CN116386121A - Personnel identification method and device based on power grid safety production - Google Patents
Personnel identification method and device based on power grid safety production Download PDFInfo
- Publication number
- CN116386121A CN116386121A CN202310619474.3A CN202310619474A CN116386121A CN 116386121 A CN116386121 A CN 116386121A CN 202310619474 A CN202310619474 A CN 202310619474A CN 116386121 A CN116386121 A CN 116386121A
- Authority
- CN
- China
- Prior art keywords
- personnel
- face
- head portrait
- information
- certificate
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 45
- 230000005540 biological transmission Effects 0.000 claims abstract description 8
- 210000003128 head Anatomy 0.000 claims description 133
- 238000012549 training Methods 0.000 claims description 40
- 230000008569 process Effects 0.000 claims description 28
- 238000001228 spectrum Methods 0.000 claims description 22
- 238000007781 pre-processing Methods 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 16
- 210000001747 pupil Anatomy 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 7
- 210000004709 eyebrow Anatomy 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000003708 edge detection Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims 1
- 238000010276 construction Methods 0.000 abstract description 24
- 238000001514 detection method Methods 0.000 abstract description 14
- 238000004422 calculation algorithm Methods 0.000 abstract description 6
- 238000012795 verification Methods 0.000 abstract description 6
- 238000007689 inspection Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000001815 facial effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 210000000887 face Anatomy 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 210000001061 forehead Anatomy 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
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
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Multimedia (AREA)
- Tourism & Hospitality (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Security & Cryptography (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a personnel identification method and a device based on power grid safety production, wherein the personnel identification method is applied to a personnel identification system, the personnel identification system comprises a certificate identification system and a face identification system, identity information of power grid safety production personnel is pre-stored in the certificate identification system, and data transmission is carried out between the certificate identification system and the face identification system through a local area network. According to the invention, on the basis of conventional identity information document identification, the head portrait identification algorithm is combined with conventional identity information document identification through the face identification system, and when constructors or related staff enter a construction site and a place which needs related evidence to enter, double verification is performed through the document identification system and the face identification system, so that the accuracy of identity authentication is improved, and meanwhile, compared with the conventional identity identification mode, the time of identity identification is reduced, and the safety detection efficiency is improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a personnel identification method and device based on power grid safety production.
Background
With the development of modern society, the requirement of the whole society on information security is also higher and higher. The production and construction of the electric power engineering are full of a plurality of uncertainty factors, meanwhile, the safety production process has a higher threshold for professional knowledge, and the construction personnel have a higher operation standard, and the construction personnel can enter the site to perform construction operation after the construction personnel have to pass the related capability authentication examination and the safety standard examination. However, some constructors do not have construction qualification, and impoundment replaces other persons to enter a construction site for construction, so that serious safety accidents can be caused.
At present, the safety inspection of the domestic significant power engineering production and construction projects mostly adopts a traditional manual inspection mode. Against the increasingly serious problem of safe production, the defects of manual verification are gradually revealed. If security inspection flow, post management, security inspection equipment and the like are not updated in time, the situation that constructors take place in place of the impersonation and pass the leak detection frequently occurs. Therefore, how to confirm the personal identity of the constructor is a social problem to be solved urgently.
Disclosure of Invention
In view of the problem that the electric power constructors impost to replace others to enter a construction site for construction and have potential safety hazards, the embodiment of the invention provides a personnel identification method and device based on power grid safety production, so that the identities of constructors can be accurately identified.
In one aspect, the embodiment of the invention discloses a personnel identification method based on power grid safety production, the personnel identification method is applied to a personnel identification system, the personnel identification system comprises a certificate identification system and a face identification system, identity information of power grid safety production personnel is pre-stored in the certificate identification system, data transmission is carried out between the certificate identification system and the face identification system through a local area network, and the personnel identification method comprises the following steps:
acquiring certificate information of a person to be identified through a certificate identification system, sending the certificate information to a face recognition system, comparing the certificate information with identity information of registered power grid safety production personnel, and judging whether the certificate information is true and effective;
if the certificate information is true and effective, acquiring a head portrait photo of a person to be identified through a camera, and extracting face characteristic information in the head portrait photo, wherein the face characteristic information comprises key characteristic points of a face;
calculating the feature value distance between key feature points of the face of the person, and judging whether the feature value distance is larger than a preset distance threshold value or not;
if the characteristic value distance is larger than a preset distance threshold, judging that the personnel to be identified is inconsistent with personnel information corresponding to the certificate information, and intercepting the personnel to be identified;
And if the distance between the characteristic values is smaller than or equal to a preset distance threshold, judging that the personnel to be identified is consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified.
Optionally, the face recognition system is provided with a face recognition module, and the calculating of the feature value distance between the key feature points of the face comprises:
positioning the input head portrait photo of the person to be identified through the face recognition module, and acquiring key feature points of the face of the person;
measuring and analyzing the key feature points, and calculating the pixel distance between two key feature points taking pixels as units;
and determining the actual distance between the two key feature points according to the pixel distance and the conversion coefficient.
Optionally, the key feature points include eyes, nose tips, mouth corner points and eyebrows, and the calculating the pixel distance between two key feature points in units of pixels includes:
the face recognition module is used for recognizing the input head portrait photo of the person to be recognized, the three-family five-eye method is used for roughly extracting the face part in the head portrait photo, then horizontal projection and vertical projection are carried out on the extracted face part, the accurate position of eyes is positioned, and the corresponding eye image is obtained;
And performing edge detection and ellipse fitting on the human eye image, taking the circle center of the ellipse as the pupil position of the eye, and obtaining a width difference delta x and a height difference delta y between the two pupils, wherein the units of delta x and delta y are pixels.
Optionally, the face recognition system is provided with a face recognition module, and before using the face recognition system to perform identity recognition, the person recognition method further includes:
training the face recognition module;
the process of training the face recognition module comprises the following steps:
acquiring identity information of m power grid security producers, wherein the identity information comprises credentials, basic information and n photos of the head portrait corresponding to the producers; wherein m and n are positive integers;
preprocessing the q-th reliable head portrait photo of the p-th person, wherein the preprocessing process comprises the steps of converting the certificate photo and the head portrait photo into a face image with a fixed size, and randomly changing the brightness of the face image during preprocessing to enhance data; wherein p is more than 0 and less than m+1, q is more than 0 and less than n+1, and p and q are positive integers;
carrying out data normalization on the face image with the fixed size, so that the pixel value of the face image is scaled to be between [ -1,1 ];
Extracting face characteristic information of each face image, wherein the face characteristic information comprises face key characteristic points, and calculating characteristic value distances among the face key characteristic points;
and correlating the characteristic value distance with the identity information of the corresponding production personnel.
Optionally, the process of training the face recognition module further includes:
and carrying out wavelet decomposition on the q-th reliable head portrait photo of the p-th person to obtain a first low-frequency face image, carrying out Fourier transformation on the first low-frequency face image, and adopting the amplitude of the first low-frequency face image as the frequency spectrum characteristic of the head portrait photo.
Optionally, after extracting the face feature information in the head portrait photo of the person to be identified, the person identification method further includes:
performing wavelet decomposition on the head portrait photo of the person to be identified, obtaining a second low-frequency face image, performing Fourier transform on the second low-frequency face image, and taking the amplitude of the second low-frequency face image as the frequency spectrum characteristic of the head portrait photo of the person to be identified;
respectively determining the similarity dpq between a plurality of first low-frequency face image spectrum feature vectors X and a plurality of second low-frequency face image spectrum feature vectors Y, wherein dpq=cos (X, Y), and the similarity dpq is the cosine value of an included angle between the feature vectors X and the feature vectors Y;
Determining a value dmax with the maximum similarity value from a plurality of similarity dpq, and judging whether dmax is larger than a preset similarity threshold value or not;
if dmax is greater than or equal to a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified;
if dmax is smaller than a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and intercepting the personnel to be identified.
Optionally, the process of training the face recognition module further includes:
acquiring a two-dimensional head portrait photo A of a power grid safety production person and a three-dimensional image photo B corresponding to the two-dimensional head portrait photo A, marking pixel values at coordinates (x, y) of the two-dimensional head portrait photo A as A (x 1, y 1), and marking pixel values at coordinates (x, y) of the three-dimensional image photo B as B (x 2, y 2);
fusing the two-dimensional head portrait photo A and the three-dimensional image photo B to obtain a fused head portrait photo C, and marking the pixel value at the coordinates (x, y) of the head portrait photo C as C (x 3, y 3);
training a face recognition module by adopting a plurality of groups of pixel values C (x 3, y 3);
the pixel value C (x 3, y 3) is calculated as follows:
C(x3,y3)=w(x,y)
A(x1,y1)/>
b (x 2, y 2), w (x, y) is a weighting coefficient when the pixel value a (x 1, y 1) and the pixel value B (x 2, y 2) are fused.
Optionally, when fusing n groups of corresponding two-dimensional head portrait photos a and three-dimensional image photos B, the weighting coefficients w (x, y) are calculated as follows:
wherein A (x i ,y i ) And B (x) i ,y i ) Two-dimensional head portrait and three-dimensional head portrait, respectively, of head portrait photo, w (x i ,y i ) N is a positive integer greater than 0, which is a weighting coefficient corresponding to the head portrait photo.
On the other hand, the embodiment of the invention also discloses a personnel identification device based on the power grid safety production, the personnel identification device is applied to a personnel identification system, the personnel identification system comprises a certificate identification system and a face identification system, the identity information of the power grid safety production personnel is prestored in the certificate identification system, the certificate identification system and the face identification system carry out data transmission through a local area network, and the personnel identification device comprises:
the certificate information acquisition module is used for acquiring the certificate information of the personnel to be identified through the certificate identification system, sending the certificate information to the face recognition system, comparing the certificate information with the identity information of registered power grid safety production personnel, and judging whether the certificate information is true and effective;
the face feature information acquisition module is used for acquiring head portrait photos of the person to be identified through a camera and extracting face feature information in the head portrait photos if the certificate information is true and effective, wherein the face feature information comprises key feature points of face parts;
The characteristic value distance processing module is used for calculating characteristic value distances among key characteristic points of the face of the person and judging whether the characteristic value distances are larger than a preset distance threshold value or not;
the first personnel identity recognition module is used for judging that personnel to be recognized are inconsistent with personnel information corresponding to the certificate information if the characteristic value distance is larger than a preset distance threshold value, and intercepting the personnel to be recognized;
and the second personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if the distance of the characteristic value is smaller than or equal to a preset distance threshold value, and releasing the personnel to be recognized.
Optionally, the face recognition system is provided with a face recognition module, and the eigenvalue distance processing module includes:
the key feature point positioning sub-module is used for positioning the input head portrait photo of the person to be identified through the face recognition module and obtaining key feature points of the face;
the pixel distance calculating sub-module is used for measuring and analyzing the key feature points and calculating the pixel distance between two key feature points taking pixels as units;
and the actual distance calculation sub-module is used for determining the actual distance between the two key feature points according to the pixel distance and the conversion coefficient.
Optionally, the key feature points include eyes, nose tips, mouth corner points and eyebrows, and the pixel distance calculating submodule includes:
the first pixel distance calculating unit is used for identifying the input head portrait photo of the person to be identified through the face recognition module, performing coarse extraction on a face part in the head portrait photo by using a three-room five-eye method, performing horizontal projection and vertical projection on the extracted face part, positioning the accurate position of eyes, and acquiring a corresponding eye image;
and the second pixel distance calculation unit is used for carrying out edge detection and ellipse fitting on the human eye image, taking the circle center of the ellipse as the pupil position of the eye, and obtaining a width difference delta x and a height difference delta y between the two pupils, wherein the units of delta x and delta y are pixels.
Optionally, the face recognition system is provided with a face recognition module, and the person recognition system further includes:
the face recognition training module is used for training the face recognition module, and the face recognition training module comprises:
the first training sub-module is used for acquiring identity information of m power grid security producers, wherein the identity information comprises credentials, basic information and n head portrait photos corresponding to the producers; wherein m and n are positive integers;
The second training sub-module is used for preprocessing the q-th reliable head portrait photo of the p-th person, the preprocessing process comprises the steps of converting the certificate photo and the head portrait photo into a face image with a fixed size, and randomly changing the brightness of the face image during preprocessing to enhance data; wherein p is more than 0 and less than m+1, q is more than 0 and less than n+1, and p and q are positive integers;
the third training sub-module is used for carrying out data normalization on the face image with fixed size so that the pixel value of the face image is scaled to be between [ -1,1 ];
a fourth training sub-module, configured to extract face feature information of each face image, where the face feature information includes face key feature points, and calculate feature value distances between the face key feature points;
and the fifth training sub-module is used for associating the characteristic value distance with the identity information of the corresponding production personnel.
Optionally, the face recognition training module further includes:
and the sixth training submodule is used for carrying out wavelet decomposition on the q-th reliable head portrait photo of the p-th person to obtain a first low-frequency face image, carrying out Fourier transform on the first low-frequency face image, and adopting the amplitude of the first low-frequency face image as the frequency spectrum characteristic of the head portrait photo.
Optionally, the person identifying device further includes:
the frequency spectrum feature processing module is used for carrying out wavelet decomposition on the head portrait photo of the person to be identified, obtaining a second low-frequency face image, carrying out Fourier transform on the second low-frequency face image, and adopting the amplitude of the second low-frequency face image as the frequency spectrum feature of the head portrait photo of the person to be identified;
the similarity calculation module is used for respectively determining similarity dpq between a plurality of first low-frequency face image spectrum feature vectors X and a plurality of second low-frequency face image spectrum features Y, wherein dpq=cos (X, Y), and the similarity dpq is a cosine value of an included angle between the feature vectors X and Y;
the similarity judging module is used for determining a value dmax with the maximum similarity value from a plurality of similarity dpq and judging whether dmax is larger than a preset similarity threshold value or not;
the third personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if dmax is larger than or equal to a preset similarity threshold value, and releasing the personnel to be recognized;
and the fourth personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if dmax is smaller than a preset similarity threshold value, and intercepting the personnel to be recognized.
Optionally, the face recognition training module further includes:
a seventh training sub-module, configured to obtain a two-dimensional head portrait photo a of a power grid security producer and a three-dimensional image photo B corresponding to the two-dimensional head portrait photo a, record a pixel value at a coordinate (x, y) of the two-dimensional head portrait photo a as a (x 1, y 1), and record a pixel value at a coordinate (x, y) of the three-dimensional image photo B as B (x 2, y 2);
the eighth training submodule is used for fusing the two-dimensional head portrait photo A and the three-dimensional image photo B, obtaining a fused head portrait photo C, and marking the pixel value at the coordinates (x, y) of the head portrait photo C as C (x 3, y 3);
a ninth training submodule, configured to train the face recognition module with a plurality of sets of pixel values C (x 3, y 3);
the pixel value C (x 3, y 3) is calculated as follows:
C(x3,y3)=w(x,y)
A(x1,y1)/>
b (x 2, y 2), w (x, y) is a weighting coefficient when the pixel value a (x 1, y 1) and the pixel value B (x 2, y 2) are fused.
Optionally, when fusing n groups of corresponding two-dimensional head portrait photos a and three-dimensional image photos B, the weighting coefficients w (x, y) are calculated as follows:
wherein A (x i ,y i ) And B (x) i ,y i ) Two-dimensional head portrait and three-dimensional head portrait, respectively, of head portrait photo, w (x i ,y i ) N is a positive integer greater than 0, which is a weighting coefficient corresponding to the head portrait photo.
On the other hand, the embodiment of the invention also discloses an electronic device, which comprises: a processor, a memory and a computer program stored on the memory and capable of running on the processor, which computer program, when being executed by the processor, implements the steps of the person identification method described above.
In another aspect, the embodiment of the present invention further discloses a computer readable storage medium, where a computer program is stored, where the computer program is executed by a processor to implement the steps of the person identification method described above.
By adopting the technical scheme, the technical scheme of the invention has the following beneficial effects:
(1) According to the invention, on the basis of conventional identity information document identification, the head portrait identification algorithm is combined with conventional identity information document identification through the face identification system, and when constructors or related staff enter a construction site and a place which needs related evidence to enter, double verification is performed through the document identification system and the face identification system, so that the accuracy of identity authentication is improved, and meanwhile, compared with the conventional identity identification mode, the time of identity identification is reduced, and the safety detection efficiency is improved.
(2) By calculating the characteristic value distance between key characteristic points of the face, the accuracy of the identity recognition of constructors can be improved, the condition that constructors impound to replace to enter construction sites can be recognized, and the safety in the power grid construction process is improved.
(3) By fusing the two-dimensional face image and the three-dimensional face image, the complete information of key feature points in the face recognition process can be constructed, so that the head images of the faces to be recognized can be recognized quickly and accurately, and the recognition efficiency of a face recognition system is improved.
Drawings
Fig. 1 is a step flowchart of a personnel identification method based on power grid safety production according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a method for rough extraction of face parts in a head portrait photograph;
fig. 3 is a block diagram of a personnel identification device based on power grid safety production according to an embodiment of the present invention;
FIG. 4 is a screenshot of the result of an embodiment of the invention re-enabling automatic acquisition of information on an identity information document by a document identification system.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Currently, in the process of power grid engineering construction operation, the identification comprises the following two parts: (1) Identification of identity information certificates, which generally comprise authenticity of certificates, expiration of certificates and the like, is a focus of attention of the current identity inspection; (2) Whether the person with the certificate is the person to whom the certificate belongs or not is a problem, namely, the consistency check of the certificates. In various places where the identity of the person needs to be verified, most of the security inspection personnel simply observe whether the gender of the person is consistent or not, and whether the obvious facial features are consistent or not. However, the manual identification of the security inspection personnel has larger error, the accuracy of the identification is not high, the identification speed is slower in the manual identification process, and when the number of the personnel to be inspected is larger, the workload of the security inspection personnel is also large, the labor is very consumed, and the efficiency is low.
In recent years, with rapid development of computer technology, face automatic recognition technology has been widely researched and developed, and the purpose of face recognition is to extract personalized features of a person from a face image and thereby recognize the identity of the person. A simple automatic face recognition system generally includes the following 4 aspects: (1) face detection: i.e. the presence of a face is detected from various different scenes and its position is determined. (2) face normalization: correcting the face variation in the aspects of scale, illumination, rotation and the like. (3) face representation: in some way, the detection of a face and the detection of a known face in the database. (4) face recognition: and comparing the face to be recognized with the known face in the database to obtain related information. In the face recognition process, the classification capability, algorithm complexity and realizability of the features are factors to be considered in determining the feature extraction method. The extracted features have a decisive influence on the final classification result. The upper limit of resolution that can be achieved by the classifier is the maximum distinguishable degree between various features. Therefore, the implementation of face recognition requires comprehensive consideration of feature selection, feature extraction and classifier design.
Because the information of the power grid constructor belongs to the information which can be collected in advance, the inventor considers that on the basis of conventional identity information certificate recognition, a head portrait recognition algorithm is combined with conventional identity information certificate recognition through a face recognition system, and when constructors or related staff enter a place where construction is heavy and the relevant staff need to be proved to enter, double verification is carried out through the certificate recognition system and the face recognition system, so that the accuracy of identity authentication is improved. Correspondingly, fig. 1 is a flowchart of steps of a personnel identification method based on power grid safety production, which is provided by the embodiment of the invention, and the personnel identification method is applied to a personnel identification system, wherein the personnel identification system comprises a certificate identification system and a face identification system, identity information of power grid safety production personnel is pre-stored in the certificate identification system, and data transmission is performed between the certificate identification system and the face identification system through a local area network, and the personnel identification method comprises the following steps:
The process of obtaining the certificate information of the person to be identified by the certificate identification system can comprise the following steps: the personnel to be identified or the safety detection personnel place the identity information certificate in an identity information card reading area of the identity information certificate reading module, and the built-in identity information reader chip reads the relevant information of the identity information certificate, wherein the identity information and the basic information are certificate information, and the basic information can comprise certificate number, name, gender, date of birth and the like. As shown in FIG. 4, the invention can automatically acquire the information on the identity information document by the document identification system.
After obtaining the certificate information and the basic information, the basic information of the identity information certificate can be compared with the identity information of registered power grid safety production personnel, and if the basic information of the identity information certificate is completely consistent with the registration information after comparison, the certificate information is truly and effectively judged; if the basic information of the identity information certificate is inconsistent with the registration information after comparison, the identity information of the person to be identified can be checked by a safety detection personnel, and if the check is correct, a head portrait photo of the person to be identified is acquired by a camera; if the checked information is inconsistent, judging that the identity information of the person to be identified is wrong, and intercepting the person to be identified.
It should be noted that in the prior art, there are many applications in the face recognition technology, and in general, the face recognition adopts a cloud comparison technology, that is, a service program is identified based on server-side identity information comparison, and the service is deployed on its own server (cloud server or local server). After deployment, identity recognition service is called through an application interface, and identity information certificate images and face images are uploaded. After the identification and comparison are completed on the server, the identification information and the comparison result are returned. However, the security requirement is met in the power grid construction process, so that the cloud-based personnel and evidence comparison technology is not suitable for the technical field of power grids, cannot be effectively integrated with the original security detection subsystem, and achieves the effect of real-time operation. According to the embodiment of the application, the certificate identification system and the face recognition system are adopted to conduct data transmission through the local area network, so that maintenance of identity information can be effectively protected, and information leakage risk is reduced.
102, if the certificate information is true and effective, acquiring a head portrait photo of a person to be identified through a camera, and extracting face feature information in the head portrait photo, wherein the face feature information comprises key feature points of a face;
In some embodiments, after the basic information in the certificate information is judged to be true and valid, the camera is called, the camera shoots towards the direction of the face of the certificate holder, and at least one face head portrait is obtained. It should be noted that, the photo collected by the camera does not necessarily contain the face information of the certificate, the detection step in the face recognition module is used for detecting the face in the image, when the face is detected, the photo of the certificate is obtained, otherwise, the face image of the certificate is continuously collected. In order to improve the accuracy of face detection and recognition, the face recognition module needs to perform image preprocessing on the photo acquired by the camera, wherein the mode of preprocessing the input face image can comprise image filtering, region segmentation, gray scale and scale normalization, face alignment, histogram equalization, local binary mode and the like.
In some embodiments, since two-dimensional image recognition is more sensitive to the quality of an input face image, such as the recognition rate of a similar image like a side face is lower, in order to improve the recognition efficiency of a face recognition module, a two-dimensional head portrait photo a of a power grid security producer and a three-dimensional image photo B corresponding to the two-dimensional head portrait photo a can be obtained, pixel values at coordinates (x, y) of the two-dimensional head portrait photo a are marked as a (x 1, y 1), and pixel values at coordinates (x, y) of the three-dimensional image photo B are marked as B (x 2, y 2); fusing the two-dimensional head portrait photo A and the three-dimensional image photo B to obtain a fused head portrait photo C, and marking the pixel value at the coordinates (x, y) of the head portrait photo C as C (x 3, y 3); the face recognition module is trained by adopting a plurality of groups of pixel values C (x 3, y 3), and the two-dimensional face image and the three-dimensional face image are fused, so that the complete information of key feature points in the face recognition process can be constructed, the face head portrait to be recognized can be rapidly and accurately recognized, and the recognition efficiency of a face recognition system is improved.
The pixel value C (x 3, y 3) is calculated as follows:
C(x3,y3)=w(x,y)
A(x1,y1)/>
b (x 2, y 2), w (x, y) is a weighting coefficient when the pixel value a (x 1, y 1) and the pixel value B (x 2, y 2) are fused. The weighting coefficients w (x, y) are calculated as follows:
wherein A (x i ,y i ) And B (x) i ,y i ) Corresponding to the pixel values of the two-dimensional head portrait and the three-dimensional head portrait of the head portrait photo respectively, w (x i ,y i ) N is a positive integer greater than 0, which is a weighting coefficient corresponding to the head portrait photo.
In some embodiments, the process of training the face recognition module may further include: acquiring identity information of m power grid security producers, wherein the identity information comprises credentials, basic information and n photos of the head portrait corresponding to the producers; wherein m and n are positive integers; preprocessing the q-th reliable head portrait photo of the p-th person, wherein the q-th reliable head portrait photo is used for preprocessing the q-th reliable head portrait photo of the p-th personThe preprocessing process comprises the steps of converting the certificate photo and the head photo into face images with fixed sizes, and randomly changing the brightness of the face images during preprocessing to enhance data; wherein p is more than 0 and less than m+1, q is more than 0 and less than n+1, and p and q are positive integers; data normalizing a face image of a fixed size such that pixel values of the face image scale to [ -1,1]Between them; extracting face characteristic information of each face image, wherein the face characteristic information comprises face key characteristic points, and calculating characteristic value distances among the face key characteristic points; and correlating the characteristic value distance with the identity information of the corresponding production personnel. Exemplary, the face image may be converted to 100 after preprocessing 
100, the convolutional neural network in the face recognition module can recognize quickly, in the course of data normalization, the pixel value obtained when processing face image can be subtracted 127.5 and divided by 128, so that the pixel value can be scaled to [ -1,1]Between them.
in some embodiments, the facial key feature points may include eyes, nose tips, mouth corner points, and eyebrows, and facial features may be labeled when extracting facial key feature points: the eyes, the nose tip and the lips are used for obtaining the minimum face inscription rectangle, and the minimum face inscription rectangle inscribes four characteristic points of eyes, the nose tip and the lips. The way of calculating the feature value distance between the key feature points of the face of the person can comprise: positioning the input head portrait photo of the person to be identified through a face recognition module, and acquiring key feature points of the face of the person; measuring and analyzing the key feature points, and calculating the pixel distance between two key feature points taking pixels as units; and determining the actual distance between the two key feature points according to the pixel distance and the conversion coefficient. Because the distances between the key feature points of the face of the person are essentially different, the accuracy of identity authentication can be improved by calculating the distances between the key feature points of the face of the person and judging through the distance values.
Fig. 2 is a schematic diagram of a method for roughly extracting a face part in a head portrait photo. In the figure, the human face is trisected through the hairline, the eyebrow line, the nose bottom line and the forehead bottom line, and is pentad through the positions of eyes, so that the characteristic information of the human face is conveniently and roughly extracted. The calculating of the pixel distance between the two key feature points in units of pixels may include: the head portrait photo of the person to be identified is identified through a face identification module, the face part in the head portrait photo is roughly extracted by a three-family five-eye method, then horizontal projection and vertical projection are carried out on the extracted face part, the accurate position of eyes is positioned, and a corresponding eye image is obtained; and performing edge detection and ellipse fitting on the human eye image, taking the circle center of the ellipse as the pupil position of the eye, and obtaining a width difference delta x and a height difference delta y between the two pupils, wherein the units of delta x and delta y are pixels. The accuracy of identity authentication can be improved by determining the distance between two pupils of the corresponding person and judging through the distance value.
104, if the characteristic value distance is larger than a preset distance threshold, judging that the personnel information corresponding to the identification information is inconsistent with the personnel information corresponding to the identification information, and intercepting the personnel to be identified;
And 105, if the distance between the characteristic values is smaller than or equal to a preset distance threshold, judging that the personnel to be identified is consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified.
It should be noted that, a person skilled in the art may set the corresponding preset distance threshold according to the actual requirement, so as to facilitate the identification of the face recognition module.
In some embodiments, the process of training the face recognition module may further include: and carrying out wavelet decomposition on the q-th reliable head portrait photo of the p-th person to obtain a first low-frequency face image, carrying out Fourier transformation on the first low-frequency face image, and adopting the amplitude of the first low-frequency face image as the frequency spectrum characteristic X of the head portrait photo. After the face characteristic information in the head portrait photo of the person to be identified is extracted, the person identification method further comprises the following steps:
performing wavelet decomposition on the head portrait photo of the person to be identified, obtaining a second low-frequency face image, performing Fourier transform on the second low-frequency face image, and taking the amplitude of the second low-frequency face image as the frequency spectrum characteristic of the head portrait photo of the person to be identified; respectively determining the similarity dpq between a plurality of first low-frequency face image spectrum feature vectors X and a plurality of second low-frequency face image spectrum feature vectors Y, wherein dpq=cos (X, Y), and the similarity dpq is the cosine value of an included angle between the feature vectors X and the feature vectors Y; determining a value dmax with the maximum similarity value from a plurality of similarity dpq, and judging whether dmax is larger than a preset similarity threshold value or not; if dmax is greater than or equal to a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified; if dmax is smaller than a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and intercepting the personnel to be identified. By calculating the similarity between the first low-frequency face image and the second low-frequency face image, the accuracy of identity authentication can be effectively improved, and the efficiency of safety detection is improved.
By adopting the technical scheme, the technical scheme of the invention has the following beneficial effects:
(1) According to the invention, on the basis of conventional identity information document identification, the head portrait identification algorithm is combined with conventional identity information document identification through the face identification system, and when constructors or related staff enter a construction site and a place which needs related evidence to enter, double verification is performed through the document identification system and the face identification system, so that the accuracy of identity authentication is improved, and meanwhile, compared with the conventional identity identification mode, the time of identity identification is reduced, and the safety detection efficiency is improved.
(2) By calculating the characteristic value distance between key characteristic points of the face, the accuracy of the identity recognition of constructors can be improved, the condition that constructors impound to replace to enter construction sites can be recognized, and the safety in the power grid construction process is improved.
(3) By fusing the two-dimensional face image and the three-dimensional face image, the complete information of key feature points in the face recognition process can be constructed, so that the head images of the faces to be recognized can be recognized quickly and accurately, and the recognition efficiency of a face recognition system is improved.
It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.
In order to realize the personnel identification method based on the power grid safety production, the embodiment of the invention discloses a personnel identification device based on the power grid safety production, fig. 3 is a structural block diagram of the personnel identification device based on the power grid safety production, which is provided by the embodiment of the invention, the personnel identification device is applied to a personnel identification system, the personnel identification system comprises a certificate photo identification system and a face identification system, identity information of a power grid safety production personnel is prestored in the certificate photo identification system, data transmission is carried out between the certificate photo identification system and the face identification system through a local area network, and the personnel identification device comprises:
The certificate information acquisition module 301 is configured to acquire certificate information of a person to be identified through a certificate identification system, send the certificate information to a face recognition system, and compare the certificate information with identity information of registered power grid security production personnel to determine whether the certificate information is truly effective;
the face feature information acquisition module 302 is configured to acquire a head portrait photograph of a person to be identified through a camera and extract face feature information in the head portrait photograph if the certificate information is true and effective, where the face feature information includes key feature points of a face;
the feature value distance processing module 303 is configured to calculate feature value distances between key feature points of the face, and determine whether the feature value distances are greater than a preset distance threshold;
the first person identification module 304 is configured to determine that the person to be identified is inconsistent with the person information corresponding to the certificate information, and intercept the person to be identified if the feature value distance is greater than a preset distance threshold;
and the second personnel identity recognition module 305 is configured to determine that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information, and pass the personnel to be recognized if the distance between the feature values is less than or equal to a preset distance threshold.
In some embodiments, the face recognition system is provided with a face recognition module, and the eigenvalue distance processing module 303 may include:
the key feature point positioning sub-module is used for positioning the input head portrait photo of the person to be identified through the face recognition module and obtaining key feature points of the face;
the pixel distance calculating sub-module is used for measuring and analyzing the key feature points and calculating the pixel distance between two key feature points taking pixels as units;
and the actual distance calculation sub-module is used for determining the actual distance between the two key feature points according to the pixel distance and the conversion coefficient.
In some embodiments, the key feature points include eyes, nose tips, mouth corner points, and eyebrows, and the pixel distance calculation sub-module may include:
the first pixel distance calculating unit is used for identifying the input head portrait photo of the person to be identified through the face recognition module, performing coarse extraction on a face part in the head portrait photo by using a three-room five-eye method, performing horizontal projection and vertical projection on the extracted face part, positioning the accurate position of eyes, and acquiring a corresponding eye image;
And the second pixel distance calculation unit is used for carrying out edge detection and ellipse fitting on the human eye image, taking the circle center of the ellipse as the pupil position of the eye, and obtaining a width difference delta x and a height difference delta y between the two pupils, wherein the units of delta x and delta y are pixels. The accuracy of identity authentication can be improved by determining the distance between two pupils of the corresponding person and judging through the distance value.
In some embodiments, the face recognition system is provided with a face recognition module, and the person recognition system further includes:
the face recognition training module is used for training the face recognition module, and the face recognition training module comprises:
the first training sub-module is used for acquiring identity information of m power grid security producers, wherein the identity information comprises credentials, basic information and n head portrait photos corresponding to the producers; wherein m and n are positive integers;
the second training sub-module is used for preprocessing the q-th reliable head portrait photo of the p-th person, the preprocessing process comprises the steps of converting the certificate photo and the head portrait photo into a face image with a fixed size, and randomly changing the brightness of the face image during preprocessing to enhance data; wherein p is more than 0 and less than m+1, q is more than 0 and less than n+1, and p and q are positive integers;
The third training sub-module is used for carrying out data normalization on the face image with fixed size so that the pixel value of the face image is scaled to be between [ -1,1 ];
a fourth training sub-module, configured to extract face feature information of each face image, where the face feature information includes face key feature points, and calculate feature value distances between the face key feature points;
and the fifth training sub-module is used for associating the characteristic value distance with the identity information of the corresponding production personnel.
In some embodiments, the face recognition training module further comprises:
and the sixth training submodule is used for carrying out wavelet decomposition on the q-th reliable head portrait photo of the p-th person to obtain a first low-frequency face image, carrying out Fourier transform on the first low-frequency face image, and adopting the amplitude of the first low-frequency face image as the frequency spectrum characteristic of the head portrait photo.
In some embodiments, the person identification device further comprises:
the frequency spectrum feature processing module is used for carrying out wavelet decomposition on the head portrait photo of the person to be identified, obtaining a second low-frequency face image, carrying out Fourier transform on the second low-frequency face image, and adopting the amplitude of the second low-frequency face image as the frequency spectrum feature of the head portrait photo of the person to be identified;
The similarity calculation module is used for respectively determining similarity dpq between a plurality of first low-frequency face image spectrum feature vectors X and a plurality of second low-frequency face image spectrum features Y, wherein dpq=cos (X, Y), and the similarity dpq is a cosine value of an included angle between the feature vectors X and Y;
the similarity judging module is used for determining a value dmax with the maximum similarity value from a plurality of similarity dpq and judging whether dmax is larger than a preset similarity threshold value or not;
the third personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if dmax is larger than or equal to a preset similarity threshold value, and releasing the personnel to be recognized;
and the fourth personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if dmax is smaller than a preset similarity threshold value, and intercepting the personnel to be recognized.
In some embodiments, the face recognition training module further comprises:
a seventh training sub-module, configured to obtain a two-dimensional head portrait photo a of a power grid security producer and a three-dimensional image photo B corresponding to the two-dimensional head portrait photo a, record a pixel value at a coordinate (x, y) of the two-dimensional head portrait photo a as a (x 1, y 1), and record a pixel value at a coordinate (x, y) of the three-dimensional image photo B as B (x 2, y 2);
The eighth training submodule is used for fusing the two-dimensional head portrait photo A and the three-dimensional image photo B, obtaining a fused head portrait photo C, and marking the pixel value at the coordinates (x, y) of the head portrait photo C as C (x 3, y 3);
a ninth training submodule, configured to train the face recognition module with a plurality of sets of pixel values C (x 3, y 3);
the pixel value C (x 3, y 3) is calculated as follows:
C(x3,y3)=w(x,y)
A(x1,y1)/>
b (x 2, y 2), w (x, y) is a weighting coefficient when the pixel value a (x 1, y 1) and the pixel value B (x 2, y 2) are fused.
Optionally, when fusing n groups of corresponding two-dimensional head portrait photos a and three-dimensional image photos B, the weighting coefficients w (x, y) are calculated as follows:
wherein A (x i ,y i ) And B (x) i ,y i ) Two-dimensional head portrait and three-dimensional head portrait, respectively, of head portrait photo, w (x i ,y i ) N is a positive integer greater than 0, which is a weighting coefficient corresponding to the head portrait photo.
By adopting the technical scheme, the technical scheme of the invention has the following beneficial effects:
(1) According to the invention, on the basis of conventional identity information document identification, the head portrait identification algorithm is combined with conventional identity information document identification through the face identification system, and when constructors or related staff enter a construction site and a place which needs related evidence to enter, double verification is performed through the document identification system and the face identification system, so that the accuracy of identity authentication is improved, and meanwhile, compared with the conventional identity identification mode, the time of identity identification is reduced, and the safety detection efficiency is improved.
(2) By calculating the characteristic value distance between key characteristic points of the face, the accuracy of the identity recognition of constructors can be improved, the condition that constructors impound to replace to enter construction sites can be recognized, and the safety in the power grid construction process is improved.
(3) By fusing the two-dimensional face image and the three-dimensional face image, the complete information of key feature points in the face recognition process can be constructed, so that the head images of the faces to be recognized can be recognized quickly and accurately, and the recognition efficiency of a face recognition system is improved.
On the other hand, the embodiment of the invention also discloses an electronic device, which comprises: a processor, a memory and a computer program stored on the memory and capable of running on the processor, which computer program, when being executed by the processor, implements the steps of the person identification method described above.
In another aspect, the embodiment of the present invention further discloses a computer readable storage medium, where a computer program is stored, where the computer program is executed by a processor to implement the steps of the person identification method described above.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein, the invention is not limited to the specific embodiments and applications disclosed herein.
Claims (8)
1. The personnel identification method based on the power grid safety production is characterized in that the personnel identification method is applied to a personnel identification system, the personnel identification system comprises a certificate identification system and a face identification system, identity information of power grid safety production personnel is pre-stored in the certificate identification system, data transmission is carried out between the certificate identification system and the face identification system through a local area network, and the personnel identification method comprises the following steps:
acquiring certificate information of a person to be identified through a certificate identification system, sending the certificate information to a face recognition system, comparing the certificate information with identity information of registered power grid safety production personnel, and judging whether the certificate information is true and effective;
if the certificate information is true and effective, acquiring a head portrait photo of a person to be identified through a camera, and extracting face characteristic information in the head portrait photo, wherein the face characteristic information comprises key characteristic points of a face;
Calculating the feature value distance between key feature points of the face of the person, and judging whether the feature value distance is larger than a preset distance threshold value or not;
if the characteristic value distance is larger than a preset distance threshold, judging that the personnel to be identified is inconsistent with personnel information corresponding to the certificate information, and intercepting the personnel to be identified;
if the distance between the characteristic values is smaller than or equal to a preset distance threshold value, judging that the personnel to be identified is consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified;
the face recognition system is provided with a face recognition module, and before the face recognition system is used for carrying out identity recognition, the person recognition method further comprises the following steps:
training the face recognition module;
the process of training the face recognition module comprises the following steps:
acquiring a two-dimensional head portrait photo A of a power grid safety production person and a three-dimensional image photo B corresponding to the two-dimensional head portrait photo A, marking pixel values at coordinates (x, y) of the two-dimensional head portrait photo A as A (x 1, y 1), and marking pixel values at coordinates (x, y) of the three-dimensional image photo B as B (x 2, y 2);
fusing the two-dimensional head portrait photo A and the three-dimensional image photo B to obtain a fused head portrait photo C, and marking the pixel value at the coordinates (x, y) of the head portrait photo C as C (x 3, y 3);
Training a face recognition module by adopting a plurality of groups of pixel values C (x 3, y 3);
the pixel value C (x 3, y 3) is calculated as follows:
C(x3,y3)=w(x,y)
A(x1,y1)/>
B(x2,y2)，w(x,y) Weighting coefficients for fusion of pixel value a (x 1, y 1) and pixel value B (x 2, y 2);
when fusing n groups of corresponding two-dimensional head portrait photos A and three-dimensional image photos B, the weighting coefficients w (x, y) are calculated as follows:
wherein A (x i ,y i ) And B (x) i ,y i ) Two-dimensional head portrait and three-dimensional head portrait, respectively, of head portrait photo, w (x i ,y i ) N is a positive integer greater than 0, which is a weighting coefficient corresponding to the head portrait photo.
2. The method for identifying personnel based on power grid safety production according to claim 1, wherein a face recognition module is arranged on the face recognition system, and the calculating of the feature value distance between key feature points of the face comprises:
positioning the input head portrait photo of the person to be identified through the face recognition module, and acquiring key feature points of the face of the person;
measuring and analyzing the key feature points, and calculating the pixel distance between two key feature points taking pixels as units;
and determining the actual distance between the two key feature points according to the pixel distance and the conversion coefficient.
3. The method for identifying personnel based on power grid security production according to claim 2, wherein the key feature points comprise eyes, nose tips, mouth corner points and eyebrows, and the calculating the pixel distance between two key feature points in pixels comprises:
the face recognition module is used for recognizing the input head portrait photo of the person to be recognized, the three-family five-eye method is used for roughly extracting the face part in the head portrait photo, then horizontal projection and vertical projection are carried out on the extracted face part, the accurate position of eyes is positioned, and the corresponding eye image is obtained;
and performing edge detection and ellipse fitting on the human eye image, taking the circle center of the ellipse as the pupil position of the eye, and obtaining a width difference delta x and a height difference delta y between the two pupils, wherein the units of delta x and delta y are pixels.
4. The method for identifying personnel based on power grid security production of claim 1, wherein the process of training the face recognition module further comprises:
acquiring identity information of m power grid security producers, wherein the identity information comprises credentials, basic information and n photos of the head portrait corresponding to the producers; wherein m and n are positive integers;
Preprocessing the q-th reliable head portrait photo of the p-th person, wherein the preprocessing process comprises the steps of converting the certificate photo and the head portrait photo into a face image with a fixed size, and randomly changing the brightness of the face image during preprocessing to enhance data; wherein p is more than 0 and less than m+1, q is more than 0 and less than n+1, and p and q are positive integers;
carrying out data normalization on the face image with the fixed size, so that the pixel value of the face image is scaled to be between [ -1,1 ];
extracting face characteristic information of each face image, wherein the face characteristic information comprises face key characteristic points, and calculating characteristic value distances among the face key characteristic points;
and correlating the characteristic value distance with the identity information of the corresponding production personnel.
5. The method for identifying personnel based on safe production of power grid according to claim 4, wherein the process of training the face recognition module further comprises:
and carrying out wavelet decomposition on the q-th reliable head portrait photo of the p-th person to obtain a first low-frequency face image, carrying out Fourier transformation on the first low-frequency face image, and adopting the amplitude of the first low-frequency face image as the frequency spectrum characteristic of the head portrait photo.
6. The method for identifying personnel based on power grid security production according to claim 5, wherein after extracting face feature information in the head portrait photo of the personnel to be identified, the method for identifying personnel further comprises:
performing wavelet decomposition on the head portrait photo of the person to be identified, obtaining a second low-frequency face image, performing Fourier transform on the second low-frequency face image, and taking the amplitude of the second low-frequency face image as the frequency spectrum characteristic of the head portrait photo of the person to be identified;
respectively determining the similarity dpq between a plurality of first low-frequency face image spectrum feature vectors X and a plurality of second low-frequency face image spectrum feature vectors Y, wherein dpq=cos (X, Y), and the similarity dpq is the cosine value of an included angle between the feature vectors X and the feature vectors Y;
determining a value dmax with the maximum similarity value from a plurality of similarity dpq, and judging whether dmax is larger than a preset similarity threshold value or not;
if dmax is greater than or equal to a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and releasing the personnel to be identified;
if dmax is smaller than a preset similarity threshold, judging that the personnel to be identified are consistent with personnel information corresponding to the certificate information, and intercepting the personnel to be identified.
7. A personnel identification device based on power grid safety production using the method of any one of claims 1 to 6, wherein the personnel identification device is applied to a personnel identification system, the personnel identification system comprises a certificate identification system and a face identification system, identity information of a power grid safety production personnel is pre-stored in the certificate identification system, data transmission is performed between the certificate identification system and the face identification system through a local area network, and the personnel identification device comprises:
the certificate information acquisition module is used for acquiring the certificate information of the personnel to be identified through the certificate identification system, sending the certificate information to the face recognition system, comparing the certificate information with the identity information of registered power grid safety production personnel, and judging whether the certificate information is true and effective;
the face feature information acquisition module is used for acquiring head portrait photos of the person to be identified through a camera and extracting face feature information in the head portrait photos if the certificate information is true and effective, wherein the face feature information comprises key feature points of face parts;
the characteristic value distance processing module is used for calculating characteristic value distances among key characteristic points of the face of the person and judging whether the characteristic value distances are larger than a preset distance threshold value or not;
The first personnel identity recognition module is used for judging that personnel to be recognized are inconsistent with personnel information corresponding to the certificate information if the characteristic value distance is larger than a preset distance threshold value, and intercepting the personnel to be recognized;
and the second personnel identity recognition module is used for judging that the personnel to be recognized is consistent with the personnel information corresponding to the certificate information if the distance of the characteristic value is smaller than or equal to a preset distance threshold value, and releasing the personnel to be recognized.
8. An electronic device, comprising: a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor performs the steps of the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310619474.3A CN116386121B (en) | 2023-05-30 | 2023-05-30 | Personnel identification method and device based on power grid safety production |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310619474.3A CN116386121B (en) | 2023-05-30 | 2023-05-30 | Personnel identification method and device based on power grid safety production |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116386121A true CN116386121A (en) | 2023-07-04 |
| CN116386121B CN116386121B (en) | 2023-08-11 |
Family
ID=86965998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310619474.3A Active CN116386121B (en) | 2023-05-30 | 2023-05-30 | Personnel identification method and device based on power grid safety production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116386121B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104598878A (en) * | 2015-01-07 | 2015-05-06 | 深圳市唯特视科技有限公司 | Multi-modal face recognition device and method based on multi-layer fusion of gray level and depth information |
| CN108846306A (en) * | 2018-03-28 | 2018-11-20 | 中科博宏（北京）科技有限公司 | A kind of identity card Compare System and method based on deep learning recognition of face |
| CN109902780A (en) * | 2019-02-14 | 2019-06-18 | 广州番禺职业技术学院 | Testimony of a witness unification verification terminal and system and method based on multi-modal recognition of face |
| CN110516649A (en) * | 2019-09-02 | 2019-11-29 | 南京微小宝信息技术有限公司 | Alumnus's authentication method and system based on recognition of face |
| CN113449694A (en) * | 2021-07-24 | 2021-09-28 | 福州大学 | Android-based certificate compliance detection method and system |
| KR20210146185A (en) * | 2020-05-26 | 2021-12-03 | (주)엔에스데블 | UBT system using face contour recognition AI to prevent the cheating behaviour and method thereof |
| US20220292772A1 (en) * | 2021-03-15 | 2022-09-15 | Tencent America LLC | Methods and systems for constructing facial position map |
| CN115100719A (en) * | 2022-07-04 | 2022-09-23 | 施明成 | Face recognition method based on fusion of Gabor binary pattern and three-dimensional gradient histogram features |
| CN115688075A (en) * | 2022-08-30 | 2023-02-03 | 中国工商银行股份有限公司 | Identity verification method and device based on face recognition |
| US20230100427A1 (en) * | 2021-08-20 | 2023-03-30 | Tencent Technology (Shenzhen) Company Limited | Face image processing method, face image processing model training method, apparatus, device, storage medium, and program product |
-
2023
- 2023-05-30 CN CN202310619474.3A patent/CN116386121B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104598878A (en) * | 2015-01-07 | 2015-05-06 | 深圳市唯特视科技有限公司 | Multi-modal face recognition device and method based on multi-layer fusion of gray level and depth information |
| CN108846306A (en) * | 2018-03-28 | 2018-11-20 | 中科博宏（北京）科技有限公司 | A kind of identity card Compare System and method based on deep learning recognition of face |
| CN109902780A (en) * | 2019-02-14 | 2019-06-18 | 广州番禺职业技术学院 | Testimony of a witness unification verification terminal and system and method based on multi-modal recognition of face |
| CN110516649A (en) * | 2019-09-02 | 2019-11-29 | 南京微小宝信息技术有限公司 | Alumnus's authentication method and system based on recognition of face |
| KR20210146185A (en) * | 2020-05-26 | 2021-12-03 | (주)엔에스데블 | UBT system using face contour recognition AI to prevent the cheating behaviour and method thereof |
| US20220292772A1 (en) * | 2021-03-15 | 2022-09-15 | Tencent America LLC | Methods and systems for constructing facial position map |
| CN113449694A (en) * | 2021-07-24 | 2021-09-28 | 福州大学 | Android-based certificate compliance detection method and system |
| US20230100427A1 (en) * | 2021-08-20 | 2023-03-30 | Tencent Technology (Shenzhen) Company Limited | Face image processing method, face image processing model training method, apparatus, device, storage medium, and program product |
| CN115100719A (en) * | 2022-07-04 | 2022-09-23 | 施明成 | Face recognition method based on fusion of Gabor binary pattern and three-dimensional gradient histogram features |
| CN115688075A (en) * | 2022-08-30 | 2023-02-03 | 中国工商银行股份有限公司 | Identity verification method and device based on face recognition |
Non-Patent Citations (2)
| Title |
|---|
| ALI MOEINI等: "2D facial expression recognition via 3D reconstruction and feature fusion", 《JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION》, vol. 35, pages 1 - 14, XP029395052, DOI: 10.1016/j.jvcir.2015.11.006 * |
| 方天红;陈庆虎;廖海斌;邱益鸣;: "融合纹理与形状的人脸加权新特征", 武汉大学学报(信息科学版), no. 03, pages 321 - 326 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116386121B (en) | 2023-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106778586B (en) | Off-line handwritten signature identification method and system | |
| CN106778525B (en) | Identity authentication method and device | |
| CN111460962B (en) | Face recognition method and face recognition system for mask | |
| US10509985B2 (en) | Method and apparatus for security inspection | |
| Chakraborty et al. | An overview of face liveness detection | |
| CN105740780B (en) | Method and device for detecting living human face | |
| Tome et al. | The 1st competition on counter measures to finger vein spoofing attacks | |
| Raghavendra et al. | Scaling-robust fingerprint verification with smartphone camera in real-life scenarios | |
| CN105740779B (en) | Method and device for detecting living human face | |
| KR20170006355A (en) | Method of motion vector and feature vector based fake face detection and apparatus for the same | |
| CN105335726B (en) | Recognition of face confidence level acquisition methods and system | |
| CN107977656A (en) | A kind of pedestrian recognition methods and system again | |
| Alheeti | Biometric iris recognition based on hybrid technique | |
| CN108960088A (en) | The detection of facial living body characteristics, the recognition methods of specific environment | |
| CN106650623A (en) | Face detection-based method for verifying personnel and identity document for exit and entry | |
| Kwaśniewska et al. | Face detection in image sequences using a portable thermal camera | |
| CN110222660B (en) | Signature authentication method and system based on dynamic and static feature fusion | |
| Naveen et al. | Face recognition and authentication using LBP and BSIF mask detection and elimination | |
| KR101151435B1 (en) | Apparatus and method of recognizing a face | |
| CN113436735A (en) | Body weight index prediction method, device and storage medium based on face structure measurement | |
| Deng et al. | Attention-aware dual-stream network for multimodal face anti-spoofing | |
| CN110866442B (en) | Real-time face recognition-based testimony-of-person integrated checking system and method | |
| CN104615985B (en) | A kind of recognition methods of human face similarity degree | |
| CN116386121B (en) | Personnel identification method and device based on power grid safety production | |
| CN115601807A (en) | Face recognition method suitable for online examination system and working method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |