CN114241590B - Self-learning face recognition terminal - Google Patents
Self-learning face recognition terminal Download PDFInfo
- Publication number
- CN114241590B CN114241590B CN202210184848.9A CN202210184848A CN114241590B CN 114241590 B CN114241590 B CN 114241590B CN 202210184848 A CN202210184848 A CN 202210184848A CN 114241590 B CN114241590 B CN 114241590B
- Authority
- CN
- China
- Prior art keywords
- face
- recognition
- identification
- user
- terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a self-learning face recognition terminal, which relates to the technical field of face recognition and comprises a recognition terminal and a control base, wherein a rotary power device is embedded in the control base, the top end of the rotary power device is connected with a support frame, a controller and a control main board are arranged in the recognition terminal, a display screen is arranged on the front side of the recognition terminal, a recognition probe is arranged at the position close to the top surface of the display screen, a flat-scanning laser probe is arranged at the central position of the display screen, the recognition terminal carries out 270-degree recognition on the face of a user and scans the flat-scanning laser probe to provide a data base for three-dimensional modeling, when the recognition terminal is used for recognizing the face of the user, after distance measurement and deflection angles of the face surface are carried out through the flat-scanning laser probe, the angle corresponding to an internal image is compared, the user carries out recognition analysis, and simultaneously, the user carries out designated feature recognition through self-setting of the face recognition features, thereby not only increasing the rapidness and richness when the recognition terminal is used, and the identification terminal is in a self-learning state during identification, so that the identification terminal is convenient to use.
Description
Technical Field
The invention relates to the technical field of face recognition, in particular to a self-learning face recognition terminal.
Background
According to the Chinese patent number CN111444802A, a face recognition method, a face recognition device and an intelligent terminal, the method comprises the following steps: collecting a face picture, cutting the face picture to be uniform in size, converting the face picture into a vector form, and generating a corresponding 0-1 label vector; constructing a predictive judgment dictionary learning model and initializing the predictive judgment dictionary learning model, wherein the model comprises a dictionary learning submodel and a predictive neural network submodel; iteratively optimizing the solution model until convergence; training a prediction neural network submodule by using a self-generating oversampling method in a convergence process; and storing the optimal model, classifying by using the model obtained by training, further acquiring a clear face image matched with the face image to be recognized and corresponding identity information thereof, and outputting a face recognition result of the image to be recognized. Compared with the prior art, especially compared with the technical scheme of deep learning face recognition, the method has higher face recognition rate and time efficiency, and obviously improves the effect of scenes with insufficient sample diversity.
According to the Chinese patent number CN111046837A, the face recognition device based on the 5G framework comprises a face recognition terminal, a 5G transmission network module, a 5G cloud server computing platform and a backup system, wherein the face recognition terminal collects face data through a face collection module and processes the face data; the face recognition terminal comprises a floodlight induction element, a dot matrix projector, an ambient light sensor, a distance sensor and an infrared lens, and 3D face model data are established by the face recognition terminal; the invention relates to the technical field of 5G face recognition. The face recognition equipment based on the 5G architecture is based on the 5G network architecture, and the transmission rate is higher; the cloud computing platform is applied for storage, so that the storage capacity is larger; the cloud server is adopted for storage, the openness and the universality of the system are better, and the backup system is arranged for cloud data storage, so that the data is safer; and 3D face model data identification is adopted, so that the identification rate can be effectively improved, and the confidentiality is better.
The first patent document stores images after the images are identified only in a two-dimensional image form, human faces are prone to being identified inaccurately and being identified wrongly due to the fact that three-dimensional characteristics are prone to occurring during two-dimensional image comparison analysis, the second patent document adopts 3D modeling user information to conduct face auxiliary identification and solves the error problem of two-dimensional identification, and single face angle identification is prone to being affected by light and shadow, makeup and bright and dark faces, so that face identification is inaccurate and not beneficial to actual use.
Disclosure of Invention
Solves the technical problem
Aiming at the defects of the prior art, the invention provides a self-learning face recognition terminal, which solves the problems that the existing face recognition cannot accurately recognize the three-dimensional face of a user, is easily influenced by the external environment during recognition and analysis, and cannot recognize multiple features aiming at single face recognition feature.
Technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme: a self-learning face recognition terminal comprises the following use flows:
sp 1: inputting a face image, aligning an identification probe on the surface of an identification terminal by a user to identify the face at 270 degrees, synchronously scanning data by a flat scanning laser probe, and supplementing a synchronous light source by a light source illuminator on the surface of the identification terminal;
sp 2: data processing and analysis, wherein the inside of the identification terminal performs uniform cutting, analysis, collection and storage on the identified data;
sp 3: the method comprises the following steps of face three-dimensional modeling, namely performing three-dimensional face modeling aiming at face data of data analysis, performing combined analysis on plane data according to a key frame of a 270-degree image of a user, and performing three-dimensional stereo data judgment by combining distance measurement data of a synchronous angle of a flat scanning laser probe;
sp 4: setting characteristics, wherein the setting characteristics comprise designated characteristic identification, random characteristic identification and conventional characteristic identification, and a user selects the characteristics to set;
sp 5: face angle feature recognition, wherein the recognition terminal starts recognition after detecting a face, one of a designated feature recognition mode, a random feature recognition mode and a conventional feature recognition mode is selected according to the face of a user entering a detection range, and exclusive feature recognition is preferentially carried out after the recognition terminal detects exclusive features;
sp 6: the method comprises the steps that a characteristic identification analysis is carried out, after the deflection angle of a user identification area is calculated by the identification terminal through a flat scanning laser probe on the surface, the deflection angle is compared with an image of an internal corresponding angle, the characteristic identification analysis carries out distance measurement scanning by the flat scanning laser probe according to the range of the face of a user entering the identification terminal, the characteristic distance measurement is selected to calculate the angle value of the face of the user at the moment and the deflection of the front face, and the angle value is compared with the image of the internal corresponding angle of the identification terminal;
sp 7: and (4) successfully identifying, namely successfully identifying after the comparative analysis is successful.
Preferably, when the face image is input, the user and the recognition terminal rotate by 270 degrees in sequence to collect the face of the user, the light source lamp on the surface of the recognition terminal is synchronously lighted up to supplement the light source when the face of the user is detected, and when the recognition terminal recognizes the user data, the flat scanning laser probe carries out synchronous face distance measurement processing.
Preferably, in the feature recognition, the designated feature recognition is that three features of a face selected by a user are used as a specific recognition area, the random feature recognition is that the recognition terminal randomly selects three features of a face of the user for recognition, and the conventional feature recognition is that the recognition terminal randomly selects one feature to match with eyes for feature recognition under the condition that two eyes are fixedly selected.
Preferably, the calculation angle of the distance measurement of the flat-scan laser probe is solved for the deflection angle by adopting a triangle solving formula and an angle theorem, and the flat-scan laser probe judges the position and the distance of the distance measurement according to whether feedback is generated after the laser emission contacts an object or not.
Preferably, when the face image is input, the user can adopt a mode of automatically following the dynamic input of the identification terminal and inputting the multi-angle picture.
The utility model provides a self-learning formula face identification terminal, includes identification terminal and control base, its characterized in that: the control base is internally embedded with a rotary power device, the top end of the rotary power device is connected with a support frame, the top end of the support frame is connected with the bottom surface of the identification terminal, the identification terminal is internally provided with a controller and a control mainboard, the control mainboard comprises an image identification module, a storage module, a calculation analysis module, a feature identification module and a data processing module, the front surface of the identification terminal is provided with a light source illuminator near the top surface, the front surface of the identification terminal is provided with a display screen near the top surface, the center of the display screen is provided with a flat scanning laser probe, the image identification module is electrically connected with the identification probe, the identification probe carries out scanning identification on the face of a user through the image identification module, the identified information is synchronously stored in the storage module, and the calculation analysis module 1003 transfers the data information of the corresponding to the user in the storage module in real time to carry out calculation analysis, the analyzed data is subjected to three-dimensional modeling by a data processing module, the modeled data and the identification data of an initial user are uniformly stored in a storage module, a user performs specified feature identification setting by a feature identification module, an identification terminal realizes specified feature identification, random feature identification and conventional feature identification of the user by the feature identification module, when an identification probe performs face identification of the user, the feature identification module performs feature analysis on the face of the user and selects a feature identification mode, an infrared distance sensor is arranged in a light source illuminator, the light source illuminator is automatically turned on to perform face supplementary lighting processing when the face of the user approaches the identification terminal in a set range, a display screen displays the identified face region of the user, the identification probe performs 270-degree scanning input on the face features of the user and performs face feature identification on the user, the laser positioning method comprises the steps that a flat-scan laser probe emits laser to be positioned, when the flat-scan laser probe emits the laser to be used, a two-dimensional plane coordinate system is established by taking the position of the flat-scan laser probe as an original point, the front direction of the position of the flat-scan laser probe is taken as a Y axis, and the horizontal direction is taken as an X axis, so that the laser positioning is carried out.
Advantageous effects
The invention provides a self-learning face recognition terminal. The method has the following beneficial effects:
1. the invention adopts the recognition terminal to perform 270-degree recognition on the face of a user and scan the flat-scan laser probe to provide a data base for three-dimensional modeling, when the face of the user is recognized and used, after the flat-scan laser probe is used for measuring the distance and the deflection angle of the face surface, the angle corresponding to an internal image is compared, recognized and analyzed, and simultaneously, the user performs specified feature recognition by self-setting face recognition features, thereby not only increasing the rapidness and richness of the recognition terminal when in use, but also enabling the recognition terminal to be in a self-learning state when in recognition, and being convenient to use.
2. The invention adopts the recognition terminal to carry out face recognition by supporting three modes of user specified feature recognition and random selection feature recognition while carrying out conventional recognition features of the user, so that the user can still carry out face analysis recognition by adopting facial features after the surface of the face is damaged, the interior of the recognition terminal is always in an active state aiming at the face feature analysis, and the accurate face recognition is convenient for face defect people.
3. The invention adopts the inside of the recognition terminal to carry out three-dimensional modeling data entry on the face of a user, changes the limitation that the front face of the traditional face recognition must be accurately aligned and recognized and the limitation of pupil recognition when carrying out the face recognition of the user, adopts the flat scanning laser probe to carry out the angle recognition and analysis of the face of the user, carries out the recognition and analysis of each angle according to the characteristics of the user, increases the range of the face recognition of the recognition terminal, and leads the recognition terminal to be simpler and quicker when in use.
Drawings
FIG. 1 is a flow chart of an identification terminal of the present invention;
FIG. 2 is a two-dimensional plan view of a flat-scan laser probe of the present invention;
FIG. 3 is a schematic perspective view of the present invention;
FIG. 4 is a front cross-sectional view of the present invention;
fig. 5 is a plan view of the control board of the present invention.
Wherein: 1. a control base; 2. a support frame; 3. a rotary power device; 4. identifying a terminal; 5. a light source lamp; 6. identifying a probe; 7. a display screen; 8. a flat scanning laser probe; 9. a controller; 10. a control main board; 1001. an image recognition module; 1002. a storage module; 1003. a calculation analysis module; 1004. a feature identification module; 1005. and a data processing 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. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The first embodiment is as follows:
as shown in fig. 1 to 5, a self-learning face recognition terminal 4 has the following usage flow:
sp 1: inputting a face image, wherein a user aims at an identification probe 6 on the surface of the identification terminal 4 to identify the face at 270 degrees, a flat scanning laser probe 8 synchronously scans data, and a light source illuminator 5 on the surface of the identification terminal 4 synchronously supplements a light source;
sp 2: data processing and analysis, namely uniformly cutting, analyzing, collecting and storing the identified data in the identification terminal 4;
sp 3: the method comprises the steps of face three-dimensional modeling, namely performing three-dimensional face modeling on face data subjected to data analysis, performing combined analysis on plane data according to a key frame of a 270-degree image of a user through the face three-dimensional modeling, and judging three-dimensional stereo data by combining ranging data of a synchronous angle of a flat scanning laser probe;
sp 4: setting characteristics, wherein the setting characteristics comprise designated characteristic identification, random characteristic identification and conventional characteristic identification, and a user selects the characteristics to set;
sp 5: face angle feature recognition, wherein the recognition terminal 4 starts recognition after detecting a face, one of a designated feature recognition mode, a random feature recognition mode and a conventional feature recognition mode is selected according to the face of a user entering a detection range, and exclusive feature recognition is preferentially carried out after the recognition terminal 4 detects exclusive features;
sp 6: the characteristic identification analysis, after calculating the deflection angle of the user identification area through the flat scanning laser probe 8 on the surface of the identification terminal 4, the identification terminal carries out the comparative analysis with the image of the corresponding internal angle, the characteristic identification analysis carries out the distance measurement scanning according to the range that the face of the user enters the identification terminal 4, the flat scanning laser probe 8 emits laser, the characteristic distance measurement is selected to calculate the angle value between the current range of the face of the user and the front deflection, and the angle value is compared with the image of the corresponding internal angle of the identification terminal 4;
sp 7: and (4) successfully identifying, namely successfully identifying after the comparative analysis is successful.
When a face image is input, a user and an identification terminal 4 rotate 270 degrees in sequence to collect the face of the user, a light source illuminator 5 on the surface of the identification terminal 4 is synchronously lightened to supplement the light source when the face of the user is detected, the face light source is kept when the identification probe 6 and the face of the user are scanned through the light source supplementation in real time, and the light source errors during different-angle identification are reduced, when the identification terminal 4 identifies the user data, a horizontal scanning laser probe 8 is used for synchronous face distance measurement processing, when the face image is input, the user can adopt the modes of dynamic input and multi-angle picture input automatically following the identification terminal 4, when the user records the initial face data, the user can adopt three modes to record the face information, the first mode is that the identification terminal 4 and the user both keep dynamic input, the user and the identification terminal 4 rotate 270 degrees in sequence to carry out face identification input, and the second mode is that the user keeps a standing state, the recognition terminal 4 rotates 270 degrees for recognition and input of the face of the user, the third is that the recognition terminal 4 keeps a static state, the user rotates 270 degrees in situ for face recognition, wherein the recognition terminal 4 in the three modes always keeps a normally-on state when recognizing the face, the flat-scan laser probe 8 is always in a laser distance detection state, the face three-dimensional modeling performs combined analysis of plane data according to a key frame of a 270-degree image of the user, three-dimensional stereo data judgment is performed by combining range data of a synchronous angle of the flat-scan laser probe 8, when the user performs basic face data input, the 270-degree angle comprises all face features in an angle from the front to the back of two side ears of the user, and facial feature recognition is performed on small upper and lower ranges from the chin of the user to the forehead of the user, so that the accuracy of the three-dimensional modeling data is improved, information input by the recognition terminal 4 enters a storage module 1002 inside the control main board 10, the method comprises the steps of performing color processing on an input user photo when the calculation analysis module 1003 performs processing, performing unified gray level processing on an input image while keeping an original color, facilitating analysis of light and shade contrast of each angle image, enabling contrast data of the identification terminal 4 to be an initial input image, a gray level image and a three-dimensional modeling corresponding angle image when face identification contrast is performed, improving identification accuracy of the identification terminal 4, selecting three features of a face as a specific identification area by the user through specified feature identification in feature identification, setting specified features after basic face input by the user, selecting three feature positions of the face as a basis for identification by the user, selecting three features of a right cheekbone, a lip and a right ear for face identification, randomly selecting three features of the face of the user for identification by the identification terminal 4 through random feature identification, and additionally randomly selecting one particular feature for the identification terminal 4 under the condition of fixedly selecting two eyes for identification through conventional feature identification Matching with eyes to perform feature recognition, when a user does not set appointed feature recognition, preferentially selecting the conventional features for recognition by the recognition terminal 4, if the conditions that two eyes required by the conventional recognition feature recognition must exist are not met, switching the recognition terminal 4 into a random feature recognition mode to perform user face recognition, randomly selecting features of three positions of the face of the user by the recognition terminal 4 to perform recognition comparison, performing face recognition by the recognition terminal 4 in a mode of supporting the user to specify the feature recognition and randomly selecting the feature recognition while performing the conventional recognition features of the user, enabling the user to still perform face analysis recognition by the facial features after the surface of the face is damaged, enabling the interior of the recognition terminal 4 to be always in an active state aiming at the face feature analysis, facilitating accurate face recognition of people with facial defects, performing the feature recognition analysis according to the condition that the face of the user enters the recognition range recognized by the recognition terminal 4, the method comprises the steps that a flat scanning laser probe 8 emits laser to carry out ranging scanning, the angle value of the current range and the front deflection of a face of a user is calculated by selecting characteristic ranging, the angle value is compared with an image of a corresponding angle in an identification terminal 4, the identification terminal 4 is adopted to carry out 270-degree identification on the face of the user and the scanning of the flat scanning laser probe 8 provides a data base for three-dimensional modeling, when the face of the user is identified and used, the angle corresponding to an internal image is compared, identified and analyzed after the ranging and deflection angle of the face surface is carried out through the flat scanning laser probe 8, and meanwhile, the user carries out specified characteristic identification through self-setting of face identification characteristics, so that the rapidness and richness of the identification terminal 4 in use are improved, the identification terminal 4 is in a self-learning state in identification, and the use is facilitated.
The distance measurement calculation angle of the flat-scan laser probe 8 adopts a triangle solving formula and an angle theorem to solve the deflection angle, the flat-scan laser probe 8 judges the position and the distance of the distance measurement according to whether feedback is generated after the emitted laser contacts an object, three-dimensional modeling data entry is carried out on the face of a user by adopting the interior of the identification terminal 4, the limitation that the front face is accurately aligned and identified and the limitation of pupil identification are required to be used in the traditional face identification when the face of the user is identified are changed, the angle identification analysis of the face of the user is carried out by adopting the flat-scan laser probe 8, each angle identification analysis is carried out according to the characteristics of the user, the range of the face identified by the identification terminal 4 is increased, the identification terminal 4 is simpler and quicker when in use, the laser scanning is carried out in the area when the flat-scan laser probe 8 is used, and the position of the face of the user is judged according to the feedback generated when the emitted laser contacts the face of the user, the flat scanning laser probe 8 establishes a planar two-dimensional coordinate system according to the position of the flat scanning laser probe 8 to scan and read laser data, the position of the center of the flat scanning laser probe 8 is taken as an O point, the horizontal direction is an X axis, the vertical direction is a Y axis, two characteristic points A and B of the face are picked up, the positions of the two points after being picked up are used for obtaining position points A (Ax, Ay) and B (Bx, By) corresponding to the XY axis according to the coordinate system established By the flat scanning laser, the distance between the two points after being picked up and the O point can be known through laser feedback of the flat scanning laser probe 8, the angle value alpha of an OA straight line and an OB straight line can be known through a laser deflection angle, the straight line distance between characteristic AB points is calculated By utilizing a triangle cosine law, the coordinate point between the two points after being picked up is used for calculating a coordinate point C (Cx, cy), the distance between the laser mark itself and the midpoint, namely the distance between OC, emitted by the flat-scan laser probe 8, thereby obtaining the deflection angle beta between the laser OC between the flat-scan laser probe 8 and the midpoint and the Y-axis of the vertical axis, further leading the identification terminal 4 to fetch the face information of the user corresponding to the deflection angle from the inside for analysis and comparison by calculating the deflection distance between the identified face of the user and the Y-axis of the centerline of the front surface of the flat-scan laser probe 8, wherein the angle error range of the obtained deflection angle beta is beta +/-3 degrees, if the identification terminal 4 is directly adopted to carry out modeling after carrying out multi-angle identification on the face of the user through the identification probe 6, because the three-dimensional characteristics of the face of the user are easily influenced by the conditions of external environment light, cosmetic face dressing, bright face and dark face when being converted into the two-dimensional image data inside the identification terminal 4, the data are generated errors and even wrong, the auxiliary point data setting when the flat-scan laser probe 8 is adopted to carry out the recording of the face of the user, the method can effectively avoid errors of different heights of the face caused by different light and shadow, improve the accuracy of face recognition, because the positions of bones of the face slightly change under the conditions of different emotions and facial expressions of a user, for example, cheekbones of the face move upwards to a certain position when smiling compared with the situation of no expression, the flat scanning laser probe 8 and the recognition probe 6 are always in a working state, the recognition probe 6 realizes a camera monitoring function when not recognizing, the real-time display is carried out through the display screen 7, the recognition probe 6 passes through the user of the recognition terminal 4 through snapshot or camera shooting, the flat scanning laser probe 8 carries out synchronous data measurement value, the synchronous comparison is carried out with the user stored in the recognition probe 6, the various expressions and emotions of the user are convenient to be synchronously grabbed, and the recognition terminal 4 is always in a continuous rich state when not carrying out face recognition, the range and the accuracy during the actual face recognition are increased, so that the recognition terminal 4 is still in the range of accurate recognition when recognizing the user with the changed facial expression.
The second embodiment is as follows:
as shown in FIGS. 3-5, a self-learning face recognition terminal comprises a recognition terminal 4 and a control base 1, wherein a rotary power device 3 is embedded in the control base 1, the top end of the rotary power device 3 is connected with a support frame 2, the top end of the support frame 2 is connected with the bottom surface of the recognition terminal 4, the control base 1 supports the bottom end of the recognition terminal 4 through the support frame 2, the recognition terminal 4 is placed in position through the control base 1, the rotary power device 3 is embedded in the control base 1, and the rotary power device 3 can be a prior art device with a rotation function, such as a servo motor, the rotary power device 3 is connected with the bottom surface of the recognition terminal 4 through the support frame 2, the rotation of the rotary power device 3 drives the recognition terminal 4 to rotate integrally, and the rotary power device 3 provides basic power for 270-degree recognition rotation of a user's face for the recognition device, the identification terminal 4 is internally provided with a controller 9 and a control mainboard 10, the control mainboard 10 comprises an image identification module 1001, a storage module 1002, a calculation analysis module 1003, a feature identification module 1004 and a data processing module 1005, when user information is recorded, the image identification module 1001 is electrically connected with the identification probe 6, the identification probe 6 carries out user face scanning identification through the image identification module 1001, the identified information is synchronously stored in the storage module 1002, the calculation analysis module 1003 invokes data information corresponding to a user in the storage module 1002 in real time for calculation and analysis, the analyzed data is subjected to three-dimensional modeling through the data processing module 1005, the modeled data and identification data of an initial user are uniformly stored in the storage module 1002, the user carries out designated feature identification setting through the feature identification module 1004, and the identification terminal 4 realizes designated feature identification of the user through the feature identification module 1004, Random feature recognition and conventional feature recognition, when the recognition probe 6 is used for recognizing a user face, the feature recognition module 1004 performs feature analysis on the user face and selects a feature recognition mode, a light source illuminator 5 is arranged on the front side of the recognition terminal 4 close to the top surface, a display screen 7 is arranged on the front side of the recognition terminal 4, the recognition probe 6 is arranged on the display screen 7 close to the top surface, a flat scanning laser probe 8 is arranged at the center of the display screen 7, the whole recognition terminal 4 is controlled to operate by a controller 9 and a control main board 10 inside the recognition terminal 4, the light source illuminator 5 provides an illumination basis for the recognition terminal 4 to ensure that the user face is not influenced by external ambient light when scanning or inputting information, an infrared distance sensor is arranged inside the light source illuminator 5, and the light source illuminator 5 is automatically turned on to perform user face light supplement processing when the user face approaches the recognition terminal 4 within a set range, the display screen 7 shows the user facial region of discernment, discernment probe 6 carries out 270 degrees scanning to user's face characteristic and types, and carry out facial feature recognition to the user and use, the laser probe 8 transmission laser of sweeping is swept flatly and is fixed a position, laser probe 8 is swept flatly and when transmission laser used, use the flatly sweep laser probe 8 place and establish two-dimensional plane coordinate system as the initial point, use self position front direction as the Y axle, use the horizontal direction as the X axle, carry out laser positioning and use, be convenient for judge the deflection distance and the user's of discernment user and discernment terminal 4 front deflection angle, thereby transfer the image of inside corresponding deflection angle and carry out the feature comparison, be convenient for carry out quick face identification to user different angles.
It should be noted that, in this document, 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. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other similar elements in a process, method, article, or apparatus that comprises the element.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. A self-learning face recognition terminal is characterized in that: the use process of the identification terminal (4) is as follows:
sp 1: inputting a face image, wherein a user aims at an identification probe (6) on the surface of an identification terminal (4) to identify the face at 270 degrees, a flat scanning laser probe (8) synchronously scans data, and a light source illuminator (5) on the surface of the identification terminal (4) synchronously supplements a light source;
sp 2: data processing and analysis, wherein the inside of the identification terminal (4) performs uniform cutting, analysis, collection and storage on the identified data;
sp 3: the method comprises the following steps of face three-dimensional modeling, namely performing three-dimensional face modeling aiming at face data of data analysis, performing combined analysis on plane data according to a key frame of a 270-degree image of a user, and performing three-dimensional stereo data judgment by combining distance measurement data of a synchronous angle of a flat scanning laser probe (8);
sp 4: setting characteristics, wherein the setting characteristics comprise designated characteristic identification, random characteristic identification and conventional characteristic identification, and a user selects the characteristics to set;
sp 5: the method comprises the steps of face angle feature recognition, wherein the recognition terminal (4) starts recognition after detecting a face, one of a designated feature recognition mode, a random feature recognition mode and a conventional feature recognition mode is selected according to the face of a user entering a detection range, and exclusive feature recognition is preferentially carried out after the recognition terminal (4) detects exclusive features;
sp 6: the method comprises the steps that a characteristic recognition analysis is carried out, the recognition terminal (4) calculates the deflection angle of a user recognition area through a flat scanning laser probe (8) on the surface, then the deflection angle is compared with an image of a corresponding angle in the recognition terminal (4), the flat scanning laser probe (8) emits laser to carry out distance measurement scanning according to the range of the face of a user entering the recognition terminal (4), the angle value of the current range and the front deflection of the face of the user is calculated through characteristic distance measurement, and the angle value is compared with the image of the corresponding angle in the recognition terminal (4);
sp 7: and (4) successfully identifying, namely successfully identifying after the comparison analysis is successful.
2. The self-learning face recognition terminal of claim 1, wherein: when the face image is input, the user and the recognition terminal (4) rotate by 270 degrees in sequence to collect the face of the user, the light source illuminator (5) on the surface of the recognition terminal (4) is synchronously lighted up to supplement the light source when the face of the user is detected, and when the recognition terminal (4) recognizes the user data, the horizontal scanning laser probe (8) carries out synchronous face ranging processing.
3. The self-learning face recognition terminal of claim 1, wherein: the specific feature recognition in the feature recognition is that three features of a face selected by a user are used as specific recognition areas, the random feature recognition is that the recognition terminal (4) randomly selects three features of the face of the user for recognition, and the conventional feature recognition is that the recognition terminal (4) additionally randomly selects one feature to match with eyes for feature recognition under the condition that two eyes are fixedly selected for recognition.
4. The self-learning face recognition terminal of claim 1, wherein: the distance measurement calculation angle of the flat-scan laser probe (8) adopts a triangle solving formula and an angle theorem to solve the deflection angle, and the flat-scan laser probe (8) judges the distance measurement position and distance according to whether feedback is generated after the laser is emitted to contact an object.
5. The self-learning face recognition terminal of claim 1, wherein: when the face image is input, the user can adopt a dynamic input and multi-angle picture input mode of the automatic following recognition terminal (4).
6. The utility model provides a self-learning formula face identification terminal, includes identification terminal (4) and control base (1), its characterized in that: the intelligent control device is characterized in that a rotary power device (3) is embedded in the control base (1), the top end of the rotary power device (3) is connected with a support frame (2), the top end of the support frame (2) is connected with the bottom surface of the identification terminal (4), a controller (9) and a control mainboard (10) are arranged in the identification terminal (4), the control mainboard (10) comprises an image identification module (1001), a storage module (1002), a calculation analysis module (1003), a feature identification module (1004) and a data processing module (1005), a light source illuminator (5) is arranged on the front surface of the identification terminal (4) close to the top surface, a display screen (7) is arranged on the front surface of the identification terminal (4), an identification probe (6) is arranged on the display screen (7) close to the top surface, a flat scanning laser probe (8) is arranged at the center of the display screen (7), and the image identification module (1001) is electrically connected with the identification probe (6), the identification probe (6) scans and identifies the face of a user through an image identification module (1001), the identified information is synchronously stored in a storage module (1002), a calculation and analysis module 1003 invokes data information of the corresponding user in the storage module (1002) in real time to carry out calculation and analysis, the analyzed data is subjected to three-dimensional modeling through a data processing module (1005), the modeled data and the identification data of an initial user are uniformly stored in the storage module (1002), the user carries out specified feature identification setting through a feature identification module (1004), the identification terminal (4) realizes specified feature identification, random feature identification and conventional feature identification of the user through the feature identification module (1004), when the identification probe (6) carries out face identification of the user, the feature identification module (1004) carries out feature analysis on the face of the user and selects to use a feature identification mode, the light source illumination lamp (5) is internally provided with an infrared distance sensor, in a set range, when a user face approaches the identification terminal (4), the light source illumination lamp (5) is automatically started to perform light supplement processing on the user face, the display screen (7) displays the identified user face area, the identification probe (6) performs 270-degree scanning and inputting on the user face characteristics, facial characteristic identification and use are performed on the user, the flat scanning laser probe (8) emits laser to perform positioning, when the flat scanning laser probe (8) emits laser to use, a two-dimensional plane coordinate system is established by taking the position of the flat scanning laser probe (8) as an original point, the self-position front direction is taken as a Y axis, the horizontal direction is taken as an X axis, laser positioning is performed, and the front deflection distance between the identified user and the identification terminal (4) and the deflection angle of the user are conveniently judged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210184848.9A CN114241590B (en) | 2022-02-28 | 2022-02-28 | Self-learning face recognition terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210184848.9A CN114241590B (en) | 2022-02-28 | 2022-02-28 | Self-learning face recognition terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114241590A CN114241590A (en) | 2022-03-25 |
| CN114241590B true CN114241590B (en) | 2022-07-22 |
Family
ID=80748277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210184848.9A Active CN114241590B (en) | 2022-02-28 | 2022-02-28 | Self-learning face recognition terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114241590B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684951A (en) * | 2018-12-12 | 2019-04-26 | 北京旷视科技有限公司 | Face identification method, bottom library input method, device and electronic equipment |
| CN110852310A (en) * | 2020-01-14 | 2020-02-28 | 长沙小钴科技有限公司 | Three-dimensional face recognition method and device, terminal equipment and computer readable medium |
| CN110930556A (en) * | 2019-12-04 | 2020-03-27 | 浙江凯迪仕实业有限公司 | 3D face identification intelligence lock |
| CN211124167U (en) * | 2019-12-31 | 2020-07-28 | 深圳市中福信息科技有限公司 | Face recognition intelligent door |
| CN112528798A (en) * | 2020-12-02 | 2021-03-19 | 广州朗国电子科技有限公司 | Face recognition device capable of automatically adjusting visual angle |
| CN113096261A (en) * | 2021-02-23 | 2021-07-09 | 广州李博士科技研究有限公司 | Face recognition attendance machine |
| CN113673329A (en) * | 2021-07-14 | 2021-11-19 | 深圳市万物互联科技发展有限公司 | 3D face information identification method and device based on big data |
| CN113971833A (en) * | 2021-11-29 | 2022-01-25 | 成都新潮传媒集团有限公司 | Multi-angle face recognition method and device, computer main equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7421097B2 (en) * | 2003-05-27 | 2008-09-02 | Honeywell International Inc. | Face identification verification using 3 dimensional modeling |
| CN104616375A (en) * | 2014-12-30 | 2015-05-13 | 苏州福丰科技有限公司 | Intelligent three-dimensional face identification lock |
| CN107958223B (en) * | 2017-12-13 | 2020-09-18 | 北京小米移动软件有限公司 | Face recognition method and device, mobile equipment and computer readable storage medium |
| CN109377616B (en) * | 2018-10-30 | 2021-02-12 | 南京邮电大学 | Access control system based on two-dimensional face recognition |
| CN112001219B (en) * | 2020-06-19 | 2024-02-09 | 国家电网有限公司技术学院分公司 | Multi-angle multi-face recognition attendance checking method and system |
| CN112613461B (en) * | 2020-12-14 | 2022-11-29 | 南京理工大学 | Intelligent gate communication and attendance checking method and system combining face recognition |
| CN112784796A (en) * | 2021-02-01 | 2021-05-11 | 中科视拓(南京)科技有限公司 | Self-learning non-inductive face recognition system |
-
2022
- 2022-02-28 CN CN202210184848.9A patent/CN114241590B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684951A (en) * | 2018-12-12 | 2019-04-26 | 北京旷视科技有限公司 | Face identification method, bottom library input method, device and electronic equipment |
| CN110930556A (en) * | 2019-12-04 | 2020-03-27 | 浙江凯迪仕实业有限公司 | 3D face identification intelligence lock |
| CN211124167U (en) * | 2019-12-31 | 2020-07-28 | 深圳市中福信息科技有限公司 | Face recognition intelligent door |
| CN110852310A (en) * | 2020-01-14 | 2020-02-28 | 长沙小钴科技有限公司 | Three-dimensional face recognition method and device, terminal equipment and computer readable medium |
| CN112528798A (en) * | 2020-12-02 | 2021-03-19 | 广州朗国电子科技有限公司 | Face recognition device capable of automatically adjusting visual angle |
| CN113096261A (en) * | 2021-02-23 | 2021-07-09 | 广州李博士科技研究有限公司 | Face recognition attendance machine |
| CN113673329A (en) * | 2021-07-14 | 2021-11-19 | 深圳市万物互联科技发展有限公司 | 3D face information identification method and device based on big data |
| CN113971833A (en) * | 2021-11-29 | 2022-01-25 | 成都新潮传媒集团有限公司 | Multi-angle face recognition method and device, computer main equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 三维人脸激光测距特征的提取与识别;张汉萍;《激光杂志》;20180725;第39卷(第07期);第87-91页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114241590A (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7715596B2 (en) | Method for controlling photographs of people | |
| US7660442B2 (en) | Method and system for capturing fingerprints, palm prints and hand geometry | |
| US20110025834A1 (en) | Method and apparatus of identifying human body posture | |
| JP6305171B2 (en) | How to detect objects in a scene | |
| CN109977813A (en) | A kind of crusing robot object localization method based on deep learning frame | |
| CN111012353A (en) | Height detection method based on face key point recognition | |
| CN108562250A (en) | Keyboard cover flatness method for fast measuring based on structure light imaging and device | |
| CN109059798B (en) | Three-dimensional body scanner based on multi-view depth information | |
| KR20120006819A (en) | Gaze detection method and system adopting the same | |
| CN102479322A (en) | System, apparatus and method for analyzing facial defect by facial image with angle | |
| CN111724445A (en) | Method and system for identifying large-view small-size identification code | |
| CN114241590B (en) | Self-learning face recognition terminal | |
| CN111178172A (en) | Laboratory mouse sniffing action recognition method, module and system | |
| KR20180092033A (en) | Component registration device | |
| CN112668539A (en) | Biological characteristic acquisition and identification system and method, terminal equipment and storage medium | |
| JP2004086929A5 (en) | ||
| Al-Qayedi et al. | An algorithm for face and facial-feature location based on grey-scale information and facial geometry | |
| CN111898552A (en) | Method and device for distinguishing person attention target object and computer equipment | |
| CN115414031A (en) | Human body parameter measuring device | |
| CN111967428B (en) | Face temperature measurement method and device and storage medium | |
| CN111210438A (en) | Mirror | |
| CN111967550A (en) | Non-contact temperature detector based on artificial intelligence and detection method thereof | |
| KR100559365B1 (en) | A computer vision based gesture recognition system and method thereof | |
| CN113469169B (en) | Steel cylinder perforation code positioning and identifying system and identifying method | |
| CN116596932B (en) | Method, device, equipment and storage medium for detecting appearance of battery top cover pole |
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 |