge, the manpower investment is large, the face recognition process is repeatedly operated, the repeated work is more, and the efficiency is low;
2. in the process of face recognition, recognition errors are easy to occur, the face recognition result is influenced, errors exist in the face recognition result, and the accuracy is low.
Disclosure of Invention
The invention aims to provide a face recognition method, a face recognition system and a face recognition device based on key point recognition and verification, and aims to solve the problems in the prior art.
In order to achieve the purpose, the invention provides the following technical scheme:
when an image to be recognized of a target object is acquired, carrying out face recognition of the target object based on a face sample image, and executing the following steps:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when the target object corresponding to the face image is a living body, then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, according to the attribute characteristic value of the image to be recognized extracted from the face contour, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result, further obtaining an evaluation index of the image to be recognized, wherein the evaluation index comprises the accuracy, the error rate and the recall rate of the extracted face characteristic information, and then entering the step E;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic values of the image to be recognized with the corresponding attribute characteristic values in the face sample images, obtaining a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, carrying out face recognition on the image to be recognized, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing the face recognition, and exiting the recognition state.
Preferably, in the step a, the living body recognition is performed twice on the image to be recognized, and when the accuracy of the living body recognition reaches a preset correct threshold, the image to be recognized is determined to be a living body, so as to obtain a living body recognition result;
when first living body recognition is carried out, taking the pixel value of an image to be recognized as a living body judgment array, and acquiring a first characteristic value corresponding to the image to be recognized;
and when the living body is identified for the second time, reducing the pixel value multiple of the image to be identified to be used as a living body judgment array, and acquiring a second characteristic value corresponding to the image to be identified.
Preferably, the target object corresponding to the image to be recognized is recognized whether to be a living body, each pixel point of the image to be recognized is respectively compared with the pixel point corresponding to the second eigenvalue and the pixel point corresponding to the face image in the face sample image one by one, when the similarity difference value between each pixel point of the image to be recognized and the pixel point corresponding to the second eigenvalue and the pixel point corresponding to the face image in the face sample image is smaller than or equal to the preset similarity, whether the recognition of the living body is completed is judged, and then the step B is executed.
Preferably, in step D, when a pixel difference between the image to be recognized and the image stored in the face sample image and between the face feature attribute value and the corresponding attribute feature value in the face sample image is less than or equal to a preset pixel threshold, a face attribute feature value comparison result is obtained, and an evaluation index of the image to be recognized is further obtained, where in the evaluation index, the accuracy, the error rate, and the recall rate are calculated according to the following formulas:
Figure BDA0002938709840000031
a, E, R represents the accuracy, error rate and recall rate of the evaluation index, N represents the number of images to be identified, and T represents the number of images to be identifiedpFor the number of faces in the image to be recognized, TnAs the number of non-faces in the image to be recognized, FpAs the number of non-faces in the number of faces, i.e. the number of errors in the face, FnThe number of faces in the number of non-faces, i.e. the number of errors in the non-faces.
Preferably, in the step E, a complete face contour is simulated by a convolutional neural network, so as to obtain a corrected image to be recognized.
According to a second aspect of the disclosure, a face recognition system based on keypoint recognition and verification is further provided, which is characterized by comprising:
one or more processors;
a memory storing instructions that are operable, when executed by the one or more processors, to cause the one or more processors to perform operations comprising:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when the target object corresponding to the face image is a living body, then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, according to the attribute characteristic value of the image to be recognized extracted from the face contour, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result, further obtaining an evaluation index of the image to be recognized, wherein the evaluation index comprises the accuracy, the error rate and the recall rate of the extracted face characteristic information, and then entering the step E;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic values of the image to be recognized with the corresponding attribute characteristic values in the face sample images, obtaining a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, carrying out face recognition on the image to be recognized, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing the face recognition, and exiting the recognition state.
According to a third aspect of the present invention, there is also provided a computer-readable medium storing software, the software including instructions executable by one or more computers, the instructions, when executed by the one or more computers, performing the operations of the face recognition method according to any one of claims 1 to 5.
According to the fourth aspect of the present invention, a face recognition apparatus based on keypoint recognition verification is further provided, which includes:
the living body judging module is used for acquiring human face images of living body target objects in a preset quantity and preset and appointed attribute characteristic values corresponding to the human face images respectively aiming at the human face sample images, and judging whether the target object is a living body or not aiming at the target object corresponding to the human face images;
the image to be recognized acquisition module is used for acquiring a face image aiming at a target object, acquiring the image to be recognized and storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into a face sample image;
the face contour recognition module is used for recognizing the face contour of the image to be recognized;
the evaluation index acquisition module is used for extracting the attribute characteristic value of the image to be identified according to the face contour, comparing the attribute characteristic value of the image to be identified with the corresponding attribute characteristic value in each face sample image, acquiring a face attribute characteristic value comparison result and further acquiring the evaluation index of the image to be identified;
the simulation correction module is used for carrying out simulation correction on the image to be recognized by utilizing a neural network aiming at the image to be recognized which cannot obtain the complete face contour to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result and further obtaining an evaluation index of the image to be recognized;
and the face recognition module is used for storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing face recognition and quitting the recognition state.
Compared with the prior art, the face recognition method based on the key point recognition and verification has the following technical effects by adopting the technical scheme:
the invention overcomes the defects of living body identification in the existing face identification method, improves the screening accuracy and identifies the living body target object when screening non-living body identification through two times of living body identification, and simultaneously screens and optimizes the face identification method process according to the evaluation index of the image, thereby improving the living body face identification efficiency, reducing the human intervention input, optimizing the face identification method, meeting the quality efficiency of the living body identification in the prior art, and being capable of being applied to the actual production in a large scale.
Drawings
FIG. 1 is a flow chart of a face recognition method according to an exemplary embodiment of the present invention;
fig. 2 is a block configuration diagram of a face recognition apparatus according to an exemplary embodiment of the present invention.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily defined to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
With reference to the face recognition method shown in fig. 1 according to the exemplary embodiment of the present invention, when an image to be recognized of a target object is acquired, face recognition of the target object is performed based on a face sample image, and the following steps are performed:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when a target object corresponding to the face image is a living body, performing living body recognition twice on the image to be recognized, and when the accuracy of the living body recognition reaches a preset correct threshold value, determining the image to be recognized as the living body to obtain a living body recognition result;
when first living body recognition is carried out, taking the pixel value of the image to be recognized as a living body judgment array, wherein the size of the living body judgment array is 1280 × 720 × 3, and acquiring a first characteristic value corresponding to the image to be recognized;
when the second living body recognition is carried out, the pixel value multiple of the image to be recognized is reduced to be used as a living body judgment array, and the size of the living body judgment array is 256 × 3 to obtain a second characteristic value corresponding to the image to be recognized;
comparing each pixel point of the image to be recognized with a pixel point corresponding to the second eigenvalue and a pixel point corresponding to the face image in the face sample image one by one, finishing the recognition of whether the image is a living body when the similarity difference value between each pixel point of the image to be recognized and the pixel point corresponding to the second eigenvalue and the pixel point corresponding to the face image in the face sample image is less than or equal to the preset similarity, and then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, extracting an attribute characteristic value of the image to be recognized according to the face contour, comparing the attribute characteristic value of the image to be recognized with the attribute characteristic value corresponding to each face sample image, obtaining a face attribute characteristic value comparison result when the pixel difference value between the face characteristic value and the attribute characteristic value corresponding to the face sample image is less than or equal to a preset pixel threshold value, and further calculating according to the following formula to obtain an evaluation index of the image to be recognized:
Figure BDA0002938709840000061
a, E, R represents the accuracy, error rate and recall rate of the evaluation index, N represents the number of images to be identified, and T represents the number of images to be identifiedpFor the number of faces in the image to be recognized, TnAs the number of non-faces in the image to be recognized, FpAs the number of non-faces in the number of faces, i.e. the number of errors in the face, FnThe number of faces in the number of non-faces, namely the number of errors in the non-faces;
the evaluation indexes comprise the accuracy, the error rate and the recall rate of the extracted face feature information, and then the step E is carried out;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network, simulating the complete face contour to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image to obtain a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, carrying out face recognition on the image to be recognized, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing the face recognition, and exiting the recognition state.
According to the embodiment disclosed in the present invention, there is also provided a face recognition system based on keypoint recognition and verification, comprising:
one or more processors;
a memory storing instructions that are operable, when executed by the one or more processors, to cause the one or more processors to perform operations comprising:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when the target object corresponding to the face image is a living body, then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, according to the attribute characteristic value of the image to be recognized extracted from the face contour, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result, further obtaining an evaluation index of the image to be recognized, wherein the evaluation index comprises the accuracy, the error rate and the recall rate of the extracted face characteristic information, and then entering the step E;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic values of the image to be recognized with the corresponding attribute characteristic values in the face sample images, obtaining a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and the face recognition result, then finishing the face recognition and exiting the recognition state.
Particularly preferably, the aforementioned processor is a processor of a computer system, including but not limited to an ARM-based embedded processor, an X86-based microprocessor, or a type-based processor.
The memory is arranged as a carrier that can store data, typically comprising RAM and ROM.
It should be understood that the computer system may communicate with each subsystem through the bus to obtain the corresponding parameters, so as to implement the control of the operation of each subsystem.
In alternative embodiments, the invention may also be configured to be implemented as follows:
with reference to fig. 2, a face recognition apparatus based on keypoint identification and verification includes:
the living body judging module is used for acquiring human face images of living body target objects in a preset quantity and preset and appointed attribute characteristic values corresponding to the human face images respectively aiming at the human face sample images, and judging whether the target object is a living body or not aiming at the target object corresponding to the human face images;
the image to be recognized acquisition module is used for acquiring a face image aiming at a target object, acquiring the image to be recognized and storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into a face sample image;
the face contour recognition module is used for recognizing the face contour of the image to be recognized;
the evaluation index acquisition module is used for extracting the attribute characteristic value of the image to be identified according to the face contour, comparing the attribute characteristic value of the image to be identified with the corresponding attribute characteristic value in each face sample image, acquiring a face attribute characteristic value comparison result and further acquiring the evaluation index of the image to be identified;
the simulation correction module is used for carrying out simulation correction on the image to be recognized by utilizing a neural network aiming at the image to be recognized which cannot obtain the complete face contour to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result and further obtaining an evaluation index of the image to be recognized;
and the face recognition module is used for storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing face recognition and quitting the recognition state.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.

Claims (8)

1. A face recognition method based on key point recognition check is used for carrying out face recognition on a living target object, and is characterized in that when an image to be recognized of the target object is acquired, the face recognition of the target object is carried out based on a face sample image, and the following steps are carried out:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when the target object corresponding to the face image is a living body, then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, according to the attribute characteristic value of the image to be recognized extracted from the face contour, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result, further obtaining an evaluation index of the image to be recognized, wherein the evaluation index comprises the accuracy, the error rate and the recall rate of the extracted face characteristic information, and then entering the step E;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic values of the image to be recognized with the corresponding attribute characteristic values in the face sample images, obtaining a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, carrying out face recognition on the image to be recognized, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing the face recognition, and exiting the recognition state.
2. The face recognition method based on key point recognition and verification according to claim 1, wherein in the step a, the image to be recognized is subjected to living body recognition twice, and when the accuracy of the living body recognition reaches a preset correct threshold, the image to be recognized is determined to be a living body, and a living body recognition result is obtained;
when first living body recognition is carried out, taking the pixel value of an image to be recognized as a living body judgment array, and acquiring a first characteristic value corresponding to the image to be recognized;
and when the living body is identified for the second time, reducing the pixel value multiple of the image to be identified to be used as a living body judgment array, and acquiring a second characteristic value corresponding to the image to be identified.
3. The face recognition method based on the key point recognition check according to claim 2, characterized in that, the target object corresponding to the image to be recognized is recognized as a living body, each pixel point of the image to be recognized is respectively compared with the pixel point corresponding to the second eigenvalue and the pixel point corresponding to the face image in the face sample image one by one, when the similarity difference between each pixel point of the image to be recognized and the pixel point corresponding to the second eigenvalue and the pixel point corresponding to the face image in the face sample image is less than or equal to the preset similarity, whether the recognition of the living body is completed is determined, and then step B is performed.
4. The method according to claim 1, wherein in step D, when a pixel difference between the image to be recognized and the image stored in the face sample image and the attribute feature value corresponding to the face sample image is less than or equal to a preset pixel threshold, a face attribute feature value comparison result is obtained, and an evaluation index of the image to be recognized is further obtained, where in the evaluation index, the accuracy, the error rate, and the recall rate are calculated according to the following formulas:
Figure FDA0002938709830000021
a, E, R represents the accuracy, error rate and recall rate of the evaluation index, N represents the number of images to be identified, and T represents the number of images to be identifiedpFor the number of faces in the image to be recognized, TnAs the number of non-faces in the image to be recognized, FpAs the number of non-faces in the number of faces, i.e. the number of errors in the face, FnThe number of faces in the number of non-faces, i.e. the number of errors in the non-faces.
5. The face recognition method based on the key point recognition and verification of claim 1, wherein in the step E, a complete face contour is simulated through a convolutional neural network to obtain a corrected image to be recognized.
6. A face recognition system based on key point recognition verification is characterized by comprising:
one or more processors;
a memory storing instructions that are operable, when executed by the one or more processors, to cause the one or more processors to perform operations comprising:
step A, acquiring face images of living body target objects in a preset number and preset and appointed attribute characteristic values corresponding to the face images respectively aiming at face sample images, and judging whether the target objects are living bodies or not aiming at the target objects corresponding to the face images;
when the target object corresponding to the face image is a non-living body, quitting the face identification step of the target object;
when the target object corresponding to the face image is a living body, then entering the step B;
b, acquiring a face image aiming at the target object to obtain an image to be recognized, storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into the face sample image, and then entering the step C;
step C, recognizing the face contour of the image to be recognized;
when the recognition result of the face contour can obtain the complete face contour, entering the step D;
when the recognition result of the face contour can not obtain the complete face contour, entering the step E;
when the face contour cannot be obtained according to the face contour recognition result, returning to the step B, and re-collecting the image to be recognized of the target object;
step D, according to the attribute characteristic value of the image to be recognized extracted from the face contour, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result, further obtaining an evaluation index of the image to be recognized, wherein the evaluation index comprises the accuracy, the error rate and the recall rate of the extracted face characteristic information, and then entering the step E;
step E, aiming at the image to be recognized, which cannot obtain the complete face contour, carrying out analog correction on the image to be recognized by utilizing a neural network to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic values of the image to be recognized with the corresponding attribute characteristic values in the face sample images, obtaining a face attribute characteristic value comparison result, further obtaining the evaluation index of the image to be recognized, and then entering the step F;
and F, carrying out face recognition on the image to be recognized, storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing the face recognition, and exiting the recognition state.
7. A computer-readable medium storing software, the software comprising instructions executable by one or more computers, the instructions when executed by the one or more computers performing the operations of the face recognition method as recited in any one of claims 1-5.
8. A face recognition device based on key point recognition and verification is characterized by comprising:
the living body judging module is used for acquiring human face images of living body target objects in a preset quantity and preset and appointed attribute characteristic values corresponding to the human face images respectively aiming at the human face sample images, and judging whether the target object is a living body or not aiming at the target object corresponding to the human face images;
the image to be recognized acquisition module is used for acquiring a face image aiming at a target object, acquiring the image to be recognized and storing the image to be recognized and the attribute characteristic value corresponding to the image to be recognized into a face sample image;
the face contour recognition module is used for recognizing the face contour of the image to be recognized;
the evaluation index acquisition module is used for extracting the attribute characteristic value of the image to be identified according to the face contour, comparing the attribute characteristic value of the image to be identified with the corresponding attribute characteristic value in each face sample image, acquiring a face attribute characteristic value comparison result and further acquiring the evaluation index of the image to be identified;
the simulation correction module is used for carrying out simulation correction on the image to be recognized by utilizing a neural network aiming at the image to be recognized which cannot obtain the complete face contour to obtain a corrected image to be recognized, updating the corrected image to be recognized into the image to be recognized, comparing the attribute characteristic value of the image to be recognized with the corresponding attribute characteristic value in each face sample image, obtaining a face attribute characteristic value comparison result and further obtaining an evaluation index of the image to be recognized;
and the face recognition module is used for storing the current image to be recognized, the attribute characteristic value corresponding to the image to be recognized and a face recognition result, then finishing face recognition and quitting the recognition state.
CN202110170397.9A 2021-02-08 2021-02-08 Face recognition method, system and device based on key point recognition verification Active CN112801013B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110170397.9A CN112801013B (en) 2021-02-08 2021-02-08 Face recognition method, system and device based on key point recognition verification

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110170397.9A CN112801013B (en) 2021-02-08 2021-02-08 Face recognition method, system and device based on key point recognition verification

Publications (2)

Publication Number Publication Date
CN112801013A true CN112801013A (en) 2021-05-14
CN112801013B CN112801013B (en) 2024-04-09

Family

ID=75814701

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110170397.9A Active CN112801013B (en) 2021-02-08 2021-02-08 Face recognition method, system and device based on key point recognition verification

Country Status (1)

Country Link
CN (1) CN112801013B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117456589A (en) * 2023-12-22 2024-01-26 珠海早安科技有限公司 Human face identification access control method based on AI and access control system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105243376A (en) * 2015-11-06 2016-01-13 北京汉王智远科技有限公司 Living body detection method and device
CN107844744A (en) * 2017-10-09 2018-03-27 平安科技（深圳）有限公司 With reference to the face identification method, device and storage medium of depth information
WO2018072102A1 (en) * 2016-10-18 2018-04-26 华为技术有限公司 Method and apparatus for removing spectacles in human face image
WO2019096008A1 (en) * 2017-11-20 2019-05-23 腾讯科技（深圳）有限公司 Identification method, computer device, and storage medium
CN110059546A (en) * 2019-03-08 2019-07-26 深圳神目信息技术有限公司 Vivo identification method, device, terminal and readable medium based on spectrum analysis
CN112149497A (en) * 2020-08-10 2020-12-29 中标慧安信息技术股份有限公司 Operating system safe login method based on face recognition

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105243376A (en) * 2015-11-06 2016-01-13 北京汉王智远科技有限公司 Living body detection method and device
WO2018072102A1 (en) * 2016-10-18 2018-04-26 华为技术有限公司 Method and apparatus for removing spectacles in human face image
CN107844744A (en) * 2017-10-09 2018-03-27 平安科技（深圳）有限公司 With reference to the face identification method, device and storage medium of depth information
WO2019071664A1 (en) * 2017-10-09 2019-04-18 平安科技（深圳）有限公司 Human face recognition method and apparatus combined with depth information, and storage medium
WO2019096008A1 (en) * 2017-11-20 2019-05-23 腾讯科技（深圳）有限公司 Identification method, computer device, and storage medium
CN110059546A (en) * 2019-03-08 2019-07-26 深圳神目信息技术有限公司 Vivo identification method, device, terminal and readable medium based on spectrum analysis
CN112149497A (en) * 2020-08-10 2020-12-29 中标慧安信息技术股份有限公司 Operating system safe login method based on face recognition

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117456589A (en) * 2023-12-22 2024-01-26 珠海早安科技有限公司 Human face identification access control method based on AI and access control system
CN117456589B (en) * 2023-12-22 2024-04-26 珠海早安科技有限公司 Human face identification access control method based on AI and access control system

Also Published As

Publication number Publication date
CN112801013B (en) 2024-04-09

Similar Documents

Publication Publication Date Title
CN108229509B (en) Method and device for identifying object class and electronic equipment
WO2021027336A1 (en) Authentication method and apparatus based on seal and signature, and computer device
CN110728225B (en) High-speed face searching method for attendance checking
CN107958230B (en) Facial expression recognition method and device
CN109871847B (en) OCR recognition method and terminal
CN105787466B (en) A kind of fine recognition methods and system of type of vehicle
CN110807491A (en) License plate image definition model training method, definition detection method and device
CN111639629B (en) Pig weight measurement method and device based on image processing and storage medium
WO2021000832A1 (en) Face matching method and apparatus, computer device, and storage medium
CN111445459A (en) Image defect detection method and system based on depth twin network
WO2015131468A1 (en) Method and system for estimating fingerprint pose
CN113936280B (en) Automatic character recognition system and method for code disc of embedded instrument
CN107786867A (en) Image identification method and system based on deep learning architecture
CN112633221A (en) Face direction detection method and related device
CN114266894A (en) Image segmentation method and device, electronic equipment and storage medium
CN113111716A (en) Remote sensing image semi-automatic labeling method and device based on deep learning
CN112307984A (en) Safety helmet detection method and device based on neural network
CN109752393B (en) Patch resistor model detection method and device based on image characteristics
CN111310837A (en) Vehicle refitting recognition method, device, system, medium and equipment
CN112801013B (en) Face recognition method, system and device based on key point recognition verification
CN111968087B (en) Plant disease area detection method
CN116630809A (en) Geological radar data automatic identification method and system based on intelligent image analysis
CN107944393B (en) Human face nose tip positioning method
CN116958724A (en) Training method and related device for product classification model
CN110751163A (en) Target positioning method and device, computer readable storage medium and electronic equipment

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
CB02 Change of applicant information

Address after: 210000 11th floor, building A1, Huizhi science and Technology Park, 8 Hengtai Road, Nanjing Economic and Technological Development Zone, Nanjing City, Jiangsu Province

Applicant after: DILU TECHNOLOGY Co.,Ltd.

Address before: Building C4, No. 55 Liyuan South Road, Moling Street, Jiangning District, Nanjing City, Jiangsu Province, 211100

Applicant before: DILU TECHNOLOGY Co.,Ltd.

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant