CN111626074A - Face classification method and device - Google Patents

Abstract

The invention discloses a face classification method and device. The method includes: extracting a target feature corresponding to the face picture from a face picture; face features, obtain the similarity between the target feature and each face group, wherein one face group corresponds to save the corresponding face features of the same person; according to the similarity corresponding to each face group, determine the target The target face group to which the feature belongs; save the target feature to the target face group, and return the identification information corresponding to the target face group. The invention can realize automatic classification of face features without collecting face samples and registering face feature marks, and returns the classified identification marks, which saves time, improves the efficiency of face feature classification, and is beneficial to ordinary users or users. Home use.

Description

A kind of face classification method and device

technical field

The present invention relates to the technical field of image recognition and classification, and in particular, to a face classification method and device.

Background technique

Currently, smart devices including cameras are becoming more and more popular, such as smart cameras, smart doorbells, smartphones, robots, etc. Generally, these smart devices are equipped with the function of face recognition. Their general workflow is to shoot an image or a video, and then process and analyze the video to identify the face in it.

The face recognition algorithm generally needs to manually mark several face pictures of each person, and then when a new picture appears, the algorithm will compare it with the marked face data, and then determine the person corresponding to the picture. Since this classification method requires a large number of face sample sets, in actual application scenarios, it takes a lot of time to collect and mark a large number of face samples to achieve registration, and finally to achieve face classification and return after classification. This classification method is generally suitable for large institutions such as public security and banks, and is not conducive to the use of ordinary users or families.

SUMMARY OF THE INVENTION

In view of the above problems, a face classification method and device proposed by the present invention can realize automatic classification of face features without collecting face samples and registering face feature marks, and returning the classified identification marks, saving energy. Time, improve the efficiency of face feature classification, and is beneficial to ordinary users or families.

In the first aspect, the present application provides the following technical solutions through an embodiment:

A face classification method, the method includes:

Extract the target feature corresponding to the face picture from the face picture; obtain the similarity between the target feature and each face group according to the target feature and the stored face feature in each face group , wherein, a face group corresponds to save the corresponding face features of the same person; according to the similarity corresponding to each face group, determine the target face group to which the target feature belongs; save the target feature to the target face group, and return the identification information corresponding to the target face group.

Preferably, obtaining the similarity between the target feature and each face group according to the target feature and the stored face features in each face group, including:

Calculate the similarity between the target feature and each face feature in each stored face group; obtain the average value between each similarity corresponding to each face feature in the same face group, and use the The average value is used as the similarity between the target feature and the current face group.

Preferably, determining the target face group to which the target feature belongs according to the similarity corresponding to each face group, including:

If the maximum value of the similarity is greater than the preset judgment threshold, the face group corresponding to the maximum similarity is used as the target face group.

Preferably, determining the target face group to which the target feature belongs according to the similarity corresponding to each face group, including:

If the maximum value of the similarity is less than the preset judgment threshold, a new face group is created; the newly created face group is used as the target face group.

Preferably, after taking the newly created face group as the target face group, it includes:

The identification information is allocated to the target face group.

Preferably, the storing the target feature to the target face group includes:

Judging whether the number of face features stored in the target face group exceeds the preset storage threshold; if the number of face features stored in the target face group exceeds the preset storage threshold, then delete the earliest storage time If the number of face features stored in the target face group does not exceed the preset storage threshold, the target features are saved to the target face group. the target face group.

In the second aspect, based on the same inventive concept, the present application provides the following technical solutions through an embodiment:

A face classification device, the device includes:

The feature extraction module is used to extract the target feature corresponding to the face picture in the face picture; the similarity calculation module is used to obtain the target feature according to the target feature and the stored face features in each face group. Describe the similarity between the target feature and each face group, wherein one face group corresponds to save the corresponding face features of the same person; the face group determination module is used to determine the similarity according to the corresponding similarity of each face group. The target face group to which the target feature belongs; a feature saving module is configured to save the target feature to the target face group, and return the identification information corresponding to the target face group.

Preferably, the similarity calculation module is also used for:

Calculate the similarity between the target feature and each face feature in each stored face group; obtain the average value between each similarity corresponding to each face feature in the same face group, and use the The average value is used as the similarity between the target feature and the current face group.

Preferably, the face group determination module is also used for:

If the maximum value of the similarity is greater than the preset judgment threshold, the face group corresponding to the maximum similarity is used as the target face group.

Preferably, the face group determination module is also used for:

If the maximum value in the similarity is less than the preset judgment threshold, create a new face group;

The newly created face group is used as the target face group.

Preferably, it also includes: an identification distribution module for:

After the newly created face group is used as the target face group, the identification information is allocated to the target face group.

Preferably, the feature preservation module is also used for:

Judging whether the number of face features stored in the target face group exceeds the preset storage threshold; if the number of face features stored in the target face group exceeds the preset storage threshold, then delete the earliest storage time If the number of face features stored in the target face group does not exceed the preset storage threshold, the target features are saved to the target face group. the target face group.

In the third aspect, based on the same inventive concept, the present application provides the following technical solutions through an embodiment:

A face classification apparatus, comprising a processor and a memory, the memory being coupled to the processor, the memory storing instructions, when the instructions are executed by the processor, the face classification apparatus is caused to perform the above-mentioned The steps of any of the methods of the first aspect.

In the fourth aspect, based on the same inventive concept, the present application provides the following technical solutions through an embodiment:

A computer-readable storage medium having a computer program stored thereon, when the program is executed by a processor, implements the steps of the method in any one of the above-mentioned first aspects.

An embodiment of the present invention provides a face classification method and device, wherein the method extracts the target feature corresponding to the face picture from the face picture; according to the target feature and the stored people in each face group face features, obtain the similarity between the target feature and each face group, wherein one face group corresponds to save the corresponding face features of the same person; according to the similarity corresponding to each face group, determine the target The target face group to which the feature belongs; save the target feature to the target face group, and return the identification information corresponding to the target face group. The invention does not need to collect and mark and register face samples in advance in the whole classification process, and directly determines the target face group through the calculation result of the similarity, so as to realize the automatic classification of the target features, and return and save it to the target face group After the corresponding identification information, the time for face sample collection and mark registration is saved, the efficiency of face feature classification is improved, and it is beneficial to ordinary users or families.

The above description is only an overview of the technical solutions of the present invention, in order to be able to understand the technical means of the present invention more clearly, it can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand , the following specific embodiments of the present invention are given.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are for the purpose of illustrating preferred embodiments only and are not to be considered limiting of the invention. Also, the same components are denoted by the same reference numerals throughout the drawings. In the attached image:

FIG. 1 shows a flowchart of a method for classifying human faces according to a first embodiment of the present invention;

Fig. 2 shows the sub-step flow chart of step S20 in Fig. 1;

3 shows a functional block diagram of a face classification method provided by a second embodiment of the present invention;

FIG. 4 shows a block diagram of a module of a face classification apparatus provided by a third embodiment of the present invention.

Detailed ways

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be more thoroughly understood, and will fully convey the scope of the present disclosure to those skilled in the art.

first embodiment

Referring to FIG. 1, FIG. 1 shows a method flowchart of a face classification method provided by the first embodiment of the present invention. The method specifically includes the following steps:

Step S10: Extract the target feature corresponding to the face picture from the face picture.

Step S20: Obtain the similarity between the target feature and each face group according to the target feature and the stored face feature in each face group, wherein one face group corresponds to save the correspondence of the same person. facial features.

Step S30: Determine the target face group to which the target feature belongs according to the similarity corresponding to each face group.

Step S40: Save the target feature to the target face group, and return the identification information corresponding to the target face group.

Before step S10, a step of obtaining a face picture may also be included, and specifically, the face picture may be obtained by taking pictures, video recording and decoding with a camera installed on the smart device.

In step S10, the face picture is a picture or video (video frame including a human face) captured by the smart device. Smart devices such as smart doorbells, smart phones, etc. The target feature corresponding to the face picture represents the feature that can identify and represent the face in the face picture, for example, through the existing LBP (Local Binary Patterns) face feature extraction method, eigenface method (Eigenface) and face The facial features are extracted by methods such as feature point position detection, and the representation form of the facial features can be a vector group representing facial texture, eigenface, and facial feature points, etc., without limitation.

It should be noted that the above feature extraction methods are all existing methods, and the specific implementation process thereof will not be repeated.

In step S20, the face group is used to store face features, and the face features stored in the same face group are all face features of the same person. In a face group, a storage threshold can be set to limit the storage upper limit of the number of corresponding face features, which can save storage resources and similarity computing resources.

For example, face group A is used to store the face features of user A, and the upper limit for saving is 10 face features, then the face features stored in face group A are less than or equal to 10.

In addition, the storage occupation size of the face group may also be limited, that is, the storage threshold may be the quantity or the content size. For example, the memory footprint of each face group does not exceed 1024K.

Specifically, please refer to FIG. 2, step S20 includes:

Step S21: Calculate the similarity between the target feature and each face feature in each stored face group.

Step S22: Obtain the average value between each similarity corresponding to each face feature in the same face group, and use the average value as the similarity between the target feature and the current face group.

For example: in the smart doorbell, or in the server of the smart doorbell, there are two face groups, face group A and face group B, and the face feature storage upper limit of the two face groups is 10. There are 10 facial features, and 7 facial features are stored in the face group B. Calculate the similarity of face groups. For face group A, calculate the similarity between the target feature and the 10 face features in face group A, and get 10 similarities, and then compare the 10 similarities corresponding to face group A. The average value obtained can be used as the similarity between the target feature and face group A; similarly, for face group B, calculate the difference between the target feature and the seven face features in face group B. Similarity, 7 similarities can be obtained, and then the 7 similarities corresponding to face group B are averaged, and the obtained average can be used as the similarity between the target feature and face group B.

The calculation of similarity can be carried out in the local smart device or in the cloud server. For example, for a smart device with weak data processing capability (weak processor performance), the data can be uploaded to the cloud for similarity calculation, such as: smart doorbell, smart bracelet, etc. For smart devices with strong data processing capabilities (strong processor performance), similarity calculation can be performed in local smart devices, such as smart phones, smart robots, and the like.

In addition, the specific measurement method of the similarity is not limited, for example, the similarity between the target feature and the face feature is measured by cosine distance, Euclidean distance, Hamming distance, etc.

In step S30, specifically, after obtaining the similarities of each face group corresponding to the target feature, the face group corresponding to the maximum similarity may be used as the target face group for storing the target feature.

However, in order to further improve the accuracy of target feature preservation, it is avoided that two face features that are not the same person but are relatively similar are erroneously stored in the same face group. A judgment threshold can be set, and when the maximum similarity is greater than or equal to the judgment threshold, the target feature can be stored in the face group. Otherwise, it is judged that all current face groups cannot be used to save the target features, and a new face group needs to be created as the target face group to save the target features.

In order to facilitate the determination of the person corresponding to each face group, corresponding identification information may be added when each face group is newly created to distinguish different face groups. For example, the corresponding number of each face group can be distinguished, or the user can customize the settings for the newly created face group. For example, after creating a new face group as the target face group, the person corresponding to the target feature is Zhang San, Then, the target face group can be marked with "Zhang San" or its corresponding ID number as identification information.

In step S40, when the target feature is stored in the target face group, it should also be judged whether the face feature stored in the target face group exceeds a preset storage threshold, wherein the storage thresholds of different face groups may be the same or Different, for example, in a smart doorbell, a larger storage upper limit can be set for the face group that saves the owner's facial features to improve the accuracy of similarity calculation and ensure the accuracy of the smart doorbell owner identification.

Further, if the face features stored in the target face group exceed the preset storage threshold, delete the face features with the earliest storage time and then save the target features to the target face group; If the face feature does not exceed the preset storage threshold, the target feature is directly saved to the target face group. For example, if the similarity between face group A and the target feature is greater than the similarity between face group B and the target feature, and is greater than the judgment threshold, the face feature with the earliest storage time in face group A needs to be deleted at least one, and then store the target feature. If the similarity between face group B and the target feature is greater than the similarity between face group A and the target feature, and is greater than the judgment threshold, since there are 7 face features stored in face group B, the storage upper limit has not been reached. , at this time, the target feature can be stored in the face group B without deleting the face feature.

In step S40, the returned identification information may be the name, user-defined number, ID number, etc. of the person corresponding to the face group, which is not limited.

To sum up, the embodiment of the present invention provides a face classification method, by extracting the target feature corresponding to the face picture from the face picture; according to the target feature and the stored data in each face group face features, obtain the similarity between the target feature and each face group, wherein one face group corresponds to save the corresponding face features of the same person; according to the similarity corresponding to each face group, determine A target face group for saving the target feature; saving the target feature to the target face group, and returning the identification information corresponding to the target face group. In the entire classification process performed by the present invention, it is not necessary to collect and register face samples in advance, and directly determine a target face group through the calculation result of the similarity, so that the automatic classification of target features can be realized, that is, no manual operation is required. Marking can realize the automatic grouping of faces; and only need to input a face image to return its identification information (such as ID), without the need to input a lot of faces to return the identification information of each face, saving face The time for sample collection and mark registration improves the efficiency of face feature classification, which is beneficial to ordinary users or families, and can be promoted on a large scale.

Second Embodiment

Based on the same inventive concept, the second embodiment of the present invention provides a face classification apparatus 400 . FIG. 3 shows a block diagram of functional modules of a face classification apparatus 400 provided by the second embodiment of the present invention.

Specifically, the device 400 includes:

The feature extraction module 401 is used for extracting target features corresponding to the face picture in the face picture.

The similarity calculation module 402 is used to obtain the similarity between the target feature and each face group according to the target feature and the stored face feature in each face group, wherein one face group corresponds to The corresponding facial features of the same person are saved.

The face group determination module 403 is configured to determine the target face group to which the target feature belongs according to the similarity corresponding to each face group.

The feature saving module 404 is configured to save the target feature to the target face group, and return the identification information corresponding to the target face group.

As an optional implementation manner, the similarity calculation module 402 is further configured to:

Calculate the similarity between the target feature and each face feature in each stored face group; obtain the average value between each similarity corresponding to each face feature in the same face group, and use the The average value is used as the similarity between the target feature and the current face group.

As an optional implementation manner, the face group determination module 403 is further configured to:

If the maximum value of the similarity is greater than the preset judgment threshold, the face group corresponding to the maximum similarity is used as the target face group.

As an optional implementation manner, the face group determination module 403 is further configured to:

If the maximum value of the similarity is less than the preset judgment threshold, a new face group is created; the newly created face group is used as the target face group.

As an optional implementation manner, it also includes: an identification allocation module for:

After the newly created face group is used as the target face group, the identification information is allocated to the target face group.

As an optional implementation manner, the feature saving module 404 is further configured to:

Judging whether the number of face features stored in the target face group exceeds the preset storage threshold; if the number of face features stored in the target face group exceeds the preset storage threshold, then delete the earliest storage time If the number of face features stored in the target face group does not exceed the preset storage threshold, the target features are saved to the target face group. the target face group.

It should be noted that the specific implementation and technical effects of the device 400 provided by the embodiments of the present invention are the same as those of the foregoing method embodiments. For brief description, for the parts not mentioned in the device embodiments, reference may be made to the foregoing method embodiments. corresponding content.

Third Embodiment

In addition, based on the same inventive concept, a third embodiment of the present invention further provides a face classification apparatus, including a processor and a memory, the memory is coupled to the processor, and the memory stores an instruction, when the instruction When executed by the processor, the face classification apparatus is caused to perform the following operations:

Extract the target feature corresponding to the face picture from the face picture; obtain the similarity between the target feature and each face group according to the target feature and the stored face feature in each face group , wherein, a face group corresponds to save the corresponding face features of the same person; according to the similarity corresponding to each face group, determine the target face group to which the target feature belongs; save the target feature to the target face group, and return the identification information corresponding to the target face group.

It should be noted that, in the face classification apparatus provided in the embodiment of the present invention, the specific implementation of each step and the technical effect produced are the same as those in the foregoing method embodiment. For brief description, the parts not mentioned in this embodiment may be Refer to the corresponding content in the foregoing method embodiments.

In the embodiment of the present invention, an operating system and a third-party application program are installed in the face classification apparatus. The face classification device may be a tablet computer, a mobile phone, a smart doorbell, a cleaning robot, a notebook computer, a PC (personal computer, personal computer), a wearable device, a vehicle terminal and other terminal devices.

FIG. 4 shows a block diagram of an exemplary face classification apparatus 500 . As shown in FIG. 4 , the face classification apparatus 500 includes a memory 502, a storage controller 504, one or more (only one is shown in the figure) processor 506, a peripheral interface 508, a network module 510, an input and output module 512, Display module 514, etc. These components communicate with each other via one or more communication bus/signal lines 516 .

The memory 502 can be used to store software programs and modules, such as the program instructions/modules corresponding to the face classification method and the device in the embodiment of the present invention. The processor 506 executes various software programs and modules by running the software programs and modules stored in the memory 502. Function application and data processing, such as the face classification method provided by the embodiment of the present invention.

Memory 502 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. Access to memory 502 by processor 506 and possibly other components may be under the control of memory controller 504 .

Peripherals interface 508 couples various input/output devices to processor 506 and memory 502 . In some embodiments, peripheral interface 508, processor 506, and memory controller 504 may be implemented in a single chip. In other instances, they may be implemented by separate chips.

The network module 510 is used for receiving and sending network signals. The above-mentioned network signals may include wireless signals or wired signals.

The input and output module 512 is used for providing input data to the user to realize the interaction between the user and the face classification apparatus. The input and output module 512 may be, but not limited to, a mouse, a keyboard, a touch screen, and the like.

The display module 514 provides an interactive interface (eg, a user operation interface) between the face classification apparatus 500 and the user or is used to display image data for the user's reference. In this embodiment, the display module 514 may be a liquid crystal display or a touch display. In the case of a touch display, it can be a capacitive touch screen or a resistive touch screen that supports single-point and multi-touch operations. Supporting single-point and multi-touch operation means that the touch display can sense the touch operation from one or more positions on the touch display at the same time, and hand over the sensed touch operation to the processor. calculation and processing.

It can be understood that the structure shown in FIG. 4 is only for illustration, and the face classification apparatus 500 may further include more or less components than those shown in FIG. 4 , or have different configurations from those shown in FIG. 4 . Each component shown in FIG. 4 can be implemented in hardware, software, or a combination thereof.

Fourth Embodiment

The fourth embodiment of the present invention provides a computer storage medium. If the functional modules integrated in the face classification device in the second embodiment of the present invention are implemented in the form of software functional modules and sold or used as independent products, they can store in a computer-readable storage medium. Based on this understanding, the present invention realizes all or part of the process in the above-mentioned first embodiment of the face classification method, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer readable. In the storage medium, when the computer program is executed by the processor, the steps of the foregoing method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Electric carrier signals and telecommunication signals are not included.

The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with teaching based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not directed to any particular programming language. It is to be understood that various programming languages may be used to implement the inventions described herein, and that the descriptions of specific languages above are intended to disclose the best mode for carrying out the invention.

In the description provided herein, numerous specific details are set forth. It will be understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it is to be understood that in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or its description. This disclosure, however, should not be construed as reflecting an intention that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. The modules or units or components in the embodiments may be combined into one module or unit or component, and further they may be divided into multiple sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method so disclosed may be employed in any combination, unless at least some of such features and/or procedures or elements are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, it will be understood by those skilled in the art that although some of the embodiments herein include certain features, but not others, included in other embodiments, that combinations of features of the different embodiments are intended to be within the scope of the present invention And form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

Various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the apparatus for face classification according to the embodiment of the present invention. The present invention can also be implemented as apparatus or apparatus programs (eg, computer programs and computer program products) for performing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such signals may be downloaded from Internet sites, or provided on carrier signals, or in any other form.

It should be noted that the above-described embodiments illustrate rather than limit the invention, and that alternative embodiments may be devised by those skilled in the art without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. do not denote any order. These words can be interpreted as names.

The present invention discloses A1. a face classification method, characterized in that the method comprises:

extracting the target feature corresponding to the face image from the face image;

According to the target feature and the stored face features in each face group, the similarity between the target feature and each face group is obtained, wherein one face group corresponds to save the corresponding face of the same person feature;

Determine the target face group to which the target feature belongs according to the similarity corresponding to each face group;

The target feature is saved to the target face group, and the identification information corresponding to the target face group is returned.

A2. The method according to A1, characterized in that, according to the target feature and the face feature in each stored face group, the similarity between the target feature and each face group is obtained, include:

Calculate the similarity between the target feature and each face feature in each stored face group;

Obtain the average value between each similarity corresponding to each face feature in the same face group, and use the average value as the similarity between the target feature and the current face group.

A3. The method according to A1, characterized in that, determining the target face group to which the target feature belongs according to the similarity corresponding to each face group, including:

If the maximum value of the similarity is greater than the preset judgment threshold, the face group corresponding to the maximum similarity is used as the target face group.

A4. The method according to A1, characterized in that, determining the target face group to which the target feature belongs according to the similarity corresponding to each face group, comprising:

If the maximum value in the similarity is less than the preset judgment threshold, create a new face group;

The newly created face group is used as the target face group.

A5. The method according to A4, characterized in that, after taking the newly created face group as the target face group, the method comprises:

The identification information is allocated to the target face group.

A6. The method according to A1, wherein the storing the target feature to the target face group comprises:

Judging whether the number of face features stored in the target face group exceeds a preset storage threshold;

If the number of face features stored in the target face group exceeds a preset storage threshold, delete the face feature with the earliest storage time, and save the target feature to the target face group;

If the number of face features stored in the target face group does not exceed a preset storage threshold, the target feature is stored in the target face group.

The present invention discloses B7. a face classification device, characterized in that the device comprises:

a feature extraction module, used for extracting the target feature corresponding to the face image in the face image;

The similarity calculation module is used to obtain the similarity between the target feature and each face group according to the target feature and the stored face feature in each face group, wherein one face group corresponds to save The corresponding facial features of the same person;

a face group determination module, configured to determine the target face group to which the target feature belongs according to the similarity corresponding to each face group;

A feature saving module, configured to save the target feature to the target face group, and return identification information corresponding to the target face group.

B8. The device according to B7, wherein the similarity calculation module is also used for:

Calculate the similarity between the target feature and each face feature in each stored face group;

Obtain the average value between each similarity corresponding to each face feature in the same face group, and use the average value as the similarity between the target feature and the current face group.

B9. The device according to B7, wherein the face group determination module is also used for:

If the maximum value of the similarity is greater than the preset judgment threshold, the face group corresponding to the maximum similarity is used as the target face group.

B10. The device according to B7, wherein the face group determination module is further used for:

If the maximum value in the similarity is less than the preset judgment threshold, create a new face group;

The newly created face group is used as the target face group.

B11. The device according to B10, further comprising: an identification distribution module for:

After the newly created face group is used as the target face group, the identification information is allocated to the target face group.

B12. The device according to B7, characterized in that the feature preservation module is also used for:

Judging whether the number of face features stored in the target face group exceeds a preset storage threshold;

If the number of face features stored in the target face group exceeds a preset storage threshold, delete the face feature with the earliest storage time, and save the target feature to the target face group;

If the number of face features stored in the target face group does not exceed a preset storage threshold, the target feature is stored in the target face group.

The present invention discloses C13. A face classification device, characterized in that it comprises a processor and a memory, the memory is coupled to the processor, and the memory stores an instruction, when the instruction is executed by the processor At the same time, the face classification apparatus is made to execute the steps of any one of the methods in A1-A6.

The present invention discloses D14. A computer-readable storage medium on which a computer program is stored, characterized in that, when the program is executed by a processor, the steps of any one of the methods in A1-A6 are implemented.

Claims (10)

1. A face classification method, characterized in that the method comprises:

extracting target features corresponding to the face picture from the face picture;

according to the target features and stored face features in each face group, obtaining the similarity between the target features and each face group, wherein one face group correspondingly stores the corresponding face features of the same person;

determining a target face group to which the target features belong according to the corresponding similarity of each face group;

and storing the target features to the target face group, and returning identification information corresponding to the target face group.

2. The method of claim 1, wherein obtaining the similarity between the target feature and each face group according to the target feature and the stored face features in each face group comprises:

calculating the similarity between the target feature and each face feature in each stored face group;

and acquiring an average value between each similarity corresponding to each face feature in the same face group, and taking the average value as the similarity between the target feature and the current face group.

3. The method according to claim 1, wherein the determining the target face group to which the target feature belongs according to the similarity corresponding to each face group comprises:

and if the maximum value in the similarity is larger than a preset judgment threshold value, taking the face group corresponding to the maximum similarity as a target face group.

4. The method according to claim 1, wherein the determining the target face group to which the target feature belongs according to the similarity corresponding to each face group comprises:

if the maximum value in the similarity is smaller than a preset judgment threshold value, a face group is newly established;

and taking the newly-built face group as the target face group.

5. The method according to claim 4, wherein the step of setting the newly created face group as the target face group comprises:

and distributing the identification information to the target face group.

6. The method of claim 1, wherein saving the target features to the target face group comprises:

judging whether the number of the face features stored in the target face group exceeds a preset storage threshold value or not;

if the number of the face features stored in the target face group exceeds a preset storage threshold value, deleting the face feature with the earliest storage time, and storing the target feature into the target face group;

and if the number of the face features stored in the target face group does not exceed a preset storage threshold, storing the target features into the target face group.

7. An apparatus for classifying a human face, the apparatus comprising:

the characteristic extraction module is used for extracting target characteristics corresponding to the face image from the face image;

the similarity calculation module is used for obtaining the similarity between the target feature and each face group according to the target feature and the stored face features in each face group, wherein one face group correspondingly stores the corresponding face features of the same person;

the face group determining module is used for determining a target face group to which the target features belong according to the corresponding similarity of each face group;

and the characteristic storage module is used for storing the target characteristics to the target face group and returning the identification information corresponding to the target face group.

8. The apparatus of claim 7, wherein the similarity calculation module is further configured to:

calculating the similarity between the target feature and each face feature in each stored face group;

and acquiring an average value between each similarity corresponding to each face feature in the same face group, and taking the average value as the similarity between the target feature and the current face group.

9. A face classification apparatus comprising a processor and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the face classification apparatus to perform the steps of the method of any of claims 1-6.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.

Images