CN111242020A - Face recognition method and device - Google Patents
Face recognition method and device Download PDFInfo
- Publication number
- CN111242020A CN111242020A CN202010027475.5A CN202010027475A CN111242020A CN 111242020 A CN111242020 A CN 111242020A CN 202010027475 A CN202010027475 A CN 202010027475A CN 111242020 A CN111242020 A CN 111242020A
- Authority
- CN
- China
- Prior art keywords
- target
- facial
- feature
- face
- coincident
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06V40/174—Facial expression recognition
Abstract
The invention discloses a face recognition encryption method, which comprises the following steps: continuously acquiring multiple frames of facial images of a target user, and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features; combining the multiple groups of target facial features, calculating the spatial distance of the non-coincident feature points in the same region, and taking the average value of the spatial distance as a target coincident feature point until all the feature points are coincident to obtain a target coincident facial feature; marking and extracting the coincidence routes of the target coincidence characteristic points to obtain a plurality of target coincidence routes; associating the multiple target coincidence routes with the target coincidence facial features, using the multiple target coincidence routes as feature comparison sources, and storing the feature comparison sources in a facial feature database; the invention also discloses a face recognition decryption method; the invention adopts a multi-frame combination encryption comparison mode to carry out dynamic comparison, and can realize the improvement of the accuracy of face recognition.
Description
Technical Field
The invention relates to the technical field of face recognition, in particular to a face recognition method and a face recognition device.
Background
With the continuous development of the technology, the face recognition technology is widely applied. For example, visitor registration, attendance check-in, registration information entry, or entrance guard recognition and the like are performed by face recognition. The face recognition technology is based on the face characteristics of people, firstly, whether a face exists in an input face image or video stream is judged, and if the face exists, the position and the size of each face and the position information of each main facial organ are further given; and further extracting the identity characteristics implied in each face according to the information, and comparing the identity characteristics with the known faces so as to identify the identity of each face.
The traditional face recognition technology adopts a single-frame image recognition mode for recognition, namely, the feature storage is carried out by collecting the single-frame facial image features as a comparison source, then the comparison is carried out by identifying the single-frame facial features of a target user, and when the identity is determined, the recognition is successful. The traditional single-frame identification mode has insufficient encryption level, and when the facial features of two persons are similar to each other or the faces are corrected by special means, the single-frame image identification has high error rate because the facial features are subjected to static comparison.
Disclosure of Invention
The invention provides a face recognition method, which comprises a face recognition encryption method and a face recognition decryption method, wherein face features are extracted and merged by continuously collecting multi-frame images of a user, and target coincident face features are obtained by coincidence and are used as a comparison source; then carrying out feature recognition on multi-frame images of the detection user, and when the error value of the combined detection image feature and the comparison source is within the threshold value range, restoring the facial feature of the comparison source through a recombination route to carry out accurate feature recognition; the technical problem that the error rate is high due to static comparison of facial features of single-frame image recognition when the facial features of two persons are similar to each other or when the faces are corrected by special means due to the fact that the encryption level of a traditional single-frame recognition mode is not enough is solved, dynamic comparison is conducted by adopting a multi-frame combination encryption comparison mode, and therefore the accuracy of face recognition is improved.
In order to solve the above technical problem, an embodiment of the present invention provides a face recognition encryption method, including:
continuously acquiring multiple frames of facial images of a target user, and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features;
combining the multiple groups of target facial features, calculating the spatial distance of the non-coincident feature points in the same region, and taking the average value of the spatial distance as a target coincident feature point until all the feature points are coincident to obtain a target coincident facial feature;
marking and extracting the coincidence routes of the target coincidence characteristic points to obtain a plurality of target coincidence routes;
and associating the plurality of target coincidence routes with the target coincidence facial features to serve as feature comparison sources, and storing the feature comparison sources in a facial feature database.
As a preferred scheme, the specific step of extracting the facial features of each frame of facial image includes:
carrying out bone feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
connecting the plurality of skeleton points to obtain a skeleton network diagram;
carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
and marking the micro-expression characteristic points in the skeleton network graph to construct a target facial characteristic network.
As a preferred scheme, the specific step of merging the multiple groups of target facial features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value thereof as a target coincident feature point includes:
constructing a space coordinate system, setting a unique datum point and setting the multiple groups of target facial features in the space coordinate system;
carrying out space coordinate identification on feature points in each group of target face features to obtain coordinate parameters;
and carrying out region division on each group of target facial features, calculating a space average value of feature points with different coordinate parameters in the same region, and taking the space average value as a target coincidence feature point of the region.
The embodiment of the invention also provides a face recognition encryption device, which comprises:
the feature extraction module is used for continuously acquiring multiple frames of facial images of a target user and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features;
the feature coincidence module is used for merging the multiple groups of target face features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value of the spatial distance as a target coincidence feature point until all the feature points are coincided to obtain the target coincidence face features;
the route extraction module is used for marking and extracting the superposed routes of the target superposed characteristic points to obtain a plurality of target superposed routes;
and the association storage module is used for associating the target coincidence routes with the target coincidence facial features, serving as a feature comparison source and storing the feature comparison source in a facial feature database.
Preferably, the feature extraction module includes:
the characteristic marking unit is used for carrying out skeleton characteristic identification on each frame of face image and marking a plurality of skeleton points in the face image;
the characteristic connection unit is used for connecting the plurality of skeleton points to obtain a skeleton network diagram;
the expression marking unit is used for carrying out facial expression feature recognition on each frame of facial image and marking a plurality of micro-expression feature points in the facial image;
and the characteristic network unit is used for marking the micro-expression characteristic points in the skeleton network graph to construct a target facial characteristic network.
Preferably, the characteristic reclosing module includes:
the spatial coordinate unit is used for constructing a spatial coordinate system, setting a unique datum point and setting the multiple groups of target facial features in the spatial coordinate system;
the coordinate parameter unit is used for carrying out space coordinate identification on the feature points in each group of target face features to obtain coordinate parameters;
and the feature coincidence unit is used for carrying out region division on each group of target face features, calculating the space average value of the feature points with different coordinate parameters in the same region, and taking the space average value as the target coincidence feature point of the region.
The embodiment of the invention also provides a face recognition decryption method, which is based on the face recognition encryption method and comprises the following steps:
continuously acquiring multiple frames of face images of a detected user, and respectively extracting face features of the frames of face images to obtain multiple groups of detected face features;
combining the multiple groups of detected facial features, calculating the spatial distance of the non-coincident feature points in the same region, and taking the average value of the spatial distance as a detected coincident feature point until all the feature points are coincident to obtain a detected coincident facial feature;
comparing and identifying the detected coincident facial features with target coincident facial features in a facial feature database through a facial feature database, and when the spatial difference value of corresponding feature points is not greater than a preset threshold value, restoring the target coincident facial features through a plurality of associated target coincident routes to obtain a multi-frame facial image of a target user;
and comparing the face features of the multi-frame face image of the target user with the multi-frame face image of the detection user, and determining that the face recognition is successful when the features of any one frame of face image are consistent.
As a preferred scheme, the specific step of extracting the facial features of each frame of facial image includes:
carrying out bone feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
connecting the plurality of skeleton points to obtain a skeleton network diagram;
carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
and marking the micro-expression characteristic points in the skeleton network graph to construct a face characteristic detection network.
As a preferred scheme, the specific step of merging the multiple groups of detected facial features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value thereof as a detected coincident feature point includes:
constructing a space coordinate system, setting a unique datum point and setting the multiple groups of detected facial features in the space coordinate system;
carrying out space coordinate identification on feature points in each group of detected face features to obtain coordinate parameters;
and carrying out region division on each group of detected facial features, calculating a space average value of feature points with different coordinate parameters in the same region, and taking the space average value as a detection coincidence feature point of the region.
The embodiment of the invention also provides a face recognition decryption device, which comprises:
the feature extraction module is used for continuously acquiring a plurality of frames of face images of a detected user, and respectively extracting the face features of the frames of face images to obtain a plurality of groups of detected face features;
the feature coincidence module is used for merging the multiple groups of detected face features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value of the spatial distance as a detected coincident feature point until all the feature points are coincident to obtain a detected coincident face feature;
the route restoration module is used for comparing and identifying the detected coincident facial features with target coincident facial features in the facial feature database through the facial feature database, and restoring the target coincident facial features through a plurality of associated target coincident routes when the spatial difference value of corresponding feature points is not greater than a preset threshold value to obtain a multi-frame facial image of a target user;
and the feature recognition module is used for comparing the face features of the multi-frame face image of the target user with the multi-frame face image of the detection user, and when the features of any one frame of face image are consistent, determining that the face recognition is successful.
Preferably, the feature extraction module includes:
the characteristic marking unit is used for carrying out skeleton characteristic identification on each frame of face image and marking a plurality of skeleton points in the face image;
the characteristic connection unit is used for connecting the plurality of skeleton points to obtain a skeleton network diagram;
the expression marking unit is used for carrying out facial expression feature recognition on each frame of facial image and marking a plurality of micro-expression feature points in the facial image;
and the characteristic network unit is used for marking the micro-expression characteristic points in the skeleton network graph to construct a face characteristic detection network.
Preferably, the characteristic reclosing module includes:
the spatial coordinate unit is used for constructing a spatial coordinate system, setting a unique datum point and setting the multiple groups of detected facial features in the spatial coordinate system;
the coordinate parameter unit is used for carrying out space coordinate identification on feature points in each group of detected facial features to obtain coordinate parameters;
and the feature coincidence unit is used for carrying out region division on each group of detected facial features, calculating the space average value of feature points with different coordinate parameters in the same region, and taking the space average value as the detected coincidence feature points of the region.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein the computer program, when running, controls an apparatus in which the computer-readable storage medium is located to perform the face recognition encryption method according to any one of the above.
An embodiment of the present invention further provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, where the processor implements the face recognition encryption method according to any one of the above items when executing the computer program.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
extracting and combining facial features by continuously collecting multi-frame images of a user, and overlapping to obtain target overlapped facial features serving as a comparison source; then carrying out feature recognition on multi-frame images of the detection user, and when the error value of the combined detection image feature and the comparison source is within the threshold value range, restoring the facial feature of the comparison source through a recombination route to carry out accurate feature recognition; the technical problem that the error rate is high due to static comparison of facial features of single-frame image recognition when the facial features of two persons are similar to each other or when the faces are corrected by special means due to the fact that the encryption level of a traditional single-frame recognition mode is not enough is solved, dynamic comparison is conducted by adopting a multi-frame combination encryption comparison mode, and therefore the accuracy of face recognition is improved.
Drawings
FIG. 1: the steps of the face recognition encryption method in the embodiment of the invention are a flow chart;
FIG. 2: the invention discloses a flow chart of steps of a face recognition decryption method in an embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a preferred embodiment of the present invention provides a face recognition encryption method, including:
s1, continuously acquiring multiple frames of facial images of the target user, and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features; in this embodiment, the specific step of extracting the facial features of each frame of facial image includes:
s11, carrying out skeleton feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
s12, connecting the skeleton points to obtain a skeleton network diagram;
s13, carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
s14, marking the micro expression characteristic points in the skeleton network graph to construct a target face characteristic network.
Specifically, the face recognition encryption and decryption method can be applied to a face recognition access control machine; firstly, collecting facial feature images of a user, keeping connection with a background in a network transmission mode, and continuously synchronizing user data including base map information, user names, residence information and the like with the background in real time. The multi-frame facial images can be facial feature images when the user changes different expressions, recognition of facial expression features of the user is added to serve as an influence parameter, the combination of the comparison sources is participated, and the accuracy of face recognition can be further improved.
S2, merging the multiple groups of target face features, calculating the space distance of the non-coincident feature points in the same region, and taking the average value of the space distance as a target coincident feature point until all the feature points are coincident to obtain the target coincident face feature; in this embodiment, the step specifically includes:
s21, constructing a space coordinate system, setting a unique datum point and setting the multiple groups of target facial features in the space coordinate system;
s22, carrying out space coordinate recognition on the feature points in each group of target face features to obtain coordinate parameters;
and S23, performing region division on each group of target facial features, calculating the space average value of the feature points with different coordinate parameters in the same region, and taking the space average value as the target overlapping feature point of the region.
Specifically, a space coordinate system is established to merge multiple groups of target facial features, because the introduction of expression influence factors in the facial images of continuous multiple frames can cause the coordinate coefficients of feature points among frames to be different, one feature point is selected in advance as a reference point, or coordinate parameters in the coordinate system are selected as the reference point, the spatial distance of the non-coincident feature points is calculated, a virtual coordinate is obtained after the average value of the spatial distance is taken, and the virtual coordinate is used as the feature point of the region.
And S3, marking and extracting the superposition routes of the target superposition characteristic points to obtain a plurality of target superposition routes. Specifically, since the number of overlapped routes of the target overlapped feature point is large, and the route needs to be restored in the later stage to obtain an original facial feature image for performing accurate feature comparison, the overlapped route needs to be marked and extracted in the step to obtain the overlapped route.
And S4, associating the target coincidence routes with the target coincidence facial features as feature comparison sources, and storing the feature comparison sources in a facial feature database. In the step, the execution process of the whole face recognition encryption scheme is completed, and a comparison source is obtained, so that the face feature recognition of the user to be detected is performed subsequently.
Referring to fig. 2, an embodiment of the present invention further provides a face recognition decryption method, where the face recognition encryption method includes:
s1, continuously acquiring multiple frames of face images of the detected user, and respectively extracting the face features of the frames of face images to obtain multiple groups of detected face features; in this embodiment, the specific step of extracting the facial features of each frame of facial image includes:
s11, carrying out skeleton feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
s12, connecting the skeleton points to obtain a skeleton network diagram;
s13, carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
s14, marking the micro expression characteristic points in the skeleton network graph to construct a face characteristic detection network.
Similarly, the multi-frame facial image can be a facial feature image when the user changes different expressions, and the accuracy of face recognition can be further improved by adding recognition of facial expression features of the user as influence parameters and participating in combination of the comparison sources.
S2, combining the multiple groups of detected face features, calculating the space distance of the non-coincident feature points in the same region, and taking the average value as the detected coincident feature points until all the feature points are coincident to obtain the detected coincident face features; in this embodiment, the step specifically includes:
s21, constructing a space coordinate system, setting a unique datum point and setting the multiple groups of detected facial features in the space coordinate system;
s22, carrying out space coordinate recognition on the feature points in each group of detected face features to obtain coordinate parameters;
and S23, carrying out region division on each group of detected facial features, calculating the space average value of the feature points with different coordinate parameters in the same region, and taking the space average value as the detected overlapped feature points of the region.
Similarly, a space coordinate system is established to merge a plurality of groups of target facial features, because the introduction of expression influence factors in the facial images of continuous multiple frames can lead the coordinate coefficients of feature points among frames to be different, one feature point is selected as a reference point in advance, or a coordinate parameter in the coordinate system is selected as the reference point, the spatial distance of the non-coincident feature points is calculated, a virtual coordinate is obtained after the average value of the spatial distance is taken, and the virtual coordinate is used as the feature point of the region.
And S3, comparing and identifying the detected coincidence facial features with the target coincidence facial features in the facial feature database through the facial feature database, and when the spatial difference value of the corresponding feature points is not greater than a preset threshold value, restoring the target coincidence facial features through a plurality of associated target coincidence routes to obtain a multi-frame facial image of the target user.
Specifically, the preset threshold may be adjusted and modified according to actual needs. When the error value of the comparison of the two combined image features is within the threshold range, the previously associated coincident route is restored to obtain the original image features, at the moment, the multi-frame images are respectively identified one by one, and if any one frame is the same, the identification can be judged to be successful. In the step, the fact that the continuous frame expression characteristics of the comparison source are possibly not completely consistent with the expression characteristics of the source to be detected is considered, for example, a user is smiling during data entry, but the user is not smiling during detection; then the comparison can be successful only by face feature recognition of the original frame image.
S4, comparing the face features of the multi-frame face image of the target user with the multi-frame face image of the detection user, and determining that the face recognition is successful when the features of any one frame of face image are consistent. In this step, the execution process of the whole face recognition decryption scheme is completed, and the face recognition success is determined, so that the whole scheme is completed.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein, when running, the computer program controls the device where the computer-readable storage medium is located to execute the face recognition encryption method according to any one of the above embodiments.
The embodiment of the present invention further provides a terminal device, where the terminal device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the face recognition encryption method according to any one of the above embodiments when executing the computer program.
Preferably, the computer program may be divided into one or more modules/units (e.g., computer program) that are stored in the memory and executed by the processor to implement the invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the terminal device.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable gate array (FPGA) or other Programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., the general purpose Processor may be a microprocessor, or the Processor may be any conventional Processor, the Processor is a control center of the terminal device, and various interfaces and lines are used to connect various parts of the terminal device.
The memory mainly includes a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function, and the like, and the data storage area may store related data and the like. In addition, the memory may be a high speed random access memory, may also be a non-volatile memory, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, or may also be other volatile solid state memory devices.
It should be noted that the terminal device may include, but is not limited to, a processor and a memory, and those skilled in the art will understand that the terminal device is only an example and does not constitute a limitation of the terminal device, and may include more or less components, or combine some components, or different components.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (10)
1. A face recognition encryption method is characterized by comprising the following steps:
continuously acquiring multiple frames of facial images of a target user, and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features;
combining the multiple groups of target facial features, calculating the spatial distance of the non-coincident feature points in the same region, and taking the average value of the spatial distance as a target coincident feature point until all the feature points are coincident to obtain a target coincident facial feature;
marking and extracting the coincidence routes of the target coincidence characteristic points to obtain a plurality of target coincidence routes;
and associating the plurality of target coincidence routes with the target coincidence facial features to serve as feature comparison sources, and storing the feature comparison sources in a facial feature database.
2. The face recognition encryption method of claim 1, wherein the specific step of extracting the facial features of each frame of face image comprises:
carrying out bone feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
connecting the plurality of skeleton points to obtain a skeleton network diagram;
carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
and marking the micro-expression characteristic points in the skeleton network graph to construct a target facial characteristic network.
3. The face recognition encryption method of claim 2, wherein the specific step of merging the multiple sets of target face features, calculating the spatial distances of non-coincident feature points in the same region, and taking the average value thereof as a target coincident feature point comprises:
constructing a space coordinate system, setting a unique datum point and setting the multiple groups of target facial features in the space coordinate system;
carrying out space coordinate identification on feature points in each group of target face features to obtain coordinate parameters;
and carrying out region division on each group of target facial features, calculating a space average value of feature points with different coordinate parameters in the same region, and taking the space average value as a target coincidence feature point of the region.
4. A face recognition encryption apparatus, comprising:
the feature extraction module is used for continuously acquiring multiple frames of facial images of a target user and respectively extracting facial features of the frames of facial images to obtain multiple groups of target facial features;
the feature coincidence module is used for merging the multiple groups of target face features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value of the spatial distance as a target coincidence feature point until all the feature points are coincided to obtain the target coincidence face features;
the route extraction module is used for marking and extracting the superposed routes of the target superposed characteristic points to obtain a plurality of target superposed routes;
and the association storage module is used for associating the target coincidence routes with the target coincidence facial features, serving as a feature comparison source and storing the feature comparison source in a facial feature database.
5. A face recognition decryption method based on the face recognition encryption method of claim 3, comprising:
continuously acquiring multiple frames of face images of a detected user, and respectively extracting face features of the frames of face images to obtain multiple groups of detected face features;
combining the multiple groups of detected facial features, calculating the spatial distance of the non-coincident feature points in the same region, and taking the average value of the spatial distance as a detected coincident feature point until all the feature points are coincident to obtain a detected coincident facial feature;
comparing and identifying the detected coincident facial features with target coincident facial features in a facial feature database through a facial feature database, and when the spatial difference value of corresponding feature points is not greater than a preset threshold value, restoring the target coincident facial features through a plurality of associated target coincident routes to obtain a multi-frame facial image of a target user;
and comparing the face features of the multi-frame face image of the target user with the multi-frame face image of the detection user, and determining that the face recognition is successful when the features of any one frame of face image are consistent.
6. The face recognition decryption method of claim 5, wherein the specific step of extracting facial features of each frame of facial image comprises:
carrying out bone feature recognition on each frame of face image, and marking a plurality of skeleton points in the face image;
connecting the plurality of skeleton points to obtain a skeleton network diagram;
carrying out facial expression feature recognition on each frame of facial image, and marking a plurality of micro-expression feature points in the facial image;
and marking the micro-expression characteristic points in the skeleton network graph to construct a face characteristic detection network.
7. The face recognition decryption method of claim 6, wherein the specific steps of merging the multiple groups of detected face features, calculating spatial distances of non-coincident feature points in the same region, and taking an average value thereof as a detected coincident feature point comprise:
constructing a space coordinate system, setting a unique datum point and setting the multiple groups of detected facial features in the space coordinate system;
carrying out space coordinate identification on feature points in each group of detected face features to obtain coordinate parameters;
and carrying out region division on each group of detected facial features, calculating a space average value of feature points with different coordinate parameters in the same region, and taking the space average value as a detection coincidence feature point of the region.
8. A face recognition decryption apparatus, comprising:
the feature extraction module is used for continuously acquiring a plurality of frames of face images of a detected user, and respectively extracting the face features of the frames of face images to obtain a plurality of groups of detected face features;
the feature coincidence module is used for merging the multiple groups of detected face features, calculating the spatial distance of non-coincident feature points in the same region, and taking the average value of the spatial distance as a detected coincident feature point until all the feature points are coincident to obtain a detected coincident face feature;
the route restoration module is used for comparing and identifying the detected coincident facial features with target coincident facial features in the facial feature database through the facial feature database, and restoring the target coincident facial features through a plurality of associated target coincident routes when the spatial difference value of corresponding feature points is not greater than a preset threshold value to obtain a multi-frame facial image of a target user;
and the feature recognition module is used for comparing the face features of the multi-frame face image of the target user with the multi-frame face image of the detection user, and when the features of any one frame of face image are consistent, determining that the face recognition is successful.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored computer program; wherein the computer program, when running, controls the device in which the computer-readable storage medium is located to perform the face recognition encryption method according to any one of claims 1 to 3.
10. A terminal device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the face recognition encryption method according to any one of claims 1 to 3 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010027475.5A CN111242020A (en) | 2020-01-10 | 2020-01-10 | Face recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010027475.5A CN111242020A (en) | 2020-01-10 | 2020-01-10 | Face recognition method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111242020A true CN111242020A (en) | 2020-06-05 |
Family
ID=70866094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010027475.5A Pending CN111242020A (en) | 2020-01-10 | 2020-01-10 | Face recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111242020A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514432A (en) * | 2012-06-25 | 2014-01-15 | 诺基亚公司 | Method, device and computer program product for extracting facial features |
| US20170083753A1 (en) * | 2015-09-22 | 2017-03-23 | ImageSleuth, Inc. | Automated methods and systems for identifying and characterizing face tracks in video |
| CN108229330A (en) * | 2017-12-07 | 2018-06-29 | 深圳市商汤科技有限公司 | Face fusion recognition methods and device, electronic equipment and storage medium |
| CN109948489A (en) * | 2019-03-09 | 2019-06-28 | 闽南理工学院 | A kind of face identification system and method based on the fusion of video multiframe face characteristic |
| CN110363187A (en) * | 2019-08-29 | 2019-10-22 | 上海云从汇临人工智能科技有限公司 | A kind of face identification method, device, machine readable media and equipment |
| CN110472459A (en) * | 2018-05-11 | 2019-11-19 | 华为技术有限公司 | The method and apparatus for extracting characteristic point |
-
2020
- 2020-01-10 CN CN202010027475.5A patent/CN111242020A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514432A (en) * | 2012-06-25 | 2014-01-15 | 诺基亚公司 | Method, device and computer program product for extracting facial features |
| US20170083753A1 (en) * | 2015-09-22 | 2017-03-23 | ImageSleuth, Inc. | Automated methods and systems for identifying and characterizing face tracks in video |
| CN108229330A (en) * | 2017-12-07 | 2018-06-29 | 深圳市商汤科技有限公司 | Face fusion recognition methods and device, electronic equipment and storage medium |
| CN110472459A (en) * | 2018-05-11 | 2019-11-19 | 华为技术有限公司 | The method and apparatus for extracting characteristic point |
| CN109948489A (en) * | 2019-03-09 | 2019-06-28 | 闽南理工学院 | A kind of face identification system and method based on the fusion of video multiframe face characteristic |
| CN110363187A (en) * | 2019-08-29 | 2019-10-22 | 上海云从汇临人工智能科技有限公司 | A kind of face identification method, device, machine readable media and equipment |
Non-Patent Citations (2)
| Title |
|---|
| ZHON XL: "Integrating face and gait for human recognition at a distance in video" * |
| 董胜: "基于人脸区域特征相关性的视频流人脸识别研究" * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376684B (en) | Face key point detection method and device, computer equipment and storage medium | |
| CN110909651B (en) | Method, device and equipment for identifying video main body characters and readable storage medium | |
| CN104050449B (en) | A kind of face identification method and device | |
| WO2018028546A1 (en) | Key point positioning method, terminal, and computer storage medium | |
| CN111950424B (en) | Video data processing method and device, computer and readable storage medium | |
| JP4951498B2 (en) | Face image recognition device, face image recognition method, face image recognition program, and recording medium recording the program | |
| CN110348331B (en) | Face recognition method and electronic equipment | |
| CN107491674B (en) | Method and device for user authentication based on characteristic information | |
| CN110197149B (en) | Ear key point detection method and device, storage medium and electronic equipment | |
| CN110781770B (en) | Living body detection method, device and equipment based on face recognition | |
| CN112016371A (en) | Face key point detection method, device, equipment and storage medium | |
| CN112613508A (en) | Object identification method, device and equipment | |
| CN109784227B (en) | image detection and identification method and device | |
| CN112270748B (en) | Three-dimensional reconstruction method and device based on image | |
| US9286707B1 (en) | Removing transient objects to synthesize an unobstructed image | |
| JP2013218605A (en) | Image recognition device, image recognition method, and program | |
| CN113505682A (en) | Living body detection method and device | |
| CN111767839A (en) | Vehicle driving track determining method, device, equipment and medium | |
| CN109598201B (en) | Action detection method and device, electronic equipment and readable storage medium | |
| CN111242020A (en) | Face recognition method and device | |
| CN116205723A (en) | Artificial intelligence-based face tag risk detection method and related equipment | |
| CN112488062B (en) | Image identification method, device, equipment and medium | |
| CN113807451B (en) | Panoramic image feature point matching model training method and device and server | |
| CN114627542A (en) | Eye movement position determination method and device, storage medium and electronic equipment | |
| CN115937921A (en) | Image processing method, device and medium based on face recognition technology |
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 |