Identity recognition method and device based on touch information, recognition terminal and medium
By combining multiple screenings of fingerprints and facial images, the problem of low identity recognition accuracy in existing technologies is solved, and higher identity authentication accuracy is achieved.
Patent Information
- Application Number
- CN202510931015.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-07
- Publication Date
- 2025-09-23
AI Technical Summary
In the prior art, when verifying identity by punching in, it is impossible to accurately identify whether the operator is the student himself, resulting in low identity recognition accuracy.
A touch-based identity recognition method is adopted. By obtaining the operator's fingerprint image and face image, an initial screening is performed in combination with the fingerprint image library and identity information library, and then a second screening is performed through the face image and identity information library to determine the operator's identity information.
The accuracy of operator identity information verification is improved, ensuring that the identified identity information is more accurate and reducing the possibility of misidentification.
Smart Images
Figure CN120689909A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of identity recognition, and in particular to an identity recognition method, device, identification terminal, and medium based on touch information. Background Art
[0002] The advancement and development of science and technology are playing an increasingly important role in the education industry. As a result, smart classrooms have emerged alongside these technological developments. Smart classrooms are a typical embodiment of a smart learning environment, a high-end form of multimedia and online classrooms, and a new type of classroom built with the help of IoT, cloud computing, and intelligent technologies. In smart classrooms, accurate student identification prevents intruders from disrupting classroom order and effectively ensures the smooth progress of lessons.
[0003] In the relevant technology, identity can be verified by punching in. The target student can punch in on a mobile terminal, and the identity information of the target student corresponding to the puncher is determined by punching in and the punching account. However, when other students hold the target student's mobile device, other students can punch in on the target student's own mobile device instead of the target student. At this time, the puncher will still be identified as the target student. It can be seen that the relevant technology can only verify the identity information corresponding to the punching account, and cannot verify whether the operator who punches in is the student himself. The identity recognition accuracy of the operator who punches in is low. Summary of the Invention
[0004] In order to improve the accuracy of identity recognition, the present application provides an identity recognition method, device, recognition terminal and medium based on touch information.
[0005] In a first aspect, the present application provides an identity recognition method based on touch information, which adopts the following technical solutions: A touch information-based identity recognition method, comprising: Acquire multiple fingerprint images generated when the operator effectively touches the display screen; Determining, from the fingerprint image library, similar fingerprint images corresponding to each fingerprint image based on the multiple fingerprint images and the fingerprint image library, and determining first identity information corresponding to each fingerprint image based on the similar fingerprint images corresponding to each fingerprint image and the first operator identity information library; Acquire the operator's facial image; Determining second identity information corresponding to the operator's facial image based on the operator's facial image and a second operator identity information database; The identity information of the operator is determined based on the first identity information and the second identity information corresponding to each fingerprint image.
[0006] By adopting the above technical solution, multiple fingerprint images generated when the operator effectively touches the display screen are obtained, and similar fingerprint images corresponding to each of the multiple fingerprint images and the fingerprint image library are determined. The first identity information is determined through the similar fingerprint images and the first operator identity information library, so that the operator's identity information is initially screened through the similar fingerprint images; then the operator's facial image is obtained, and the operator's second identity information is determined based on the facial image and the second operator identity information library, so that the operator's identity information is secondly screened through the facial image; and then, based on the first identity information and the second identity corresponding to the fingerprint image, the identity information after the two screenings is determined as the operator's identity information, thereby effectively improving the accuracy of the operator's identity information verification.
[0007] In one possible implementation, the process of confirming a valid touch includes: Acquiring click screen information corresponding to an operator's touch operation on a displayed page, the click screen information at least including: click screen position information; Obtaining area location information corresponding to all operable areas of the display page, where the operable areas are areas where fingerprint collection can be performed; Determining whether the clicked screen position is within the operable area based on the clicked screen position and the area position information corresponding to all operable areas; If so, it is determined that the touch operation corresponding to the click screen information is a valid touch.
[0008] By adopting the above technical solution, the click screen position corresponding to the operator's touch operation on the display page and the area position information corresponding to all the operable areas of the display page are obtained to determine whether it is a valid touch. When the operator's click screen position is within the operable area, the operable area of the display interface can collect the operator's fingerprint image, so the touch information within the operable area is a valid touch; further, the click screen position within the operable area is determined to be a valid touch, which effectively improves the accuracy of collecting fingerprint images.
[0009] In a possible implementation, the screen click information further includes: a screen click duration. If yes, determining that the touch operation corresponding to the screen click information is a valid touch includes: If so, determine whether the screen click duration is greater than a preset screen click duration threshold; If so, it is determined that the touch operation corresponding to the click screen information is a valid touch.
[0010] By adopting the above technology, when the operator accidentally touches the display interface, the corresponding screen click time is shorter. When the operator touches the display interface normally, there is a certain screen click time. Therefore, by judging whether the screen click time is greater than the preset screen click time threshold, the problem of determining the accidental touch operation as a valid touch operation is reduced, thereby effectively improving the accuracy of fingerprint image collection.
[0011] In one possible implementation, obtaining the operator's facial image includes: Acquire multi-angle facial images of the operator; Based on multi-angle facial images, three-dimensional facial modeling is performed to generate a three-dimensional facial model corresponding to the operator, and the facial image of the operator is obtained based on the three-dimensional facial model.
[0012] By adopting the above technical solution, multi-angle facial images of the operator are obtained so that a more comprehensive facial image can be obtained; further, three-dimensional facial modeling is performed and a three-dimensional facial model is generated based on the multi-angle facial images to obtain the operator's facial image. The three-dimensional facial model is closer to people's intuitive feelings, and thus a more accurate facial image can be obtained based on the three-dimensional facial model.
[0013] In one possible implementation, acquiring facial images of the operator from multiple angles includes: Acquire the operator's initial facial images from multiple angles; For multiple facial images corresponding to a certain angle, a preset first feature extraction neural network is used to extract multiple first feature facial images corresponding to the facial images, where the first feature facial images are facial images with feature calibration frames; Obtaining the clarity corresponding to each of the plurality of first feature facial images, and filtering the first feature facial images based on the clarity corresponding to each of the plurality of first feature facial images to obtain a second feature facial image; performing feature extraction on the second feature facial image to determine facial features corresponding to each facial image; If the number of facial features of each facial image is not less than one, determining whether the facial features are identical to any preset facial feature image based on the facial features and a plurality of preset facial key features; If so, it is determined to be a face image at a certain angle corresponding to a valid operator.
[0014] By adopting the above technical solution, a first facial feature image is obtained to calibrate the features in the facial image, and the calibrated facial features are filtered to obtain a second feature facial image to obtain a clearer facial feature image; then, feature extraction is performed on the second feature facial image to obtain accurate facial features; when the number of extracted facial features is not less than 1, it indicates that facial features can be extracted, and it is determined whether the extracted facial features are the required facial features to obtain a valid facial image. By screening valid facial images through facial features, the amount of calculation can be effectively reduced and the calculation efficiency can be improved.
[0015] In one possible implementation, before determining similar fingerprint images corresponding to each fingerprint image from the fingerprint image library based on the multiple fingerprint images and the fingerprint image library, the method further includes: Preprocessing multiple fingerprint images, wherein the preprocessing methods include: image enhancement and binarization processing; Accordingly, the determining, from the fingerprint image library, similar fingerprint images corresponding to each fingerprint image based on the multiple fingerprint images and the fingerprint image library includes: According to the pre-processed multiple fingerprint images and the fingerprint image library, similar fingerprint images corresponding to each fingerprint image are determined from the fingerprint image library.
[0016] By adopting the above technical solution, multiple fingerprint images are enhanced to improve the clarity of the fingerprint images, and then the multiple fingerprint images are binarized to highlight the contours of the fingerprints in the fingerprint images. Similar fingerprint matching is performed based on the clear fingerprint images, effectively improving the accuracy of fingerprint matching.
[0017] In one possible implementation, after determining the identity information of the operator based on the first identity information and the second identity information corresponding to each fingerprint image, the method further includes: Obtain the current classroom identification information and the seat arrangement information corresponding to the current classroom identification information; A seat reminder signal is generated based on the seat entry information and the operator's identity information, and the seat reminder signal is used to remind the operator to take a seat.
[0018] By adopting the above technical solution, the current classroom identification information and the corresponding seat arrangement information are obtained, and a seat reminder signal is generated based on the seat arrangement information and the operator's identity information to remind the operator to reach the correct seat. By generating the seat reminder signal, the operator can enter a fixed seat and further ensure classroom order.
[0019] In a second aspect, the present application provides an identity recognition device based on touch information, which adopts the following technical solution: An identity recognition device based on touch information, comprising: A fingerprint image acquisition module is used to acquire multiple fingerprint images generated when the operator effectively touches the display screen; a first identity information determination module configured to determine, from the fingerprint image library, a similar fingerprint image corresponding to each fingerprint image based on the plurality of fingerprint images and the fingerprint image library, and to determine first identity information corresponding to each fingerprint image based on the similar fingerprint image corresponding to each fingerprint image and the first operator identity information library; A face image acquisition module is used to acquire the face image of the operator; a second identity information acquisition module, configured to determine second identity information corresponding to the operator's facial image based on the operator's facial image and a second operator identity information database; The operator identity information determination module is configured to determine the operator's identity information based on the first identity information and the second identity information corresponding to each fingerprint image.
[0020] On the third side, this application provides an identification terminal, which adopts the following technical solutions: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the identity recognition method based on touch information as described in any one of the first aspects.
[0021] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program thereon, which, when executed in a computer, causes the computer to execute the identity recognition method based on touch information as described in any one of the first aspects.
[0022] In summary, this application has the following beneficial technical effects: The system obtains multiple fingerprint images generated when the operator effectively touches the display screen, determines the corresponding similar fingerprint images according to the multiple fingerprint images and the fingerprint image library, determines the first identity information through the similar fingerprint images and the first operator identity information library, so as to perform a primary screening of the operator's identity information through the similar fingerprint images; then obtains the operator's face image, determines the operator's second identity information based on the face image and the second operator identity information library, so as to perform a second screening of the operator's identity information through the face image; and then determines the identity information after the two screenings as the operator's identity information according to the first identity information and the second identity corresponding to the fingerprint image, thereby effectively improving the accuracy of the operator's identity information verification. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 A schematic flow chart of a touch-based identity recognition method provided in an embodiment of the present application.
[0024] Figure 2 Schematic diagram of the structure of an identity recognition device based on touch information provided in an embodiment of the present application.
[0025] Figure 3 A schematic diagram of the structure of an identification terminal provided in an embodiment of the present application. DETAILED DESCRIPTION
[0026] The following combination Figures 1 to 3 This application is described in further detail.
[0027] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed, but as long as they are within the scope of the present application, they are protected by patent law.
[0028] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0029] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.
[0030] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.
[0031] An embodiment of the present application provides an identity recognition method based on touch information, which is executed by an identification terminal.
[0032] Combine Figure 1 , Figure 1 A schematic flow chart of a touch information-based identity recognition method provided in an embodiment of the present application, wherein the method includes steps S101, S102, S103, S104, and S105, wherein: Step S101: Acquire multiple fingerprint images generated when an operator effectively touches a display screen.
[0033] Specifically, multiple fingerprint images can be obtained through the fingerprint acquisition module in the identification terminal. The fingerprint recognition module is integrated inside the display screen of the identification terminal, and the fingerprint image can be recognized at any position of the display screen, wherein the recognized fingerprint image can be an overall fingerprint image or a local fingerprint image; a valid touch is a touch that can generate a touch instruction, that is, a touch with an obvious operation intention of the operator; and when the identification terminal detects an invalid touch, there is no need to execute the corresponding operation instruction, wherein an invalid touch is that the operator may accidentally touch the display screen due to a large amplitude of the movement. In the embodiment of the present application, the preferred operator is a student.
[0034] Step S102: Based on the multiple fingerprint images and the fingerprint image library, determine similar fingerprint images corresponding to each fingerprint image from the fingerprint image library, and determine first identity information corresponding to each fingerprint image based on the similar fingerprint images corresponding to each fingerprint image and the first operator identity information library.
[0035] Specifically, a plurality of fingerprint images are stored in the fingerprint image library, and the plurality of fingerprint images in the fingerprint image library are pre-stored in the identification terminal. In the embodiment of the present application, in order to improve the recognition accuracy, the fingerprint images in the fingerprint image library are the fingerprint images of ten fingers corresponding to each operator. In the embodiment of the present application, the operator can directly input the corresponding ten fingerprint images into the identification terminal, or upload the fingerprint images on the mobile terminal. The identification terminal can obtain the uploaded fingerprint images from the mobile terminal through wireless transmission or wired transmission. Because the operator may cause different degrees of deformation of the fingerprint image due to finger movement, different orientations of each print, or different points of force when touching the display screen during the process of touching the display screen, and produce blurred fingerprints or incomplete fingerprints. When a blurred fingerprint image is generated, the identification terminal cannot perform fingerprint recognition based on the blurred fingerprint or incomplete fingerprint, and therefore needs to determine the corresponding similar fingerprint image.
[0036] An image matching algorithm can be used to match similar fingerprint images corresponding to each of the multiple fingerprint images from the fingerprint image library. A histogram method, an average hash algorithm, or a PSNR peak signal-to-noise ratio algorithm can be used to determine similar fingerprints. The embodiments of this application do not limit the specific image matching algorithm. It is understood that in the process of matching similar fingerprint images, each fingerprint will be matched to multiple similar fingerprint images. Based on the image matching algorithm, the similarity between each fingerprint image and the corresponding multiple similar fingerprint images can be determined. To improve recognition accuracy, in the implementation of this application, the fingerprint corresponding to the highest similarity is determined as the corresponding similar fingerprint.
[0037] The first operator identity information database stores the identity information of each operator and the corresponding fingerprint image, wherein the operator's identity information can be: the operator's student ID number, the operator's ID number or the operator's mobile phone number. This embodiment of the present application does not limit this. It can be understood that in the first operator identity information database, the operator's student ID number, the operator's ID number and the operator's mobile phone number corresponding to each operator are all uniquely corresponding.
[0038] After determining multiple similar fingerprint images, the identification terminal can identify the similar fingerprints using fingerprint recognition technology and compare the identified fingerprints with the first operator identity information database to determine the first identity information corresponding to each similar fingerprint image, that is, the first identity information corresponding to each fingerprint image. Fingerprint recognition can be performed using a fingerprint recognition model. Fingerprint recognition based on the fingerprint recognition model can specifically include: inputting the fingerprint image to be identified into a convolutional neural network to obtain a fingerprint ridge image of the fingerprint image to be identified; extracting fingerprint features of the fingerprint to be identified from the fingerprint ridge image; and matching the extracted fingerprint features with fingerprint features in the operator information database to determine whether the fingerprint image to be identified is identical to any fingerprint image in the first operator identity information database. During this fingerprint recognition process, the fingerprint image to be identified is considered a similar fingerprint image. If it is determined that the similar fingerprint image is identical to the fingerprint corresponding to an operator in the first operator identity information database, the first identity information corresponding to the similar fingerprint image can be determined.
[0039] Step S103: Acquire the operator's facial image.
[0040] Specifically, a facial image of the operator can be obtained through 3D facial modeling. This can be accomplished by acquiring facial images of the operator from multiple angles, constructing a 3D facial model based on these multi-angle facial images, and obtaining the facial image based on the 3D facial model. 3D facial modeling based on multi-angle facial images can produce a more similar 3D facial model. Furthermore, this more similar 3D facial model can yield a more accurate facial image, effectively improving the accuracy of identity recognition.
[0041] Step S104: Determine second identity information corresponding to the operator's facial image based on the operator's facial image and the second operator identity information database.
[0042] Specifically, the second operator identity information database includes facial information corresponding to multiple operators and second identity information corresponding to each operator, wherein the second identity information can be: the operator's student ID number, the operator's ID number or the operator's mobile phone number, which is not limited in the embodiment of the present application. Each operator corresponds to a unique second identity information, and for the same operator, the identity information in the first operator identity information database is the same as the identity information in the second operator identity information database, which is pre-stored in the identification terminal. The operator's facial image can be uploaded by the operator on the mobile terminal, and the identification terminal obtains the operator's facial image from the mobile terminal through wireless transmission or wired transmission. The identification terminal matches the obtained operator's facial image with the facial images in the second operator identity information database one by one to determine the operator's second identity information; wherein, the facial image can be matched according to image matching technology, wherein the image matching technology is the same as the image matching technology of the similar fingerprint image in step S102, which is not limited in the embodiment of the present application, and the user can set it by himself.
[0043] Step S105: Determine the identity information of the operator based on the first identity information and the second identity information corresponding to each fingerprint image.
[0044] Specifically, the identification terminal compares the identified first identity information and the second identity information to determine whether the first identity information is the same as the second identity information. It can be understood that when the first identity information is the operator's mobile phone number and the second identity information is the operator's student number, the identification terminal can obtain the student number corresponding to the operator's mobile phone number in the first operator identity information database from the first operator identity information database, and determine whether the student number in the first operator identity information database is the same as the operator's student number in the second identity information. If they are the same, it can be determined that the operator's identity information is the identity information corresponding to the student number.
[0045] Based on the above embodiment, multiple fingerprint images generated when the operator effectively touches the display screen are obtained, and similar fingerprint images corresponding to each of the multiple fingerprint images and the fingerprint image library are determined. The first identity information is determined through the similar fingerprint images and the first operator identity information library, so that the operator's identity information is initially screened through the similar fingerprint images; then the operator's facial image is obtained, and the operator's second identity information is determined based on the facial image and the second operator identity information library, so that the operator's identity information is secondly screened through the facial image; and then, based on the first identity information and the second identity corresponding to the fingerprint image, the identity information after the two screenings is determined as the operator's identity information, thereby effectively improving the accuracy of the operator's identity information verification.
[0046] Furthermore, in the embodiment of the present application, the process of confirming a valid touch includes steps S1011 to S1014 (not shown in the drawings), wherein: Step S1011: Acquire the click screen information corresponding to the operator's touch operation on the displayed page, where the click screen information includes: the click screen position.
[0047] Specifically, under normal working conditions, the voltage pressure values corresponding to all positions on the display screen of the identification terminal are the same. When the operator starts to click on the display screen of the identification terminal, and the display screen of the identification terminal just senses the contact, the voltage pressure value corresponding to the clicked screen position begins to increase, and the identification terminal detects the changed voltage pressure value in real time. When the operator's finger leaves the display screen of the identification terminal, the voltage pressure value at the clicked screen position returns to its initial state. Among them, a two-dimensional coordinate system corresponding to the display page can be obtained, and the display interface can be divided into multiple position points with the origin of the two-dimensional coordinate system as the center. Each position point corresponds to an voltage pressure value. When the display screen is not operated, the voltage pressure values corresponding to all position points are the same. When the voltage pressure value corresponding to any point or any area of the display screen changes, and the maximum voltage pressure value is greater than the preset voltage pressure threshold, the clicked screen position can be determined according to the two-dimensional coordinate system. The embodiment of the present application does not limit the preset voltage pressure threshold, and the user can set it by himself.
[0048] Step S1012: Acquire area position information corresponding to all operable areas of the display page, where the operable areas are areas where fingerprint collection can be performed.
[0049] Specifically, the display page is the page presented to the operator by the current identification terminal display screen. Different display pages correspond to different operable areas. Each display page includes: an operable area and an inoperable area. The operable area is: when the operator clicks on the area, the identification terminal can perform corresponding page transformations according to the corresponding information in the operable area. The page transformation can be a change in the entire display screen page, or a text jump and image transformation on the display page; the inoperable area is: when the operator clicks on the area, the display page of the identification terminal does not perform any operation, that is, the display page does not change. Among them, the area position information corresponding to all the operable areas of each display page is pre-stored in the identification terminal. The area position information corresponding to the operable area can be a coordinate set. Each coordinate set includes multiple coordinates. The area position information corresponding to each operable area is different. The area position information of the operable areas corresponding to different display pages can be the same or different. This embodiment of the application is no longer limited.
[0050] Step S1013: Based on the clicked screen position and the area position information corresponding to all operable areas, determine whether the clicked screen position is within the operable area.
[0051] Step S1014: If yes, determine that the touch operation corresponding to the click screen information is a valid touch.
[0052] Specifically, the identification terminal matches the clicked screen position with the area position information corresponding to all operable areas one by one. When the clicked screen position is the same as any of the position information in the operable area, step S014 is executed, and the identification terminal determines it as a valid touch; if the clicked screen position is not within the operable area, it indicates that the touch operation is an invalid touch, that is, the operator accidentally touches the display screen of the identification terminal, or the operator's clicked screen position is within the invalid operation area.
[0053] Based on the above embodiment, the click screen position corresponding to the operator's touch operation on the display page and the area position information corresponding to all the operable areas of the display page are obtained to determine whether it is a valid touch. When the operator's click screen position is within the operable area, the operable area of the display interface can collect the operator's fingerprint image, so the touch information within the operable area is a valid touch; further, the click screen position within the operable area is determined to be a valid touch, which effectively improves the accuracy of collecting fingerprint images.
[0054] Furthermore, in an embodiment of the present application, the screen click information further includes: the duration of the screen click. If yes, determining that the touch operation corresponding to the screen click information is a valid touch includes steps S10141 to S10142 (not shown in the drawings), wherein: Step S10141: If yes, determine whether the screen click duration is greater than a preset screen click duration threshold.
[0055] Step S10142: If yes, determine that the touch operation corresponding to the click screen information is a valid touch.
[0056] Specifically, when the identification terminal detects that the voltage pressure value of the display screen has changed, it starts to count the duration of the screen click until the voltage pressure value at the click screen position returns to the initial voltage pressure value, and then stops counting to obtain the duration of the screen click; the identification terminal determines whether the screen click duration is greater than the preset screen click duration threshold. If so, step S10142 is executed to determine that the touch operation is a valid touch operation; otherwise, the touch operation is determined to be an invalid touch operation. The embodiment of the present application does not limit the preset screen click duration threshold, and the user can set it by himself. It is understandable that by judging the screen click duration, the valid touch operation is re-determined from the perspective of the screen click duration to improve the accuracy of identifying the touch operation.
[0057] Based on the above embodiment, when the operator accidentally touches the display interface, the corresponding screen click time is shorter. When the operator touches the display screen interface normally, there is a certain screen click time. Therefore, by judging whether the screen click time is greater than the preset screen click time threshold, the problem of determining the accidental touch operation as a valid touch operation is reduced, thereby effectively improving the accuracy of fingerprint image acquisition.
[0058] Furthermore, in the embodiment of the present application, obtaining the operator's facial image includes steps S1031 and S1032 (not shown in the drawings), wherein: Step S1031: Acquire facial images of the operator from multiple angles.
[0059] Specifically, facial images of the operator at multiple angles can be obtained based on facial image feature extraction. The feasible method of obtaining facial images of the operator at multiple angles based on facial image feature extraction may specifically include: obtaining initial facial images of the operator at multiple angles; for multiple facial images corresponding to a certain angle, using a preset first feature extraction neural network to extract multiple first feature facial images corresponding to the facial image, the first feature facial image being a facial image with a feature calibration frame; obtaining the clarity corresponding to each of the multiple first feature facial images, filtering the first feature facial image based on the clarity corresponding to each of the multiple first feature facial images to obtain a second feature facial image; performing feature extraction on the second feature facial image to determine the facial features corresponding to each facial image; if the number of facial features of each facial image is not less than 1, determining whether the facial features are the same as any preset facial feature image based on the facial features and multiple preset facial key features; if so, determining it as a facial image of a certain angle corresponding to a valid operator.
[0060] Step S1032: Perform three-dimensional facial modeling based on the multi-angle facial images to generate a three-dimensional facial model corresponding to the operator, and obtain the operator's facial image based on the three-dimensional facial model.
[0061] Specifically, three-dimensional modeling of a face can be performed based on facial key points. The achievable method for performing three-dimensional modeling of a face based on facial key points can specifically include: using a preset first image recognition model to identify the facial key points of the operator, wherein the facial key points are feature points of the facial contours, and the first image recognition model can be a neural network model or a convolutional neural network model obtained by training a large number of training samples of facial images marked with facial key points. The embodiment of the present application is not limited thereto, and the user can set it by himself; obtaining first coordinates of head key points corresponding to each of the multiple head models and second coordinates of facial key points corresponding to each of the multiple head models, wherein the first coordinate system is a coordinate system set according to the head model, and the second coordinate system is a coordinate system set according to the facial image. The process of constructing the head model is: obtaining multiple groups of head bone point parameters, wherein each group of head bone point parameters includes coordinates of multiple bone points that can characterize the head skeleton features, and each bone point is set with a corresponding skin weight. Based on each group of head bone point parameters and the corresponding skin weight, a complete three-dimensional head model can be constructed. Multiple three-dimensional head models constructed using skeletal point parameters are stored as basic head models in a database of a recognition terminal, and skinning weights are corresponding positions of the skeleton and the three-dimensional model points. Multiple first coordinates and second coordinates are matched to determine the degree of matching between the multiple first coordinates and the second coordinates, that is, the coordinates of the key points of the head in the first coordinate system are placed in the second coordinate system, and then the key points of the face are matched with the coordinates of the second coordinate system. For example, the skeletal points of the chin in the basic head model are compared with the feature points of the chin in the facial contour feature points to determine whether the coordinate difference between the skeletal points and the facial contour feature points of the basic head model is within a preset coordinate difference threshold. The more cases in which the coordinate difference falls within the set threshold, the higher the degree of matching. The embodiment of the present application does not limit the coordinate difference threshold. Based on the multiple matching degrees and the corresponding head models, the head model corresponding to the highest matching degree is selected. Based on the multiple valid facial images, a corresponding facial map image is generated. The facial map image is overlaid on the head model to obtain a three-dimensional face model with a skull, and a facial image with a strong sense of three-dimensionality is captured from the three-dimensional face model.
[0062] Based on the above embodiment, facial images of the operator are obtained from multiple angles so that a more comprehensive facial image can be obtained. Furthermore, three-dimensional facial modeling is performed based on the facial images from multiple angles and a three-dimensional facial model is generated to obtain the facial image of the operator. The three-dimensional facial model is closer to people's intuitive feelings, and thus a more accurate facial image can be obtained based on the three-dimensional facial model.
[0063] Furthermore, in the embodiment of the present application, obtaining multi-angle facial images of the operator includes steps SA1 to SA6 (not shown in the drawings), wherein: Step SA1: Acquire initial facial images of the operator from multiple angles.
[0064] Specifically, in an embodiment of the present application, the identification terminal may start acquiring the operator's initial facial images from multiple angles when it detects the operator's effective touch for the first time, or acquire the operator's initial facial images from multiple angles when the distance between the operator and the identification terminal is within a preset distance and the standing time exceeds a preset standing time threshold. The embodiment of the present application does not limit the preset distance and preset standing time thresholds, and users can set them by themselves.
[0065] Step SA2: for multiple facial images corresponding to a certain angle, use a preset first feature extraction neural network to extract multiple first feature facial images corresponding to the facial images, where the first feature facial images are facial images with feature calibration frames.
[0066] Specifically, the first feature facial image is a feature map with multiple calibrated borders, and the calibrated borders are used to calibrate facial features in the facial image. In an embodiment of the present application, the first feature extraction neural network can be a multi-task convolutional neural network, which performs Bounding-Box Regression (border regression vector) on the facial features based on the first feature extraction layer to adjust the borders and uses NMS (non-maximum suppression) to filter most of the borders, that is, merge overlapping borders, thereby obtaining the first feature facial image.
[0067] Step SA3: Obtain the clarity corresponding to each of the plurality of first feature facial images, and filter the first feature facial images based on the clarity corresponding to each of the plurality of first feature facial images to obtain a second feature facial image.
[0068] Specifically, the clarity of each first feature facial image can be obtained by the Tenengrad gradient method or the Laplacian gradient method, and the first feature facial images are sorted in descending order according to the clarity corresponding to each first feature facial image, and a preset number of first feature facial images are filtered out to obtain second feature facial images with higher clarity. The embodiment of the present application does not limit the preset number.
[0069] Step SA4: perform feature extraction on the second feature facial image to determine the facial features corresponding to each facial image.
[0070] Specifically, a feature extraction model can be used to extract features from the second feature facial image, and the second feature facial image can be input into the feature extraction model to obtain features corresponding to each second feature facial image, wherein the feature extraction model is obtained by training a neural network based on multiple feature facial images and facial feature training samples.
[0071] Step SA5: If the number of facial features of each facial image is not less than 1, determine whether the facial features are the same as any preset facial feature image based on the facial features and multiple preset facial key features.
[0072] Step SA6: If yes, determine the face image at a certain angle corresponding to the valid operator.
[0073] Specifically, if the number of facial features in the facial image is not less than 1, it indicates that the facial image carries facial features, and it is determined whether there are facial features that are the same as any preset facial key features. When it is determined that the facial features are not any of the above, step SA6 is executed to determine that it is a valid facial image; otherwise, it is determined that it is not a valid facial image; among which the preset facial features can be: left eye, right eye, nose, left corner of the mouth, right corner of the mouth and eyebrows.
[0074] Based on the above embodiment, a first facial feature image is obtained to calibrate the features in the facial image, and the calibrated facial features are filtered to obtain a second feature facial image to obtain a clearer facial feature image; then, feature extraction is performed on the second feature facial image to obtain accurate facial features; when the number of extracted facial features is not less than 1, it indicates that facial features can be extracted, and it is determined whether the extracted facial features are the required facial features to obtain a valid facial image. By screening the valid facial image by facial features, the amount of calculation can be effectively reduced and the calculation efficiency can be improved.
[0075] Furthermore, in the embodiment of the present application, before determining similar fingerprint images corresponding to each fingerprint image from the fingerprint image library based on the multiple fingerprint images and the fingerprint image library, the method further includes: Preprocessing multiple fingerprint images, wherein the preprocessing methods include: image enhancement and binarization processing; Accordingly, based on the plurality of fingerprint images and the fingerprint image library, determining similar fingerprint images corresponding to each fingerprint image from the fingerprint image library includes: According to the pre-processed multiple fingerprint images and the fingerprint image library, similar fingerprint images corresponding to each fingerprint image are determined from the fingerprint image library.
[0076] Specifically, the identification terminal can enhance the image of multiple fingerprint images through grayscale transformation methods, histogram expansion methods, or filtering. The embodiments of this application do not limit the specific image enhancement technology, and users can set it themselves. The fingerprint image after image enhancement is binarized, which can specifically include grayscale conversion of all pixels, converting color images into grayscale images, and then dividing the fingerprint images into black and white according to a preset pixel threshold, thereby highlighting the fingerprint outline of each fingerprint image.
[0077] Based on the above embodiment, image enhancement is performed on multiple fingerprint images to improve the clarity of the fingerprint images, and then the multiple fingerprint images are binarized to highlight the contours of the fingerprints in the fingerprint images. Similar fingerprint matching is performed based on the clear fingerprint images, effectively improving the accuracy of fingerprint matching.
[0078] Furthermore, in the embodiment of the present application, after determining the identity information of the operator based on the first identity information and the second identity information corresponding to each fingerprint image, the method further includes: Get the current classroom identification information and the seat arrangement information corresponding to the current classroom identification information.
[0079] Specifically, the current classroom identification information is the classroom number. Each classroom corresponds to a unique classroom number. The classroom number and the corresponding seat arrangement information are both pre-stored in the identification terminal. It is understandable that there are different students in the classroom at different times, so it is necessary to arrange the students' seat arrangement information in advance to avoid classroom confusion. An identification terminal is set in front of each classroom, and each identification terminal is set with a unique device number. Each device number uniquely corresponds to the classroom number of each classroom. In an embodiment of the present application, the seat arrangement information can be pre-entered into the identification terminal by the teacher.
[0080] Based on the seat entry information and the operator's identity information, a seat reminder signal is generated, and the seat reminder signal is used to remind the operator to take a seat.
[0081] Specifically, after the identification terminal determines the operator's identity information, it obtains the seat number corresponding to the operator's identity information from the seat arrangement information, and can send the seat number to the operator's mobile terminal, or generate the seat number on the display interface to remind the operator of the seat.
[0082] Based on the above embodiment, the current classroom identification information and the corresponding seat arrangement information are obtained, and a seat reminder signal is generated according to the seat arrangement information and the operator's identity information to remind the operator to reach the correct seat. By generating a seat reminder signal, the operator can enter a fixed seat and further ensure classroom order.
[0083] The above example introduces an identity recognition method based on touch information from the perspective of method flow. The following embodiment introduces an identity recognition device based on touch information from the perspective of a virtual module or virtual unit. Please refer to the following embodiment for details.
[0084] The embodiment of the present application provides an identity recognition device based on touch information, such as Figure 2 As shown, the identity recognition device based on touch information may specifically include: Fingerprint image acquisition module 210, used to acquire multiple fingerprint images generated when the operator effectively touches the display screen; a first identity information determination module 220 for determining, based on the plurality of fingerprint images and the fingerprint image library, a similar fingerprint image corresponding to each fingerprint image from the fingerprint image library, and determining first identity information corresponding to each fingerprint image based on the similar fingerprint image corresponding to each fingerprint image and the first operator identity information library; A facial image acquisition module 230 is used to acquire a facial image of the operator; The second identity information acquisition module 240 is used to determine the second identity information corresponding to the operator's face image based on the operator's face image and the second operator identity information database. The operator identity information determination module 250 is configured to determine the operator's identity information based on the first identity information and the second identity information corresponding to each fingerprint image.
[0085] For the embodiment of the present application, the fingerprint image acquisition module 210 acquires multiple fingerprint images generated when the operator effectively touches the display screen, the first identity information determination module 220 determines the corresponding similar fingerprint images based on the multiple fingerprint images and the fingerprint image library, and determines the first identity information through the similar fingerprint images and the first operator identity information library, so as to perform an initial screening of the operator's identity information through the similar fingerprint images; the face image acquisition module 230 then acquires the operator's face image, and the second identity information acquisition module 240 determines the operator's second identity information based on the face image and the second operator identity information library, so as to perform a second screening of the operator's identity information through the face image; the operator identity information determination module 250 then determines the identity information after two screenings as the operator's identity information based on the first identity information and the second identity corresponding to the fingerprint image, thereby effectively improving the accuracy of the operator's identity information verification.
[0086] In one possible implementation of the embodiment of the present application, the identity recognition device based on touch information further includes: Valid touch confirmation module, used for: Acquire click screen information corresponding to the operator's touch operation on the displayed page, the click screen information including: click screen position; Obtaining the area location information corresponding to all operable areas of the display page, where the operable area is the area where fingerprint collection can be performed; Determine whether the clicked screen position is within the operable area based on the clicked screen position and the area position information corresponding to all operable areas; If so, the touch operation corresponding to the click screen information is determined to be a valid touch.
[0087] In one possible implementation of the embodiment of the present application, when the valid touch confirmation module determines that the touch operation corresponding to the screen click information is a valid touch, it is used to: If so, determine whether the screen click duration is greater than the preset screen click duration threshold; If so, the touch operation corresponding to the click screen information is determined to be a valid touch.
[0088] In one possible implementation of the embodiment of the present application, the facial image acquisition module 230, when acquiring the facial image of the operator, is configured to: Acquire multi-angle facial images of the operator; Based on multi-angle facial images, three-dimensional facial modeling is performed to generate a three-dimensional facial model corresponding to the operator, and the operator's facial image is obtained based on the three-dimensional facial model.
[0089] In one possible implementation of the embodiment of the present application, the facial image acquisition module 230, when acquiring facial images of the operator from multiple angles, is configured to: Acquire the operator's initial facial images from multiple angles; For multiple face images corresponding to a certain angle, a preset first feature extraction neural network is used to extract multiple first feature face images corresponding to the face images, where the first feature face images are face images with feature calibration frames; Obtaining the clarity corresponding to each of the plurality of first feature facial images, and filtering the first feature facial images based on the clarity corresponding to each of the plurality of first feature facial images to obtain a second feature facial image; Performing feature extraction on the second feature facial image to determine facial features corresponding to each facial image; If the number of facial features of each facial image is not less than one, determining whether the facial features are identical to any of the preset facial feature images based on the facial features and a plurality of preset facial key features; If so, it is determined to be a face image at a certain angle corresponding to a valid operator.
[0090] In one possible implementation of the embodiment of the present application, the identity recognition device based on touch information further includes: Image preprocessing module, used for: Preprocessing multiple fingerprint images, wherein the preprocessing methods include: image enhancement and binarization processing; Accordingly, when determining similar fingerprint images corresponding to each fingerprint image from the fingerprint image library based on the multiple fingerprint images and the fingerprint image library, the first identity information determining module 220 is configured to: According to the pre-processed multiple fingerprint images and the fingerprint image library, similar fingerprint images corresponding to each fingerprint image are determined from the fingerprint image library.
[0091] In one possible implementation of the embodiment of the present application, the identity recognition device based on touch information further includes: Seat reminder signal generation module, used for: Obtain the current classroom identification information and the seat arrangement information corresponding to the current classroom identification information; Based on the seat entry information and the operator's identity information, a seat reminder signal is generated, and the seat reminder signal is used to remind the operator to take a seat.
[0092] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the above-described identity recognition device based on touch information can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.
[0093] An identification terminal provided in an embodiment of the present application is introduced below. The identification terminal described below and the identity recognition device based on touch information described above can refer to each other.
[0094] The embodiment of the present application provides an identification terminal, such as Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of an identification terminal provided in an embodiment of the present application. Figure 3 The identification terminal 300 shown includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the identification terminal 300 may further include a transceiver 304. It should be noted that in actual applications, the number of transceivers 304 is not limited to one, and the structure of the identification terminal 300 does not constitute a limitation on the embodiments of the present application.
[0095] Processor 301 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the embodiments disclosed herein. Processor 301 can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.
[0096] Bus 302 may include a path for transmitting information between the above components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.
[0097] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.
[0098] The memory 303 is used to store application code for executing the solution of the embodiment of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the above method embodiment.
[0099] in, Figure 3 The identification terminal shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.
[0100] The embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed on a computer, enables the computer to execute the corresponding contents of the aforementioned method embodiment. Compared with the related art, a plurality of fingerprint images generated when the operator effectively touches the display screen are obtained, similar fingerprint images corresponding to each of the plurality of fingerprint images and a fingerprint image library are determined, and first identity information is determined by using the similar fingerprint images and a first operator identity information library, so as to perform a primary screening of the operator's identity information by using the similar fingerprint images; then, a facial image of the operator is obtained, and the second identity information of the operator is determined based on the facial image and a second operator identity information library, so as to perform a second screening of the operator's identity information by using the facial image; and then, based on the first identity information and the second identity corresponding to the fingerprint image, the identity information after the two screenings is determined as the operator's identity information, thereby effectively improving the accuracy of the operator's identity information verification.
[0101] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
[0102] The above are only some of the implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.
Claims
1. A touch information-based identity recognition method, characterized in that: include: Acquire multiple fingerprint images generated when the operator effectively touches the display screen; Determining, from the fingerprint image library, similar fingerprint images corresponding to each fingerprint image based on the multiple fingerprint images and the fingerprint image library, and determining first identity information corresponding to each fingerprint image based on the similar fingerprint images corresponding to each fingerprint image and the first operator identity information library; Acquire the operator's facial image; Determining second identity information corresponding to the operator's facial image based on the operator's facial image and a second operator identity information database; The identity information of the operator is determined based on the first identity information and the second identity information corresponding to each fingerprint image.
2. The identity recognition method based on touch information according to claim 1, characterized in that: The confirmation process of valid touch includes: Acquiring click screen information corresponding to an operator's touch operation on a displayed page, the click screen information at least including: click screen position information; Obtaining area location information corresponding to all operable areas of the display page, where the operable areas are areas where fingerprint collection can be performed; Determining whether the clicked screen position is within the operable area based on the clicked screen position and the area position information corresponding to all operable areas; If so, it is determined that the touch operation corresponding to the click screen information is a valid touch.
3. The identity recognition method based on touch information according to claim 2, characterized in that: The screen click information also includes: the duration of the screen click; If so, determining that the touch operation corresponding to the screen click information is a valid touch includes: If so, determine whether the screen click duration is greater than a preset screen click duration threshold; If so, it is determined that the touch operation corresponding to the click screen information is a valid touch.
4. The identity recognition method based on touch information according to claim 1, characterized in that: The step of obtaining the operator's facial image includes: Acquire multi-angle facial images of the operator; Based on multi-angle facial images, three-dimensional facial modeling is performed to generate a three-dimensional facial model corresponding to the operator, and the facial image of the operator is obtained based on the three-dimensional facial model.
5. The identity recognition method based on touch information according to claim 4, characterized in that: The acquiring of multi-angle facial images of the operator includes: Acquire the operator's initial facial images from multiple angles; For multiple facial images corresponding to a certain angle, a preset first feature extraction neural network is used to extract multiple first feature facial images corresponding to the facial images, where the first feature facial images are facial images with feature calibration frames; Obtaining the clarity corresponding to each of the plurality of first feature facial images, and filtering the first feature facial images based on the clarity corresponding to each of the plurality of first feature facial images to obtain a second feature facial image; performing feature extraction on the second feature facial image to determine facial features corresponding to each facial image; If the number of facial features of each facial image is not less than one, determining whether the facial features are identical to any preset facial feature image based on the facial features and a plurality of preset facial key features; If so, it is determined to be a face image at a certain angle corresponding to a valid operator.
6. The identity recognition method based on touch information according to claim 1, characterized in that: Before determining similar fingerprint images corresponding to each fingerprint image from the fingerprint image library based on the multiple fingerprint images and the fingerprint image library, the method further includes: Preprocessing multiple fingerprint images, wherein the preprocessing methods include: image enhancement and binarization processing; Accordingly, the determining, from the fingerprint image library, similar fingerprint images corresponding to each fingerprint image based on the multiple fingerprint images and the fingerprint image library includes: According to the pre-processed multiple fingerprint images and the fingerprint image library, similar fingerprint images corresponding to each fingerprint image are determined from the fingerprint image library.
7. The identity recognition method based on touch information according to claim 1, characterized in that: After determining the identity information of the operator based on the first identity information and the second identity information corresponding to each fingerprint image, the method further includes: Obtain the current classroom identification information and the seat arrangement information corresponding to the current classroom identification information; A seat reminder signal is generated based on the seat entry information and the operator's identity information, and the seat reminder signal is used to remind the operator to take a seat.
8. An identity recognition device based on touch information, characterized in that: include: A fingerprint image acquisition module is used to acquire multiple fingerprint images generated when the operator effectively touches the display screen; a first identity information determination module configured to determine, from the fingerprint image library, a similar fingerprint image corresponding to each fingerprint image based on the plurality of fingerprint images and the fingerprint image library, and to determine first identity information corresponding to each fingerprint image based on the similar fingerprint image corresponding to each fingerprint image and the first operator identity information library; A face image acquisition module is used to acquire the face image of the operator; a second identity information acquisition module, configured to determine second identity information corresponding to the operator's facial image based on the operator's facial image and a second operator identity information database; The operator identity information determination module is configured to determine the operator's identity information based on the first identity information and the second identity information corresponding to each fingerprint image.
9. An identification terminal, characterized in that: include: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the identity recognition method based on touch information according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the touch information identity recognition method according to any one of claims 1 to 7.