WO2021004137A1

WO2021004137A1 - Information pushing method and apparatus based on face recognition and computer device

Info

Publication number: WO2021004137A1
Application number: PCT/CN2020/087133
Authority: WO
Inventors: 张娟
Original assignee: 深圳壹账通智能科技有限公司
Priority date: 2019-07-05
Filing date: 2020-04-27
Publication date: 2021-01-14
Also published as: CN110442783A

Abstract

An information pushing method and apparatus based on face recognition and a computer device. The method comprises: if user information input by an administrator is received, extracting a feature vector from each image of the user information according to a preset vector extraction model to obtain a feature vector set (S110); if a preset information pushing time point is reached, obtaining face information of a current user (S120); recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result about whether the current user is a certain user in the user information (S130); and if the recognition result is that the current user is a certain user in the user information, obtaining, according to the recognition result, target push information matched with the face information from a preset information base to be pushed, so as to push the target push information to the current user (S140). According to the method, the target push information matched with the user information can be obtained on the basis of the face recognition technology, so that information is accurately pushed, and the information push effect is greatly improved.

Description

Information push method, device and computer equipment based on face recognition

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on July 5, 2019, the application number is 201910605454.4, and the invention title is "Information Push Method, Apparatus, and Computer Equipment Based on Face Recognition". The reference is incorporated in this application.

Technical field

This application relates to the field of computer technology, and in particular to an information push method, device, computer equipment and storage medium based on face recognition.

Background technique

Enterprises can commercialize their products by pushing pictures, videos and other information. However, in order to increase the audience's reach, the traditional information push methods are to push pictures and video information in public places with a large number of people, or send information Inserted in TV series, movies and other videos and pushed on TV, computer, mobile phone and other terminals. Although the number of people receiving information is large, the inventor found that this information push method is not targeted and leads to information push effect Due to restrictions, the pushed information cannot be accurately pushed to the target population. Therefore, the existing information push method has a problem that it cannot accurately push information.

Summary of the invention

The embodiments of the application provide an information push method, device, computer equipment and storage medium based on face recognition, which are suitable for the field of artificial intelligence and aim to solve the problem of the inability to accurately push information in the prior art methods .

In the first aspect, an embodiment of the present application provides an information push method based on face recognition, which includes: if user information input by an administrator is received, extracting a model from each photo of the user information according to a preset vector Extract feature vectors in the, to obtain a feature vector set, where the user information includes multiple photos, and each photo corresponds to a user; if the preset information push time point is reached, the face information of the current user is obtained; The preset face recognition model and the feature vector set recognize the face information to obtain the recognition result of whether the current user is a certain user in the user information; if the recognition result is that the current user is A certain user in the user information obtains target push information that matches the face information from a preset to-be-push information database according to the recognition result to push the current user.

In the second aspect, an embodiment of the present application provides an information pushing device based on face recognition, which includes: a feature vector extraction unit, configured to extract a model from all users according to a preset vector if the user information input by the administrator is received The feature vector is extracted from each photo of the user information to obtain a feature vector set, wherein the user information includes a plurality of the photos, and each photo corresponds to a user; the face information acquisition unit is used to obtain a feature vector set. Set the information push time point to obtain the face information of the current user; the face information recognition unit is used to recognize the face information according to the preset face recognition model and the feature vector set to obtain whether the current user is Is the recognition result of a certain user in the user information; the target push information pushing unit is configured to, if the recognition result is that the current user is a certain user in the user information, information to be pushed from a preset according to the recognition result The target push information matching the face information is acquired from the library to push the current user.

In a third aspect, an embodiment of the present application further provides a computer device, which includes: one or more processors; a memory; one or more computer programs, wherein the one or more computer programs are stored in the memory And configured to be executed by the one or more processors, and the one or more computer programs are configured to execute an information push method based on face recognition, wherein the information push based on face recognition The method includes the following steps: if the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain a feature vector set, wherein the user information includes multiple For the photos, each of the photos corresponds to a user; if the preset information push time point is reached, the face information of the current user is obtained; the face information is performed according to the preset face recognition model and the feature vector set Recognition to obtain the recognition result of whether the current user is a certain user in the user information; if the recognition result is that the current user is a certain user in the user information, the information to be pushed from the preset according to the recognition result The target push information matching the face information is acquired from the library to push the current user.

In a fourth aspect, the embodiments of the present application also provide a computer-readable storage medium with a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, a type of information push based on face recognition is realized The method, wherein the information push method based on face recognition includes the following steps: if the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain A feature vector set, wherein the user information includes a plurality of the photos, each of the photos corresponds to a user; if the preset information push time point is reached, the current user's face information is obtained; according to the preset face recognition model And the feature vector set to recognize the face information to obtain the recognition result of whether the current user is a certain user in the user information; if the recognition result is that the current user is a certain user in the user information The user obtains target push information that matches the face information from a preset to-be-push information database according to the recognition result to push the current user.

The embodiment of the present application can obtain target push information matching user information based on the face recognition technology through the above method, so as to realize accurate push of information and greatly improve the effect of pushing information.

Description of the drawings

FIG. 1 is a schematic flowchart of an information push method based on face recognition provided by an embodiment of the application;

FIG. 2 is a schematic diagram of a sub-process of a method for pushing information based on face recognition provided by an embodiment of the application;

FIG. 3 is a schematic diagram of another sub-flow of the method for pushing information based on face recognition according to an embodiment of the application;

4 is a schematic diagram of another process of the method for pushing information based on face recognition provided by an embodiment of the application;

FIG. 5 is a schematic diagram of another sub-flow of the method for pushing information based on face recognition according to an embodiment of the application;

FIG. 6 is a schematic block diagram of an information pushing device based on face recognition provided by an embodiment of the application;

Fig. 7 is a schematic block diagram of a subunit of a face recognition-based information pushing device provided by an embodiment of the application;

FIG. 8 is a schematic block diagram of another subunit of the information pushing device based on face recognition according to an embodiment of the application;

FIG. 9 is another schematic block diagram of an information push device based on face recognition provided by an embodiment of the application;

FIG. 10 is a schematic block diagram of another subunit of the information pushing device based on face recognition according to an embodiment of the application;

FIG. 11 is a schematic block diagram of a computer device provided by an embodiment of the application.

Detailed ways

Please refer to FIG. 1, which is a schematic flowchart of a method for pushing information based on face recognition provided by an embodiment of the present application. The information push method based on face recognition is applied to a user terminal, and the method is executed by application software installed in the user terminal. The user terminal is used to execute the information push method based on face recognition to push information Terminal equipment. The user terminal is a terminal equipment with camera or video function. The above-mentioned information push method based on face recognition can be applied to vehicles that have been set up to push user terminals such as pictures and videos, such as high-speed trains, airplanes, etc. . As shown in Figure 1, the method includes steps S110 to S140.

S110: If the user information input by the administrator is received, extract a feature vector from each photo of the user information according to a preset vector extraction model to obtain a feature vector set.

If the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain a feature vector set. User information can be uniformly input to the user terminal by the administrator. The administrator is the person who inputs user information and manages the pushed information. The administrator can be the staff on the plane, the staff on the high-speed rail, etc. The user needs to provide detailed information when ticket information). Among them, the input user information includes multiple users' information, a user's information includes a photo of the user (such as the user's ID photo) and the user's personal information, and the user's personal information includes gender , Age, education, marriage, occupation and residence information. Through the vector extraction model, the feature vector corresponding to each user can be obtained from the photos contained in the user information, and all feature vectors can be obtained to obtain the feature vector set. Specifically, the vector extraction model includes vector conversion rules and feature vector calculations Formula, the feature vector set contains the feature vector corresponding to each photo in the user information. The feature vector can be used to digitize each photo in the user information. Based on the above feature vector set, the photos of other users are calculated and analyzed. It is the specific process of face recognition.

In one embodiment, as shown in FIG. 2, step S110 includes sub-steps S111, S112, and S113.

S111. Decolorize each photo in the user information to obtain a photo collection.

Decolorization processing is performed on each photo in the user information to obtain a photo collection. The photos included in the user information are all color photos. In order to facilitate subsequent processing of the photos, while keeping the information in the photos to the maximum extent, the photos can be decolorized to obtain decolorized photos containing only grayscale pixels. After all photos are decolorized, the photo collection can be obtained.

S112: Convert the photos in the photo set according to the vector conversion rule in the vector extraction model to obtain a vector set.

The photos in the photo collection are converted according to the vector conversion rules in the vector extraction model to obtain a vector collection. Specifically, the vector conversion rule is the conversion rule used to convert the decolorized photos in the photo collection to obtain a one-dimensional vector. The decolorized photo contains multiple pixels, and each pixel corresponds to a gray value. That is, each pixel in the decolorized photo can be represented by a numerical value. By obtaining the gray value of each pixel from the decolorized photo in a row-by-row or column-by-column manner, a one-dimensional vector can be obtained, and one-dimensional vectors corresponding to all photos in the photo set can be obtained to obtain a vector set S. If a vector set contains M one-dimensional vectors, it can be represented by S={C ₁ , C ₂ ,..., C _M }, where C is the one-dimensional vector contained in the vector set.

Among them, the gray value is represented by a non-negative integer, the value range of the corresponding gray value of the pixel is [0, 255], a gray value of 0 indicates that the pixel is black, and a gray value of 255 indicates that the pixel If the point is white, and the gray value is other values, the pixel point is a specific gray level between white and black.

For example, a decolorized photo is composed of a×b pixels, a is the number of pixels included in the horizontal direction, and b is the number of pixels included in the vertical direction, then the number of gray values contained in the corresponding one-dimensional vector is N=a×b, that is, the one-dimensional vector corresponding to the decolorized photo is C1={c1, c2,..., cN}.

S113: Calculate the feature vector corresponding to each photo according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.

The feature vector corresponding to each photo is calculated according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set. Specifically, the feature vector calculation formula is a formula used to calculate the vector set to obtain the feature vector corresponding to each photo. The steps for calculating the feature vector corresponding to each photo are: (1) Calculate the one-dimensional contained in the vector set S The average value of the vector to obtain the one-dimensional average vector Ψ,

The obtained Ψ is a 1×N (1 row and N column) one-dimensional vector; (2) Calculate the difference between each one-dimensional vector in the vector set S and the above-mentioned one-dimensional average vector Ψ, Φ _i = C _i －Ψ, i=1, 2, 3,..., M, the corresponding vector matrix Φ={Φ ₁ , Φ ₂ ,..., Φ _M } can be obtained; the obtained Φ _i is a 1×N one-dimensional Vector, the obtained Φ is a vector matrix with M rows and N columns; (3) Calculate M orthogonal unit vectors u _i , where i = 1, 2, 3,..., M, the unit vector can be used to describe The distribution of each Φ _i . u _i can be obtained by calculating the covariance matrix, the covariance matrix X=Φ ^T ×Φ, where X={x ₁ , x ₂ ,..., x _M }, Φ ^T is the vector matrix after transforming Φ, Φ Is a vector matrix with M rows and N columns, then Φ ^T is a vector matrix with N rows and M columns, where X is a vector matrix with M rows and M columns. The eigenvector U of the covariance matrix is calculated, where U={ u ₁ , u ₂ ,..., u _M }, u _i is a vector of M×1, and u ^T _i ×u _i =1, u ^T _i ×u _k =0 (i≠k), i=1, 2,3, ..., M; (4 ) calculate each vector corresponding eigenvectors Φ _i Ω _{_i,} Ω _i that is, with the picture corresponding to each feature _{^{_{vector, Ω i = u T i ×}}} Φ i, i=1, 2, 3,..., M, and the obtained Ω _i is a 1×N one-dimensional vector. By obtaining the feature vector corresponding to each photo, a feature vector set Ω={Ω ₁ , Ω ₂ ,..., Ω _M } can be formed.

S120: If the preset information push time point is reached, obtain the face information of the current user.

If it arrives at the preset information push time point in the user terminal, the face information of the current user is obtained. The preset information push time point can be a certain time node when the user uses the application in the user terminal, for example, the information push time point preset by the user before playing the video through the video player program, or the time point preset in the terminal display on the flight or high-speed rail. The set information push time point. When the preset information push time point is reached, the current user’s face information can be acquired through the image acquisition device in the user terminal. The image acquisition device can acquire pictures or videos. The image acquisition device can be a camera in the user terminal. A picture obtained or a picture intercepted from a video is used as the face information of the current user.

S130: Recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information.

Recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result. The face recognition model is a model used to identify whether the acquired face information is a certain user in the user information. The face recognition model includes vector extraction rules, difference calculation formulas and difference thresholds.

In an embodiment, as shown in FIG. 3, step S130 includes sub-steps S131, S132, and S133.

S131. Obtain a face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model.

Obtain the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model. Specifically, the face information includes an acquired picture, and the vector extraction rules can extract the face feature vector corresponding to the face information from the picture. The specific steps include cropping the size of the photo To obtain photos with the same size as the photos in the photo collection, perform decolorization processing on the photos after size adjustment to obtain decolorized photos corresponding to face information, and extract them from decolorized photos corresponding to face information according to the vector extraction rules The face feature vector Ω _x corresponding to the photo is obtained, and the face feature vector Ω _x is a 1×N one-dimensional vector. The decolorized photo corresponding to the face information contains multiple pixels, and each pixel corresponds to a gray value. Through the gray value, each pixel in the decolorized photo can be represented by a numerical value, and the facial feature vector is extracted The specific process is the same as the step of obtaining the one-dimensional vector corresponding to each photo from the photo collection, and will not be repeated here.

S132: Calculate the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model.

The difference value between the face feature vector and each feature vector in the feature vector set is calculated according to the difference calculation formula in the face recognition model. The difference calculation formula is the calculation formula used to calculate the difference between the face feature vector and each feature vector in the feature vector set. The difference between the face feature vector and the feature vector can be used to determine the difference between the face information and the feature vector. The difference is quantified. The smaller the difference value, the more similar the face information and the photo corresponding to the feature vector. The difference calculation formula can be expressed as ε _i =||Ω _x- Ω _i || ² , Ω _{x is the} face feature vector, ε _{i is} the difference value between the face feature vector and the i-th feature vector in the feature vector set, and Ω _i is the feature The i-th feature vector in the vector set. The face feature vector Ω _x is a 1×N one-dimensional vector, and Ω _i is also a 1×N one-dimensional vector. The difference calculation formula is to calculate the Euclidean distance of the above two vectors.

S133. Judging the difference value corresponding to each feature vector according to the difference threshold in the face recognition model to obtain the recognition result of whether the current user in the user information is a certain user in the user information .

The difference value corresponding to each feature vector is judged according to the difference threshold in the face recognition model to obtain the recognition result of whether the current user in the user information is a certain user in the user information. Specifically, the recognition result can be obtained by judging the difference value corresponding to each feature vector through the difference threshold. If the difference value corresponding to a certain feature vector is less than the difference threshold, and the difference value of other feature vectors is not less than the difference threshold, Then the recognition result is that the current user is a certain user in the user information; if the difference value corresponding to all feature vectors is not less than the difference threshold, it indicates that the photo in the face information matches the photo in the user information failed to match , The recognition result is that the current user is not a certain user in the user information; if the difference value corresponding to multiple feature vectors is smaller than the difference threshold, it indicates that the photo in the obtained face information is not a face, The recognition result is also that the current user is not a certain user in the user information.

For example, if the preset difference threshold is 500, one of the difference values ε ₃ corresponding to all feature vectors is 362, and the difference values corresponding to other feature vectors are not less than 500, the recognition result is that the current user is the user The third user in the message.

In one embodiment, as shown in FIG. 4, after step S130, a step S130a is further included.

S130a: If the recognition result is that the current user is not a certain user in the user information, send a prompt message to the current user to obtain face information again and obtain the face information of the current user again.

If the recognition result is that the current user is not a certain user in the user information, a prompt message for obtaining the face information again is issued to the current user and the face information of the current user is obtained again. If the recognition result is that the current user is not a certain user in the user information, a prompt message may be sent to the current user to obtain the face information of the current user again. After obtaining the new face information, the new face information can be recognized again by the above method to obtain the recognition result.

S140. If the recognition result is that the current user is a certain user in the user information, according to the recognition result, obtain target push information that matches the face information from a preset to-be-push information database, to target the The current user pushes.

If the recognition result is that the current user is a certain user in the user information, the target push information matching the facial information is obtained from the preset to-be-push information database according to the recognition result to send information to the current user Push it. Specifically, the information library to be pushed is an information library for storing information to be pushed. The information library to be pushed contains multiple information push categories and user classification rules, and each information push category contains at least one piece of information to be pushed. Information, the information to be pushed in the information database to be pushed can be files such as pictures, videos, audios, etc. According to the recognition results, the target push information that matches the facial information can be obtained, and the target push information is displayed to the current user in the user terminal. Pushing to achieve accurate information push can greatly improve the effect of information push.

In an embodiment, as shown in FIG. 5, step S140 includes sub-steps S141, S142, and S143.

S141. Acquire personal information corresponding to the face information in the user information according to the recognition result.

Acquire personal information corresponding to the face information in the user information according to the recognition result. Specifically, the user information includes multiple users' information. Among them, a user's information includes a photo of the user (for example, a photo of the user's ID card) and personal information of the user. The personal information includes gender, age, Education, marriage, occupation and residence information, etc. According to the recognition result, the personal information corresponding to the face information can be obtained from the user information.

For example, the recognition result is that the current user is the third user in the user information, and the personal information of the third user in the user information includes gender: male, age: 29, education: high school, marriage: unmarried, occupation: White-collar workers can use the above personal information as personal information corresponding to the face information.

S142. Obtain a user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed.

Acquire a user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed. According to the user group classification rules, the corresponding users can be classified into the corresponding user groups. Each information push category corresponds to a user group, and each information push category contains at least one piece of information to be pushed, that is, through information Push category can classify all the information to be pushed contained in the information database to be pushed. Through user classification rules and personal information corresponding to the face information, a user group matching the personal information can be obtained, and each user group corresponds to an information push category, that is, the target user group, specifically, According to user classification rules, users can be classified into the first category according to age, users are classified into the first category of the first category according to gender, and users are classified into the first subcategory of the first category according to marriage and love. According to occupation, users are classified into the first sub-category in the first sub-category, and a user group matching the user in the first sub-category is obtained as the target user group according to the educational background.

For example, user classification rules are shown in Table 1.

Table 1

According to the user classification rules shown in Table 1, to obtain personal information corresponding to the above-mentioned face information (male, age: 29, education: high school, marriage: unmarried, occupation: white-collar), the target user group can be matched with After the user corresponding to the personal information is classified, the target user group matching the user is obtained as the user group d1H6p.

S143. Obtain the to-be-push information contained in the information push category corresponding to the target user group in the to-be-push information database as the target push information to push the current user.

Obtain the to-be-push information contained in the information push category corresponding to the target user group in the to-be-push information database as the target push information to push the current user. Since each user group corresponds to an information push category, after obtaining the target user group that matches the personal information, obtain the information to be pushed contained in the information push category corresponding to the target user group, and you can get the information that matches the personal information Target push information, where the target push information contains at least one piece of information to be pushed, and the target push information is sequentially pushed in the user terminal to the current user to achieve accurate push of the information.

For example, if the information push category corresponding to the user group d1H6p is marriage and love information, the information to be pushed contained in the information push category is used as the target push information that matches the above personal information and pushes the current user.

In the information push method based on face recognition provided by the embodiment of this application, the feature vector set is extracted from the user information input by the administrator according to the vector extraction model, and the current user is automatically acquired when the preset information push time point is reached. According to the recognition result, obtain the target push information from the preset to-be-push information database and push the current user. Through the above method, it is possible to obtain target push information that matches the user information based on the face recognition technology, so as to achieve accurate push of information, which greatly improves the push effect of information.

The embodiment of the present application also provides an information pushing device based on face recognition, and the information pushing device based on face recognition is used to execute any embodiment of the aforementioned information pushing method based on face recognition. Specifically, please refer to FIG. 6, which is a schematic block diagram of an information pushing device based on face recognition provided by an embodiment of the present application. The information pushing device based on face recognition can be configured in a user terminal.

As shown in FIG. 6, the information pushing device 100 based on face recognition includes a feature vector extraction unit 110, a face information acquisition unit 120, a face information recognition unit 130 and a target push information pushing unit 140.

The feature vector extraction unit 110 is configured to, if user information input by the administrator is received, extract feature vectors from each photo of the user information according to a preset vector extraction model to obtain a feature vector set.

In other embodiments of the invention, as shown in FIG. 7, the feature vector extraction unit 110 includes sub-units: a decolorization unit 111, a vector conversion unit 112 and a feature vector calculation unit 113.

The decolorization unit 111 is configured to decolorize each photo in the user information to obtain a photo collection.

The vector conversion unit 112 is configured to convert the photos in the photo set according to the vector conversion rules in the vector extraction model to obtain a vector set.

The feature vector calculation unit 113 is configured to calculate the feature vector corresponding to each photo according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.

The face information acquiring unit 120 is configured to acquire the face information of the current user if the preset information push time point is reached.

The face information recognition unit 130 is configured to recognize the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information.

In other embodiments of the invention, as shown in FIG. 8, the face information recognition unit 130 includes sub-units: a face feature vector acquisition unit 131, a vector difference value calculation unit 132 and a difference value judgment unit 133.

The face feature vector acquiring unit 131 is configured to acquire the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model.

The vector difference value calculation unit 132 is configured to calculate the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model.

The difference value judgment unit 133 is configured to judge the difference value corresponding to each feature vector according to the difference threshold value in the face recognition model to obtain whether the current user in the user information is in the user information The recognition result of a certain user.

In other embodiments of the invention, as shown in FIG. 9, the information pushing device 100 based on face recognition further includes a subunit: a face information reacquiring unit 130a.

The face information reacquiring unit 130a is configured to, if the recognition result is that the current user is not a certain user in the user information, send a prompt message to the current user to obtain face information again and obtain the current user’s information again. Face information.

The target push information playback unit 140 is configured to, if the recognition result is that the current user is a certain user in the user information, obtain the information that matches the face information from a preset to-be-push information database according to the recognition result The target pushes information to push the current user.

In other embodiments of the invention, as shown in FIG. 10, the target push information playing unit 140 includes sub-units: a personal information acquisition unit 141, a user group matching unit 142, and a target push information acquisition unit 143.

The personal information obtaining unit 141 is configured to obtain personal information corresponding to the face information in the user information according to the recognition result.

The user group matching unit 142 is configured to obtain a user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed.

According to the user classification rules of the information database to be pushed, a user group matching the personal information is obtained as a target user group. According to the user group classification rules, the corresponding users can be classified into the corresponding user groups. Each information push category corresponds to a user group, and each information push category contains at least one piece of information to be pushed, that is, through information Push category can classify all the information to be pushed contained in the information database to be pushed. Through user classification rules and personal information corresponding to the face information, a user group matching the personal information can be obtained, and each user group corresponds to an information push category, that is, the target user group, specifically, According to user classification rules, users can be classified into the first category according to age, users are classified into the first category of the first category according to gender, and users are classified into the first subcategory of the first category according to marriage and love. According to occupation, users are classified into the first sub-category in the first sub-category, and a user group matching the user in the first sub-category is obtained as the target user group according to the educational background.

The target push information acquiring unit 143 is configured to acquire the to-be-pushed information contained in the push category of the information corresponding to the target user group in the to-be-push information library as target push information to push the current user.

The aforementioned information pushing device based on face recognition can be implemented in the form of a computer program, and the computer program can be run on a computer device as shown in FIG. 11. Please refer to FIG. 11, which is a schematic block diagram of a computer device according to an embodiment of the present application.

Referring to FIG. 11, the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.

The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute an information push method based on face recognition.

The processor 502 is used to provide calculation and control capabilities, and support the operation of the entire computer device 500.

The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can make the processor 502 execute an information push method based on face recognition.

The network interface 505 is used for network communication, such as providing data information transmission. Those skilled in the art can understand that the structure shown in FIG. 11 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

Wherein, the processor 502 is configured to run a computer program 5032 stored in the memory to realize the following function: if user information input by the administrator is received, a model is extracted from each photo of the user information according to a preset vector Extract feature vectors in the, to obtain a feature vector set, where the user information includes multiple photos, and each photo corresponds to a user; if the preset information push time point is reached, the face information of the current user is obtained; The preset face recognition model and the feature vector set recognize the face information to obtain the recognition result of whether the current user is a certain user in the user information; if the recognition result is that the current user is A certain user in the user information obtains target push information that matches the face information from a preset to-be-push information database according to the recognition result to push the current user.

In one embodiment, when the processor 502 executes the step of extracting feature vectors from each photo of the user information according to a preset vector extraction model to obtain a feature vector set if the user information input by the administrator is received, Perform the following operations: decolorize each photo in the user information to obtain a photo set; convert the photos in the photo set to obtain a vector set according to the vector conversion rule in the vector extraction model; The feature vector calculation formula in the vector extraction model and the vector set are calculated to obtain the feature vector corresponding to each photo to obtain the feature vector set.

In an embodiment, the processor 502 is performing recognition of the face information according to a preset face recognition model and the feature vector set to obtain the recognition of whether the current user is a certain user in the user information In the result step, perform the following operations: obtain the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model; calculate according to the difference calculation formula in the face recognition model The difference value between the face feature vector and each feature vector in the feature vector set; the difference value corresponding to each feature vector is judged according to the difference threshold value in the face recognition model to obtain the difference value in the user information The identification result of whether the current user is a certain user in the user information.

In an embodiment, the processor 502 is performing recognition of the face information according to a preset face recognition model and the feature vector set to obtain the recognition of whether the current user is a certain user in the user information After the result step, the following operations are also performed: if the recognition result is that the current user is not a certain user in the user information, a prompt message is issued to the current user to obtain face information again, and the current user's information is obtained again. Face information.

In an embodiment, the processor 502 executes the step of acquiring target push information matching the face information from a preset to-be-pushing information database according to the recognition result to push the current user The operations are as follows: obtain personal information corresponding to the face information in the user information according to the recognition result; obtain the user group matching the personal information according to the user classification rule of the information database to be pushed As a target user group; acquiring the information to be pushed contained in the information push category corresponding to the target user group in the information database to be pushed as the target push information to push the current user.

Those skilled in the art can understand that the embodiment of the computer device shown in FIG. 11 does not constitute a limitation on the specific configuration of the computer device. In other embodiments, the computer device may include more or less components than those shown in the figure. Or combine certain components, or different component arrangements. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structures and functions of the memory and the processor are the same as those of the embodiment shown in FIG. 11, which will not be repeated here.

It should be understood that, in the embodiment of the present invention, the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Wherein, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.

In another embodiment of the present application, a computer-readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, where the computer program is executed by the processor to implement the following steps: if user information input by the administrator is received, a model is extracted from each photo of the user information according to a preset vector The feature vector is extracted to obtain a feature vector set, wherein the user information includes multiple photos, and each photo corresponds to a user; if the preset information push time point is reached, the face information of the current user is obtained; Set the face recognition model and the feature vector set to recognize the face information to obtain the recognition result of whether the current user is a certain user in the user information; if the recognition result is that the current user is the A certain user in the user information obtains target push information that matches the face information from a preset to-be-push information database according to the recognition result to push the current user.

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Anyone familiar with the technical field can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements shall be covered within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

An information push method based on face recognition includes:

If the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain a feature vector set, wherein the user information includes a plurality of the photos, each One said photo corresponds to one user;

If it reaches the preset information push time point, obtain the current user's face information;

Recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information;

If the recognition result is that the current user is a certain user in the user information, the target push information matching the facial information is obtained from the preset to-be-push information database according to the recognition result to send information to the current user Push it.
The method for pushing information based on face recognition according to claim 1, wherein the vector extraction model includes a vector conversion rule and a feature vector calculation formula, and if the user information input by the administrator is received, the model is extracted according to a preset vector Extracting a feature vector from each photo of the user information to obtain a feature vector set includes:

Decolorize each photo in the user information to obtain a photo collection;

Transform the photos in the photo collection according to the vector conversion rules in the vector extraction model to obtain a vector collection;

The feature vector corresponding to each photo is calculated according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.
The information push method based on face recognition according to claim 1, wherein the face recognition model includes vector extraction rules, difference calculation formula calculations and difference thresholds, and the preset face recognition model and the feature vector set Recognizing the face information to obtain the recognition result of whether the current user is a certain user in the user information includes:

Acquiring the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model;

Calculating the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model;

The difference value corresponding to each feature vector is judged according to the difference threshold in the face recognition model to obtain the recognition result of whether the current user in the user information is a certain user in the user information.
The method for pushing information based on face recognition according to claim 1, wherein the face information is recognized according to a preset face recognition model and the feature vector set to obtain whether the current user is the user After the identification result of a certain user in the information, it also includes:

If the recognition result is that the current user is not a certain user in the user information, a prompt message for obtaining face information again is issued to the current user and the face information of the current user is obtained again.
The method for pushing information based on face recognition according to claim 1, wherein the information library to be pushed includes multiple information pushing categories and user classification rules, and the information is obtained from a preset information library to be pushed according to the recognition result Pushing target information matching the face information to push the current user includes:

Acquiring personal information corresponding to the face information in the user information according to the recognition result;

Acquiring the user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed;

Obtain the to-be-push information contained in the information push category corresponding to the target user group in the to-be-push information database as the target push information to push the current user.
An information push device based on face recognition includes:

The feature vector extraction unit is configured to extract feature vectors from each photo of the user information according to the preset vector extraction model to obtain a feature vector set if the user information input by the administrator is received, wherein the user information includes A plurality of said photos, each said photo corresponds to a user;

The face information acquiring unit is used to acquire the face information of the current user when the preset information push time point is reached;

The face information recognition unit is configured to recognize the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information;

The target push information push unit is configured to, if the recognition result is that the current user is a certain user in the user information, obtain a target matching the face information from a preset to-be-push information database according to the recognition result Push information to push the current user.
The information pushing device based on face recognition according to claim 6, wherein the feature vector extraction unit comprises:

A decolorization unit for decolorizing each photo in the user information to obtain a photo collection;

A vector conversion unit, configured to convert the photos in the photo collection according to the vector conversion rules in the vector extraction model to obtain a vector collection;

The feature vector calculation unit is configured to calculate the feature vector corresponding to each photo according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.
The information pushing device based on face recognition according to claim 6, wherein the face information recognition unit comprises:

A face feature vector obtaining unit, configured to obtain a face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model;

The vector difference value calculation unit is configured to calculate the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model;

The difference value judging unit is configured to judge the difference value corresponding to each feature vector according to the difference threshold in the face recognition model to obtain whether the current user in the user information is a certain one in the user information The recognition result of a user.
The information pushing device based on face recognition according to claim 6, the information pushing device based on face recognition further comprising:

The face information reacquisition unit is configured to, if the recognition result is that the current user is not a certain user in the user information, send a prompt message to the current user to obtain face information again and obtain the person of the current user again Face information.
According to the information pushing device based on face recognition of claim 6, the target pushing information pushing unit comprises:

The personal information obtaining unit is configured to obtain personal information corresponding to the face information in the user information according to the recognition result.

The user group matching unit is configured to obtain a user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed.

The target push information acquiring unit is configured to acquire the to-be-pushed information contained in the push category of the information corresponding to the target user group in the to-be-push information library as target push information to push the current user.
A computer device including:

One or more processors;

Memory

One or more computer programs, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, and the one or more computer programs are configured to execute An information push method based on face recognition, wherein the information push method based on face recognition includes the following steps:

If the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain a feature vector set, wherein the user information includes a plurality of the photos, each One said photo corresponds to one user;

If it reaches the preset information push time point, obtain the current user's face information;

Recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information;

If the recognition result is that the current user is a certain user in the user information, the target push information matching the facial information is obtained from the preset to-be-push information database according to the recognition result to send information to the current user Push it.
The computer device according to claim 11, wherein the vector extraction model includes a vector conversion rule and a feature vector calculation formula, and if the user information input by the administrator is received, the vector extraction model from the user information is The feature vector is extracted from each photo to obtain the feature vector set, including:

Decolorize each photo in the user information to obtain a photo collection;

Transform the photos in the photo collection according to the vector conversion rules in the vector extraction model to obtain a vector collection;

The feature vector corresponding to each photo is calculated according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.
The computer device according to claim 11, wherein the face recognition model includes a vector extraction rule, a difference calculation formula calculation, and a difference threshold, and the face information is compared with the preset face recognition model and the feature vector set. Performing identification to obtain the identification result of whether the current user is a certain user in the user information includes:

Acquiring the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model;

Calculating the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model;

The difference value corresponding to each feature vector is judged according to the difference threshold in the face recognition model to obtain the recognition result of whether the current user in the user information is a certain user in the user information.
The computer device according to claim 11, wherein the face information is recognized according to a preset face recognition model and the feature vector set to obtain whether the current user is a certain user in the user information After identifying the result, it also includes:

If the recognition result is that the current user is not a certain user in the user information, a prompt message for obtaining face information again is issued to the current user and the face information of the current user is obtained again.
The computer device according to claim 11, wherein the information library to be pushed includes multiple information push categories and user classification rules, and the facial information is obtained from a preset information library to be pushed according to the recognition result. The matched target push information to push the current user includes:

Acquiring personal information corresponding to the face information in the user information according to the recognition result;

Acquiring the user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed;

Obtain the to-be-push information contained in the information push category corresponding to the target user group in the to-be-push information database as the target push information to push the current user.
A computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, a method for pushing information based on face recognition is implemented, wherein the The information push method includes the following steps:

If the user information input by the administrator is received, a feature vector is extracted from each photo of the user information according to a preset vector extraction model to obtain a feature vector set, wherein the user information includes a plurality of the photos, each One said photo corresponds to one user;

If it reaches the preset information push time point, obtain the current user's face information;

Recognizing the face information according to a preset face recognition model and the feature vector set to obtain a recognition result of whether the current user is a certain user in the user information;

If the recognition result is that the current user is a certain user in the user information, the target push information matching the facial information is obtained from the preset to-be-push information database according to the recognition result to send information to the current user Push it.
The computer-readable storage medium according to claim 16, wherein the vector extraction model includes a vector conversion rule and a feature vector calculation formula, and if the user information input by the administrator is received, the vector extraction model is selected from the The feature vector is extracted from each photo of the user information to obtain a feature vector set, including:

Decolorize each photo in the user information to obtain a photo collection;

Transform the photos in the photo collection according to the vector conversion rules in the vector extraction model to obtain a vector collection;

The feature vector corresponding to each photo is calculated according to the feature vector calculation formula in the vector extraction model and the vector set to obtain a feature vector set.
The computer-readable storage medium according to claim 16, wherein the face recognition model includes a vector extraction rule, a difference calculation formula calculation, and a difference threshold, and the face recognition model and the feature vector set are used to compare the Recognizing face information to obtain the recognition result of whether the current user is a certain user in the user information includes:

Acquiring the face feature vector corresponding to the face information according to the vector extraction rule in the face recognition model;

Calculating the difference value between the face feature vector and each feature vector in the feature vector set according to the difference calculation formula in the face recognition model;

The difference value corresponding to each feature vector is judged according to the difference threshold in the face recognition model to obtain the recognition result of whether the current user in the user information is a certain user in the user information.
The computer-readable storage medium according to claim 16, wherein the face information is recognized according to a preset face recognition model and the feature vector set to obtain whether the current user is one of the user information After a user’s identification result, it also includes:

If the recognition result is that the current user is not a certain user in the user information, a prompt message for obtaining face information again is issued to the current user and the face information of the current user is obtained again.
The computer-readable storage medium according to claim 16, wherein the to-be-pushed information database contains a plurality of information push categories and user classification rules, and the identification result is obtained from a preset to-be-pushed information database and the The target pushing information that matches the facial information to push the current user includes:

Acquiring personal information corresponding to the face information in the user information according to the recognition result;

Acquiring the user group matching the personal information as a target user group according to the user classification rule of the information database to be pushed;

Obtain the to-be-push information contained in the information push category corresponding to the target user group in the to-be-push information database as the target push information to push the current user.