CN111695434A - Face recognition access control system and recognition method - Google Patents
Face recognition access control system and recognition method Download PDFInfo
- Publication number
- CN111695434A CN111695434A CN202010425462.3A CN202010425462A CN111695434A CN 111695434 A CN111695434 A CN 111695434A CN 202010425462 A CN202010425462 A CN 202010425462A CN 111695434 A CN111695434 A CN 111695434A
- Authority
- CN
- China
- Prior art keywords
- face
- client
- control processor
- module
- definition camera
- 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.)
- Withdrawn
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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a face recognition access control system and a recognition method, relates to the field related to face recognition, and aims to solve the problems that when face data are recorded in the prior art, due to the fact that the heights and postures of users are different, the recording angles are different, and the users need to change face positions repeatedly for recording in order to perfect the face data. The face recognition unit comprises a high-definition camera and an electric control guide rail, the high-definition camera is installed on the electric control guide rail, the output end of the high-definition camera is connected with the input end of the straightening module, the output end of the straightening module is connected with the input end of the control processor, the output end of the control processor is connected with the input end of the electric control guide rail, the control processor is in two-way connection with the storage module, the storage module comprises a temporary storage area and a permanent storage area, and the control processor is in two-way connection with the correction module.
Description
Technical Field
The invention relates to the field related to facial recognition, in particular to a facial recognition access control system and a recognition method.
Background
The information security problem in the current society is concerned by people. Self-security and personal privacy protection are the hot topics of this age. Traditional security measures based on passwords, personal identification codes, magnetic cards, keys and the like cannot completely meet social requirements. Under such a background, people aim at a biological feature recognition technology, and identify or confirm identities by using inherent physiological features or behavior features of human bodies, so that recognition technologies such as palm shape recognition, fingerprint recognition, voiceprint recognition, pupil recognition, face recognition and the like appear. As a new biological feature recognition technology, compared with technologies such as iris recognition, fingerprint scanning and palm shape scanning, the face recognition technology has unique advantages in application: the use is convenient, and the user acceptance is high; the intuition is outstanding; the recognition accuracy is high, and the speed is high; using a universal device; the cost is low, and the popularization and the use are easy. Face recognition uses a general-purpose camera as the identification information acquisition means. The face image of the identification object is acquired in a non-contact mode, and the computer system is compared with the database image after acquiring the image to complete the identification process.
The face recognition is applied to an access control system, generally, a person image is detected in one frame of a video stream and is separated from a background, the captured person image is compared with all person images registered in a database, whether a specified person image exists is searched and searched, and when face data are recorded in the database, due to different recording angles of users, the users need to change face positions for many times for completing the face data and repeatedly record the face data; therefore, the market urgently needs to develop a facial recognition access control system and a recognition method to help people solve the existing problems.
Disclosure of Invention
The invention aims to provide a face recognition access control system and a recognition method, which aim to solve the problems that when face data are recorded, the recording angles are different due to different heights and postures of users, and the users need to change face positions for many times to record repeatedly in order to perfect the face data.
In order to achieve the purpose, the invention provides the following technical scheme: a face recognition access control system comprises a face recognition unit, wherein the face recognition unit comprises a high-definition camera and an electric control guide rail, the high-definition camera is installed on the electric control guide rail, the output end of the high-definition camera is connected with the input end of a straightening module, the output end of the straightening module is connected with the input end of a control processor, the output end of the control processor is connected with the input end of the electric control guide rail, the control processor is bidirectionally connected with a storage module, the storage module comprises a temporary storage area and a permanent storage area, and the control processor is bidirectionally connected with a correction module;
the face recognition unit is used for recognizing face data of the client;
the high-definition camera is a main recognition mechanism of the face recognition unit and is used for shooting the face of a client;
the electric control guide rail is used for adjusting the position of the high-definition camera and realizing the up-and-down movement of the position of the high-definition camera;
the straightening module is used for straightening the face photo of the client;
the control processor is used for processing the data of the system and realizing the smooth operation of the face recognition access control system;
the storage module is used for receiving the data information obtained by the processing of the control processor and realizing the storage of the facial data of the client;
the temporary storage area is used for temporarily storing local face data of the client and realizing the rapid construction of face coordinates of the client;
the permanent storage area is used for storing and inputting a three-dimensional image constructed by the face of the client;
and the correction module is used for correcting the coordinates of the coordinate system constructed by the face of the client so as to match the (X, Y, Z) coordinates.
Preferably, the input end of the control processor is connected with the output end of the high-definition camera, the input end of the control processor is connected with the output end of the permanent storage area, the control processor is bidirectionally connected with the coordinate comparison module, the control processor is bidirectionally connected with the coordinate system construction module, and the output end of the control processor is connected with the input end of the electronic door lock;
the coordinate comparison module is used for comparing the face coordinate input and the face coordinate of the shot photo, and actually comparing the captured face data with the client;
the electronic door lock is an access control mechanism of the face recognition access control system;
and the coordinate system building module builds a coordinate system through the face data of the client and inputs the face data of the client to form a virtual character face model.
Preferably, the moving path of the high-definition camera on the electric control guide rail is in the vertical direction.
Preferably, the output end of the control processor is further connected with the input end of the speaker module.
The identification method of the face identification access control system comprises the following steps of:
the method comprises the following steps: a client stands at the front end of the high-definition camera, the front face faces the camera, the electric control guide rail drives the high-definition camera to move up and down, and the high-definition camera shoots a complete front face picture of the client;
step two: the straightening module identifies the forehead, the eyes, the nose, the mouth and the chin of the face of the client, connects the center positions of the eyes with the forehead, the nose, the mouth and the chin to form a cross net, compares the cross net with a standard vertical cross net, and rotationally cuts the shot picture to straighten the face;
step three: the control processor adds characteristic points to the straightened face data, constructs X-Y quadrant coordinates of each characteristic point by taking the nose as the center, and stores (X, Y) of each characteristic point in a temporary storage area;
step four: the control processor controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, and controls the electric control guide rail to drive the high-definition camera to move up and down to shoot a complete left side face picture of the client;
step five: the straightening module identifies the positions of the neck, the vertebra, the shoulders and the arms of the client, and rotationally cuts the picture of the left face of the client according to the positions so as to straighten the left face;
step six: adding characteristic points to the straightened left face surface data by the control processor, constructing Y-Z quadrant coordinates of each characteristic point by taking the nose as the center, and calling (X, Y) coordinates of each characteristic point in the temporary storage area by the control processor;
step seven: when the Y coordinates of the same feature points of the front face and the left face obtained in the step six are different, the correction module corrects the left side photo by taking the Y coordinate of the front face as a reference, so that the Y coordinates of each feature point are the same, then the Y-Z quadrant coordinates of each feature point are re-established, and the (Y, Z) of each feature point is stored in a temporary storage area;
step eight: when the Y coordinates of the same feature points of the front face and the left face obtained in the step six are the same, directly storing (Y, Z) of each feature point in a temporary storage area;
step nine: the control processor controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, controls the electric control guide rail to drive the high-definition camera to move up and down, shoots a complete picture of the right side face of the client, and repeats the fifth step to the ninth step;
step ten: the control processor calls the (X, Y) coordinates, (Y, Z) of each feature point in the temporary storage areaLeft side of) Coordinates and (Y, Z)Right side) Coordinate, the coordinate system constructing module is used for constructing (X, Y) coordinate, (Y, Z) according to each characteristic pointLeft side of) Coordinates and (Y, Z)Right side) The coordinates construct a three-dimensional model of the face of the virtual character, and the customer data is stored in the permanent storage area together with the constructed three-dimensional model of the face of the virtual character.
Preferably, the face recognition of the face recognition access control system comprises the following steps:
the method comprises the following steps: a user stands in front of the high-definition camera at a free angle, the electric control guide rail drives the high-definition camera to move up and down, and the high-definition camera scans incomplete facial photos of the user;
step two: identifying characteristic points which can be identified by the face of a client, and establishing an incomplete (X, Y, Z) coordinate system according to the existing facial characteristic points;
step three: the coordinate comparison module overlaps the (X, Y, Z) coordinate system of the feature points with the three-dimensional model of the face of the person called by the control processor from the permanent storage area;
step four: if the number of the characteristic points is identified as W, the number of the characteristic points which can be overlapped with the characteristic points in the three-dimensional model of the object face is S, and the control processor processes and calculates the value of S/W;
step five: in the fourth step, if the S/W > =80%, the control processor controls the electronic door lock to be opened;
step six: and in the fourth step, if the S/W is less than 80%, the control processor controls the electronic door lock to be continuously closed.
Preferably, in the third step, when the (X, Y, Z) coordinate system overlaps with the three-dimensional model of the face of the person, two identifiable special feature points are selected as base points, the two special feature points are first nested into the three-dimensional model of the face of the person, if the difference between the positions of the two feature points is large, the two special feature points cannot overlap with the three-dimensional model of the face of the person, the two special feature points are selected again as base points, and if the second special feature points cannot overlap, the subsequent overlapping comparison is not performed; and when the two special feature points are overlapped, other feature points automatically fall into the three-dimensional model of the face of the person, and the overlapped feature points are counted.
Preferably, the special feature points include all special nodal positions relating to eyes, eyebrows, nose, ears and mouth.
Compared with the prior art, the invention has the beneficial effects that:
1. in the invention, a client stands at the front end of a high-definition camera, the front face or the side face faces the camera, an electric control guide rail drives the high-definition camera to move up and down, the high-definition camera shoots a complete front face or side face picture of the client, a straightening module identifies the forehead, eyes, nose, mouth and chin of the face of the client, the center position of the eyes, the forehead, the nose, the mouth and the chin are connected to form a cross net, the cross net is compared with the standard vertical cross net, the shot picture is cut in a rotating way, the front face is straightened, the positions of the neck, the vertebra, the shoulders and the arms of the client are identified by the straightening module, the left side face picture of the client is cut in a rotating way according to the positions, the left side face of the client is straightened, and the left side face is cut in a rotating way, so that the automatic straightening of the straightening module can reduce the trouble that the position, for the client, the entry experience is better;
2. in the invention, the control processor controls the electric control guide rail to drive the high-definition camera to move up and down, so that the problem of facial distortion caused by the fact that the high-definition camera rotates up and down to shoot is avoided, the generation of distortion can be reduced, the pictures shot by the face are standard, and the trouble of recording caused by inconsistent heights of clients can be reduced;
3. in the invention, when the Y coordinates of the same characteristic points of the front face and the left face are different, the correction module can correct the left side photo by taking the Y coordinate of the front face as a reference, so that the Y coordinate of each characteristic point is the same, then the Y-Z quadrant coordinate of each characteristic point is re-established, and finally the (X, Y) coordinate, (Y, Z) coordinate of each characteristic point are obtainedLeft side of) Coordinates and (Y, Z)Right side) The coordinate system is used for constructing a virtual human face three-dimensional model, the Y coordinate of the side face picture is corrected through the Y coordinate of the front face, the side face picture is straightened again, and the obtained Z coordinate is the Z coordinate which accords with the face data of the client, so that the finally generated human face three-dimensional model is more consistent with the actual face of the client, and the door control is opened more accurately by face recognition.
Drawings
FIG. 1 is a system diagram of a face recognition access control system and recognition method according to the present invention;
FIG. 2 is a system schematic diagram of facial entry for a facial recognition access control system of the present invention;
FIG. 3 is a system diagram of the face recognition access control system of the present invention;
figure 4 is a diagram of the straightening module of the present invention for facial orthotics.
In the figure: 1. a face recognition unit; 2. a high-definition camera; 3. an electrically controlled guide rail; 4. a straightening module; 5. a control processor; 6. a storage module; 7. a temporary storage area; 8. a permanent storage area; 9. a rectification module; 10. a coordinate comparison module; 11. an electronic door lock; 12. and a coordinate system building module.
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.
Referring to fig. 1-4, an embodiment of the present invention is shown: a face recognition access control system comprises a face recognition unit 1, wherein the face recognition unit 1 comprises a high-definition camera 2 and an electric control guide rail 3, the high-definition camera 2 is installed on the electric control guide rail 3, the output end of the high-definition camera 2 is connected with the input end of a straightening module 4, the output end of the straightening module 4 is connected with the input end of a control processor 5, the output end of the control processor 5 is connected with the input end of the electric control guide rail 3, the control processor 5 is bidirectionally connected with a storage module 6, the storage module 6 comprises a temporary storage area 7 and a permanent storage area 8, and the control processor 5 is bidirectionally connected with a correction module 9;
wherein, the face recognition unit 1 is used for recognizing the face data of the client;
the high-definition camera 2 is a main recognition mechanism of the face recognition unit 1 and is used for shooting the face of a client; when the high-definition camera 2 is selected, the highest definition is selected so as to control the number of the characteristic points constructed by the processor 5 to be as large as possible;
the electric control guide rail 3 is used for adjusting the position of the high-definition camera 2 and realizing the up-and-down movement of the position of the high-definition camera 2;
the straightening module 4 is used for straightening the face photos of the clients;
the control processor 5 is used for data processing of the system and achieving smooth operation of the face recognition access control system;
the storage module 6 is used for receiving the data information obtained by the processing of the control processor 5 and realizing the storage of the face data of the client; the control processor 5 not only stores the data in the storage module 6, but also sends the data to the background database through the communication module for backup storage;
the temporary storage area 7 is used for temporarily storing local face data of the client and realizing the rapid construction of face coordinates of the client;
the permanent storage area 8 is used for storing and inputting a three-dimensional image constructed by the face of a client;
the temporary storage area 7 and the permanent storage area 8 are the attributes of the storage module 6, most of the storage modules 6 are provided with partitions, and the system separately gives out the partitions for presentation and tells the partitions;
and the correction module 9 is used for correcting the coordinates of the coordinate system constructed by the face of the client so as to match the X, Y and Z coordinates.
Further, the input end of the control processor 5 is connected with the output end of the high-definition camera 2, the input end of the control processor 5 is connected with the output end of the permanent storage area 8, the control processor 5 is bidirectionally connected with the coordinate comparison module 10, the control processor 5 is bidirectionally connected with the coordinate system construction module 12, and the output end of the control processor 5 is connected with the input end of the electronic door lock 11;
the coordinate comparison module 10 is used for comparing the face coordinate input and the face coordinate of the shot photo, and actually comparing the captured face data with the client; the coordinate comparison module 10 overlaps the (X, Y, Z) coordinate system of the feature points with the three-dimensional model of the face of the person called by the control processor 5 from the permanent memory area 8 to determine whether the coming customer is an actual customer;
the electronic door lock 11 is an entrance guard mechanism of the face recognition entrance guard system;
the coordinate system constructing module 12 constructs a coordinate system according to the face data of the client and inputs the face data of the client to form a virtual character face model.
Further, the moving path of the high-definition camera 2 on the electric control guide rail 3 is in the vertical direction.
Further, the output end of the control processor 5 is also connected with the input end of the speaker module.
A recognition method of a face recognition access control system is provided, wherein the face input of the face recognition access control system comprises the following steps:
the method comprises the following steps: a client stands at the front end of the high-definition camera 2, the high-definition camera 2 has a certain distance in the recorded standing position, the client faces the camera when reaching the front face, the electric control guide rail 3 drives the high-definition camera 2 to move up and down, and the high-definition camera 2 shoots a complete front face picture of the client;
step two: the straightening module 4 identifies the forehead, the eyes, the nose, the mouth and the chin of the face of the client, connects the center positions of the eyes with the forehead, the nose, the mouth and the chin to form a cross net, compares the cross net with a standard vertical cross net, and rotationally cuts the shot picture, so that the shot picture is larger for subsequent cutting treatment, and the face is straightened;
step three: the control processor 5 adds feature points to the straightened frontal face data, the feature points are in pixel units, as many as possible, and the X-Y quadrant coordinates of each feature point are constructed by taking the nose as the center (the position of the nose is not easy to change, so that the construction of coordinates by taking the nose as the center point is more accurate and feasible), and stores (X, Y) of each feature point in the temporary storage area 7;
step four: the control processor 5 controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, controls the electric control guide rail 3 to drive the high-definition camera 2 to move up and down, and shoots a complete picture of the left side face of the client;
step five: the straightening module 4 identifies the positions of the neck, the vertebra, the shoulders and the arms of the client, and rotationally cuts the picture of the left face of the client according to the positions to straighten the left face;
step six: the control processor 5 adds characteristic points to the straightened left face data, constructs Y-Z quadrant coordinates of each characteristic point by taking the nose as the center, and the control processor 5 calls the (X, Y) coordinates of each characteristic point in the temporary storage area 7;
step seven: when the Y coordinates of the same feature points of the front face and the left face obtained in the sixth step are different, the correction module 9 corrects the left-side photograph on the basis of the Y coordinate of the front face to make the Y coordinates of each feature point the same, then re-establishes the Y-Z quadrant coordinates of each feature point, and stores (Y, Z) of each feature point in the temporary storage area 7;
step eight: when the Y coordinates of the same feature points of the front face and the left face obtained in the step six are the same, directly storing (Y, Z) of each feature point in a temporary storage area 7;
step nine: the control processor 5 controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, controls the electric control guide rail 3 to drive the high-definition camera 2 to move up and down, shoots a complete picture of the right side face of the client, and repeats the fifth step to the ninth step;
step ten: the control processor 5 calls the (X, Y) coordinates, (Y, Z left) coordinates and (Y, Z right) coordinates of each feature point in the temporary storage area 7, and the coordinate system construction module 12 constructs a virtual character face three-dimensional model based on the (X, Y) coordinates, (Y, Z left) coordinates and (Y, Z right) coordinates of each feature point, and stores the customer information together with the constructed virtual character face three-dimensional model in the permanent storage area 8.
Further, the facial recognition of the facial recognition access control system comprises the following steps:
the method comprises the following steps: a user stands in front of the high-definition camera 2 at a free angle, the electric control guide rail 3 drives the high-definition camera 2 to move up and down, and the high-definition camera 2 scans incomplete facial photos of the client;
step two: identifying characteristic points which can be identified by the face of a client, and establishing an incomplete (X, Y, Z) coordinate system according to the existing facial characteristic points;
step three: the coordinate comparison module 10 overlaps the (X, Y, Z) coordinate system of the feature points with the three-dimensional model of the face of the person called by the control processor 5 from the permanent memory area 8;
the three-dimensional model of the face of the person does not have a fixed coordinate system, is a finished virtual head, and needs to determine (X, Y, Z) of a plurality of characteristic points when being overlapped so as to put a three-dimensional coordinate system generated by all the characteristic points into the virtual head;
step four: if the number of the characteristic points is identified as W and the number which can be overlapped with the characteristic points in the three-dimensional model of the object face is S, the control processor 5 processes and calculates the value of S/W;
step five: in the fourth step, if S/W > =80%, the control processor 5 controls the electronic door lock 11 to open;
step six: in the fourth step, if the S/W is less than 80%, the control processor 5 controls the electronic door lock 11 to be closed continuously.
Further, in the third step, when the (X, Y, Z) coordinate system is overlapped with the three-dimensional model of the face of the person, two identifiable special feature points are selected as base points, the two special feature points are nested into the three-dimensional model of the face of the person, if the position difference of the two feature points is large, the two special feature points cannot be overlapped with the three-dimensional model of the face of the person, the two special feature points are selected as the base points again, and if the special feature points cannot be overlapped for the second time, subsequent overlapping comparison is not carried out; and when the two special feature points are overlapped, other feature points automatically fall into the three-dimensional model of the face of the person, and the overlapped feature points are counted.
And selecting the feature points, wherein the coordinate system can be superposed on the three-dimensional model of the face of the person only by determining the positions of the two feature points, and partial special feature points are exposed when most positions stand.
Further, the special feature points include all special nodal positions related to the eyes, eyebrows, nose, ears, and mouth, such as the corners of the eyes, tails of the eyes, eyebrows, pinna points, and corners of the mouth, which are clearly distinctive points.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (8)
1. The utility model provides a facial recognition access control system, includes facial recognition unit (1), its characterized in that: the face recognition unit (1) comprises a high-definition camera (2) and an electric control guide rail (3), the high-definition camera (2) is installed on the electric control guide rail (3), the output end of the high-definition camera (2) is connected with the input end of a straightening module (4), the output end of the straightening module (4) is connected with the input end of a control processor (5), the output end of the control processor (5) is connected with the input end of the electric control guide rail (3), the control processor (5) is in bidirectional connection with a storage module (6), the storage module (6) comprises a temporary storage area (7) and a permanent storage area (8), and the control processor (5) is in bidirectional connection with a straightening module (9);
wherein, the face recognition unit (1) is used for recognizing the face data of the client;
the high-definition camera (2) is a main recognition mechanism of the face recognition unit (1) and is used for shooting the face of a client;
the electric control guide rail (3) is used for adjusting the position of the high-definition camera (2) and realizing the up-and-down movement of the position of the high-definition camera (2);
a straightening module (4) for straightening the face picture of the customer;
the control processor (5) is used for processing the data of the system and realizing the smooth operation of the face recognition access control system;
the storage module (6) is used for receiving the data information obtained by the processing of the control processor (5) and realizing the storage of the face data of the client;
the temporary storage area (7) is used for temporarily storing local face data of the client and realizing the rapid construction of face coordinates of the client;
the permanent storage area (8) is used for storing and recording a three-dimensional image constructed by the face of the client;
and the correction module (9) is used for correcting the coordinates of the client face construction coordinate system to enable the (X, Y, Z) coordinates to be matched.
2. The facial recognition access control system of claim 1, wherein: the input end of the control processor (5) is connected with the output end of the high-definition camera (2), the input end of the control processor (5) is connected with the output end of the permanent storage area (8), the control processor (5) is bidirectionally connected with the coordinate comparison module (10), the control processor (5) is bidirectionally connected with the coordinate system construction module (12), and the output end of the control processor (5) is connected with the input end of the electronic door lock (11);
the coordinate comparison module (10) is used for comparing the face coordinate input and the face coordinate of the shot photo, and actually comparing the captured face data with the client;
the electronic door lock (11) is an entrance guard mechanism of the face recognition entrance guard system;
and the coordinate system construction module (12) constructs a coordinate system according to the face data of the client and inputs the face data of the client to form a virtual character face model.
3. The facial recognition access control system of claim 1, wherein: the moving path of the high-definition camera (2) on the electric control guide rail (3) is in the vertical direction.
4. The facial recognition access control system of claim 1, wherein: the output end of the control processor (5) is also connected with the input end of the loudspeaker module.
5. The identification method of the facial recognition access control system is characterized in that the facial input of the facial recognition access control system comprises the following steps:
the method comprises the following steps: a client stands at the front end of the high-definition camera (2), the front face faces the camera, the electric control guide rail (3) drives the high-definition camera (2) to move up and down, and the high-definition camera (2) shoots a complete front face picture of the client;
step two: the straightening module (4) identifies the forehead, the eyes, the nose, the mouth and the chin of the face of the client, connects the center positions of the eyes with the forehead, the nose, the mouth and the chin to form a cross net, compares the cross net with a standard vertical cross net, and rotationally cuts the shot picture to straighten the face;
step three: the control processor (5) adds characteristic points to the straightened frontal face data, constructs the X-Y quadrant coordinates of each characteristic point by taking the nose as the center, and stores the (X, Y) of each characteristic point in the temporary storage area (7);
step four: the control processor (5) controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, and controls the electric control guide rail (3) to drive the high-definition camera (2) to move up and down to shoot a complete picture of the left side face of the client;
step five: the straightening module (4) identifies the positions of the neck, the vertebra, the shoulders and the arms of the client, and rotationally cuts the picture of the left face of the client according to the positions to straighten the left face;
step six: the control processor (5) adds characteristic points to the straightened left face data, constructs Y-Z quadrant coordinates of each characteristic point by taking the nose as the center, and calls the (X, Y) coordinates of each characteristic point in the temporary storage area (7);
step seven: when the Y coordinates of the same feature points of the front face and the left face obtained in the step six are different, the correction module (9) corrects the left side photo by taking the Y coordinate of the front face as a reference, so that the Y coordinates of each feature point are the same, then the Y-Z quadrant coordinates of each feature point are re-established, and the (Y, Z) of each feature point is stored in the temporary storage area (7);
step eight: when the Y coordinates of the same feature points of the front face and the left face obtained in the step six are the same, directly storing (Y, Z) of each feature point in a temporary storage area (7);
step nine: the control processor (5) controls the loudspeaker module to make a sound to remind a client to rotate ninety degrees, controls the electric control guide rail (3) to drive the high-definition camera (2) to move up and down, shoots a complete picture of the right side face of the client, and repeats the fifth step to the ninth step;
step ten: the control processor (5) calls the (X, Y) coordinates, (Y, Z) of each feature point in the temporary storage area (7)Left side of) Coordinates and (Y, Z)Right side) A coordinate system building module (12) builds (Y, Z) coordinates from the (X, Y) coordinates of each feature pointLeft side of) Coordinates and (Y, Z)Right side) Coordinates construct a three-dimensional model of the face of the virtual character, and customer data is stored in a permanent storage area (8) together with the constructed three-dimensional model of the face of the virtual character.
6. The method of claim 5, wherein the face recognition of the face recognition access control system comprises the following steps:
the method comprises the following steps: a user stands in front of the high-definition camera (2) at a free angle, the electric control guide rail (3) drives the high-definition camera (2) to move up and down, and the high-definition camera (2) scans incomplete facial pictures of the user;
step two: identifying characteristic points which can be identified by the face of a client, and establishing an incomplete (X, Y, Z) coordinate system according to the existing facial characteristic points;
step three: the coordinate comparison module (10) overlaps the (X, Y, Z) coordinate system of the feature points with the three-dimensional model of the face of the person called by the control processor (5) from the permanent storage area (8);
step four: if the number of the characteristic points is identified to be W, the number which can be overlapped with the characteristic points in the three-dimensional model of the object face part is S, and the control processor (5) processes and calculates the value of S/W;
step five: in the fourth step, if S/W > =80%, the control processor (5) controls the electronic door lock (11) to be opened;
step six: in the fourth step, if the S/W is less than 80%, the control processor (5) controls the electronic door lock (11) to be continuously closed.
7. The identification method of the face recognition access control system according to claim 6, characterized in that: in the third step, when the (X, Y, Z) coordinate system is overlapped with the three-dimensional model of the face of the person, two identifiable special feature points are selected as base points, the two special feature points are nested into the three-dimensional model of the face of the person, if the position difference of the two feature points is large, the two special feature points cannot be overlapped with the three-dimensional model of the face of the person, the two special feature points are selected as the base points again, and if the second special feature points cannot be overlapped, the subsequent overlapping comparison is not carried out; and when the two special feature points are overlapped, other feature points automatically fall into the three-dimensional model of the face of the person, and the overlapped feature points are counted.
8. The identification method of the face recognition access control system according to claim 7, characterized in that: the special feature points include all special nodal positions relating to eyes, eyebrows, nose, ears and mouth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010425462.3A CN111695434A (en) | 2020-05-19 | 2020-05-19 | Face recognition access control system and recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010425462.3A CN111695434A (en) | 2020-05-19 | 2020-05-19 | Face recognition access control system and recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111695434A true CN111695434A (en) | 2020-09-22 |
Family
ID=72477956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010425462.3A Withdrawn CN111695434A (en) | 2020-05-19 | 2020-05-19 | Face recognition access control system and recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695434A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116260901A (en) * | 2023-03-16 | 2023-06-13 | 中国工商银行股份有限公司 | User multi-angle image acquisition method, device and system |
| CN116596466A (en) * | 2023-05-12 | 2023-08-15 | 广州龙信至诚数据科技有限公司 | Government system-based data management and data analysis system and analysis method thereof |
| CN117456584A (en) * | 2023-11-13 | 2024-01-26 | 江苏创斯达智能科技有限公司 | Face recognition equipment applied to intelligent safe |
-
2020
- 2020-05-19 CN CN202010425462.3A patent/CN111695434A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116260901A (en) * | 2023-03-16 | 2023-06-13 | 中国工商银行股份有限公司 | User multi-angle image acquisition method, device and system |
| CN116596466A (en) * | 2023-05-12 | 2023-08-15 | 广州龙信至诚数据科技有限公司 | Government system-based data management and data analysis system and analysis method thereof |
| CN116596466B (en) * | 2023-05-12 | 2024-03-19 | 广州龙信至诚数据科技有限公司 | Government system-based data management and data analysis system and analysis method thereof |
| CN117456584A (en) * | 2023-11-13 | 2024-01-26 | 江苏创斯达智能科技有限公司 | Face recognition equipment applied to intelligent safe |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11074436B1 (en) | Method and apparatus for face recognition | |
| CN109446981B (en) | Face living body detection and identity authentication method and device | |
| CN111695434A (en) | Face recognition access control system and recognition method | |
| WO2021027537A1 (en) | Method and apparatus for taking identification photo, device and storage medium | |
| WO2019104930A1 (en) | Identity authentication method, electronic device and computer-readable storage medium | |
| CN100438580C (en) | Method and apparatus for converting a photo to a caricature image | |
| CN108470169A (en) | Face identification system and method | |
| US6775397B1 (en) | Method and apparatus for user recognition using CCD cameras | |
| JP4406547B2 (en) | ID card creation device, ID card, face authentication terminal device, face authentication device and system | |
| US20050084140A1 (en) | Multi-modal face recognition | |
| KR101810190B1 (en) | User authentication method and apparatus using face identification | |
| US20040218070A1 (en) | Method and apparatus for user recognition using CCD cameras | |
| CN105809415A (en) | Human face recognition based check-in system, method and device | |
| Javed | Face recognition based on principal component analysis | |
| GB2560340A (en) | Verification method and system | |
| JP2006114018A (en) | Security system | |
| US20210224523A1 (en) | Face liveness detection systems and methods | |
| JP3888676B2 (en) | Three-dimensional object recognition apparatus and method | |
| CN107622246A (en) | Face identification method and Related product | |
| CN208351494U (en) | Face identification system | |
| CN114511915B (en) | Trusted certificate photo acquisition system and method based on mobile client | |
| CN107423687B (en) | Identity authentication method and device based on face recognition technology | |
| KR100795360B1 (en) | A Method Of Face Recognizing | |
| CN106980818B (en) | Personalized preprocessing method, system and terminal for face image | |
| US20060056667A1 (en) | Identifying faces from multiple images acquired from widely separated viewpoints |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200922 |