A method and system for processing a face image

Abstract

The application relates to a face image processing method and system, and belongs to the technical field of computers, wherein the method comprises the following steps: acquiring a face image of a property owner uploaded by a terminal; measuring a face length, a face width and a position of an intersection of a straight line where the face length and the face width are located of the face image of the property owner; calculating face proportion data of the property owner based on the face length, the face width and the position of the intersection and storing the face proportion data of the property owner into a preset face database of the property owner; before playing a target recording video, calculating face proportion data of all to-be-measured face images based on face lengths, face widths and positions of intersections of all to-be-measured face images in the target recording video; and performing fuzzy processing on to-be-measured face images in the target recording video, which satisfy the face proportion data of the face database of the property owner. The application has the effect of reducing the operation cost of a property.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to a face image processing method and system. BACKGROUND

[0002] At present, more and more people begin to pay attention to their privacy safety, and the cameras in the community often leak the portraits and activity tracks of the owners.

[0003] In the related art, some high-end apartment entrances are equipped with recording cameras, and owners can upload their own face images to the server of the property. When it is necessary to screen outsiders, the recording video in the recording camera needs to be called, and the server compares all the stored face images with the face in the recording video one by one. The server pre-fuzzes the face images of all the owners appearing in the recording video. On the one hand, the demand for screening outsiders is met, and on the other hand, the privacy safety of the owners' portraits is ensured.

[0004] The related art in the above has the following defects: the server needs to store a large number of face images of the owners, and the face images occupy a large storage space, resulting in a high operating cost of the property. SUMMARY

[0005] In order to improve the problem that the face image occupies a large storage space and results in a high operating cost of the property, the present application provides a face image processing method and system.

[0006] In a first aspect, the present application provides a face image processing method, which adopts the following technical solution:

[0007] A face image processing method, the method comprising:

[0008] obtaining a face image of an owner uploaded by a terminal;

[0009] measuring the face length, face width and intersection position of a straight line on which the face length and face width are located of the face image of the owner;

[0010] based on the face length, face width and intersection position, calculating owner face ratio data and storing the owner face ratio data into a preset owner face database;

[0011] before playing a target recording video, based on the face length, face width and intersection position of all the face images to be measured in the target recording video, calculating the face ratio data to be measured of all the face images to be measured;

[0012] fuzzing the face images to be measured in the target recording video, the face ratio data to be measured of which meets the owner face database.

[0013] By adopting the technical scheme, the server needs to pre-process the owner face image uploaded by the terminal and generate corresponding owner face proportion data, so that when the server intercepts the to-be-tested face image in the recorded video, it can determine whether the face image is the owner according to the owner face proportion data. If the server identifies that the to-be-tested face image is the owner, the face image is blurred. Further, it can effectively reduce the storage of a large number of face images by the server, and reduce the operation cost of the property under the condition of satisfying the blurring of the owner face.

[0014] Optionally, the storing of the owner face proportion data into the preset owner face database comprises:

[0015] Obtaining a family account of the terminal, the family account at least comprising a family identifier;

[0016] Adding the family identifier into the owner face proportion data;

[0017] Storing the owner face proportion data into the preset owner face database.

[0018] By adopting the technical scheme, each household has a corresponding family account, and the householder can upload the face images of other members in the family to the server. The server obtains the family account and the corresponding face images, calculates the face images to obtain the owner face proportion data, and adds the family identifier of the family account to the owner face proportion data. It is convenient to call or change the face proportion data in units of families.

[0019] Optionally, the blurring of the to-be-tested face image in the target recorded video, which satisfies the to-be-tested face proportion data of the owner face database, comprises:

[0020] When the to-be-tested face proportion data of the first to-be-tested face image is consistent with the owner face proportion data of the first owner face image, obtaining a first family identifier corresponding to the first owner face image;

[0021] Obtaining the relative face proportion data corresponding to the first family identifier;

[0022] Blurring the to-be-tested face image in the target recorded video, which satisfies the relative face proportion data.

[0023] By adopting the technical scheme, since the family members of the same household have a high probability of entering and exiting the community together, when the server identifies that the face proportion data of one of the to-be-tested face images is consistent with the face proportion data of one of the property owners, the household identifier corresponding to the face proportion data of the property owner is acquired, each relative face proportion data belonging to the household identifier is selected, and each relative face proportion data is compared with the to-be-tested face proportion data of the to-be-tested face image in the recorded video. The efficiency of the server in screening the to-be-tested face image can be effectively improved, so as to shorten the time that the property or the property owner needs to wait for the recorded video to be processed.

[0024] Optionally, the method further comprises:

[0025] In each recorded video, if the spatial distance of the plurality of property owner face images mapped is less than the preset close distance threshold, a suspected close relationship is established between the plurality of property owner face images.

[0026] In each recorded video, if the number of times that the plurality of property owner face images of the suspected close relationship appear simultaneously exceeds the preset neighborhood relationship close threshold, a neighborhood relationship close identifier is added to the face proportion data corresponding to the plurality of property owner face images.

[0027] By adopting the technical scheme, the server extracts the property owner face images less than the close distance threshold in the plurality of recorded videos and counts the number of times that the plurality of property owners appear simultaneously, and when the number of times that the plurality of property owners appear simultaneously exceeds the preset neighborhood relationship close threshold, the neighborhood relationship close identifier is added to the face proportion data of the plurality of property owners, which helps the server to preferentially investigate other property owners in close relationship with the property owner when the property owner face image is identified, and further shortens the time of the server in screening the to-be-tested face image.

[0028] Optionally, the blurring processing of the to-be-tested face image in the target recorded video, whose to-be-tested face proportion data satisfies the to-be-tested face image of the property owner face database, comprises:

[0029] When the to-be-tested face proportion data of the second to-be-tested face image is identified as being consistent with the property owner face proportion data of the second property owner face image, a second neighborhood relationship close identifier corresponding to the second property owner face image is acquired.

[0030] The close face proportion data corresponding to the second neighborhood relationship close identifier is acquired.

[0031] The to-be-tested face image in the target recorded video, whose to-be-tested face proportion data satisfies the close face proportion data, is subjected to blurring processing.

[0032] By adopting the technical scheme, when the server identifies a face image of an owner in a recorded video, the corresponding close neighbor relationship identifier of the owner is obtained, and the face proportion data of other owners carrying the close neighbor relationship identifier is compared with the to-be-detected face proportion data in the recorded video, so that the efficiency of screening the to-be-detected face in the recorded video by the server is improved.

[0033] Optionally, after the face proportion data is calculated based on the face length, the face width and the intersection position, the method further includes:

[0034] When it is identified that the same owner face proportion data exists, facial features of the face image of the owner are extracted;

[0035] The position of the facial features in the face image of the owner is determined, and corresponding facial feature data is generated;

[0036] The facial feature data is added to the owner face proportion data.

[0037] By adopting the technical scheme, when the server obtains that the to-be-recorded owner face proportion data is the same as the owner face proportion data stored in the owner face database, the facial features of the to-be-recorded owner face image are extracted, and then the server generates corresponding facial feature data and adds the facial feature data to the to-be-recorded owner face image data, so that the situation that the same owner face proportion data stored between owners is difficult to distinguish is effectively reduced, and the possibility that the property is difficult to investigate the corresponding owner through the owner face proportion data is reduced.

[0038] Optionally, the blurring processing of the target recorded video, the to-be-detected face proportion data satisfying the to-be-detected face image of the owner face database includes:

[0039] When the corresponding facial feature data does not exist in the target owner face data in the face proportion database, if the target to-be-detected face data in the target recorded video satisfies the target to-be-detected face image of the owner face database, the target owner face proportion data corresponding to the target to-be-detected face image is obtained;

[0040] The pre-stored target terminal identifier corresponding to the target owner face proportion data is selected, and the target terminal identifier carries the unit building information of the owner;

[0041] The target to-be-detected face image is subjected to blurring processing, and a label of a high-similarity person and the unit building information of the owner are added to the target to-be-detected face image.

[0042] By adopting the above technical solution, when the server identifies a face that matches the proportions of a resident's face but does not store corresponding facial feature data, it first blurs the face image. Then, it adds the resident's building information (unit information) to the face image and adds a high-similarity tag for manual screening by property management. This reduces the chances of a face matching the resident's proportions but not being recognized by the server, improving the accuracy of the server's screening of outsiders.

[0043] Secondly, this application provides a face image processing apparatus, which adopts the following technical solution:

[0044] An apparatus for facial image processing, the apparatus comprising:

[0045] The acquisition module is used to acquire the facial images of the homeowners uploaded by the terminal;

[0046] The measurement module is used to measure the face length, face width, and the intersection point of the straight lines containing the face length and face width in the owner's face image.

[0047] The first calculation module is used to calculate the owner's face ratio data based on the face length, face width and intersection position, and store the owner's face ratio data in a preset owner face database.

[0048] The second calculation module is used to calculate the proportion data of the face to be tested in all face images to be tested before playing the target recorded video, based on the face length, face width and intersection position of all face images to be tested in the target recorded video.

[0049] The processing module is used to blur the face image of the target recorded video that meets the proportion data of the owner's face database.

[0050] By adopting the above technical solution, the server processes the owner's facial image to generate corresponding owner facial proportion data. Then, the server calculates and generates the test facial proportion data in the recorded video of the face image to be tested, which can be compared with the owner's facial proportion data. In this way, while meeting the requirements for identifying owners, the server reduces the need to store a large number of facial images, improves storage utilization, and correspondingly reduces the property's operating costs.

[0051] Thirdly, this application provides a system for processing facial images, employing the following technical solution:

[0052] Optionally, the face image processing system includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the face image processing method as described in the first aspect.

[0053] By adopting the above technical solution, a face image processing system can implement the above face image processing method according to the relevant computer program stored in the memory, thereby improving the cooperation between information from different sources when comparing face images, and thus improving the utilization of server storage space.

[0054] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0055] Optionally, the storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement a face image processing method as described in the first aspect.

[0056] By adopting the above technical solution, the corresponding program can be stored, thereby improving the collaboration between information from different sources during face image comparison and thus improving the utilization rate of server storage space.

[0057] In summary, this application includes at least one of the following beneficial technical effects:

[0058] By comparing facial proportion data, the server can blur facial images that match the facial proportion data of the owner, thereby effectively reducing the amount of facial images stored on the server. While satisfying the requirement of blurring the owner's face, this can reduce the property's operating costs.

[0059] By setting up a family account, the homeowner can upload facial images of other family members to the server. The server obtains the family account and the corresponding facial images, calculates the facial image to obtain the homeowner's facial proportion data, and adds the family identifier of the family account to the homeowner's facial proportion data, making it easy to retrieve or change the facial proportion data of the family as a unit.

[0060] By setting family identifiers, the server compares the facial proportion data of each relative with the facial proportion data of the person to be tested in the recorded video. This can effectively improve the server's efficiency in screening the facial images to be tested, thereby shortening the time that property management or owners need to wait for the recorded video to be processed. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a framework structure diagram of a face image processing system according to an embodiment of this application.

[0063] Figure 2 This is a flowchart illustrating a face image processing method according to an embodiment of this application.

[0064] Figure 3 This is a schematic diagram of the server process in an embodiment of this application. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0066] This invention provides a method for face image processing, which can be applied to a face image processing system. The framework structure of the face image processing system can be as follows: Figure 1 As shown, it can include a server and multiple terminals. Specifically, the execution entity of the method can be the server, assisted by the terminals. The server is used to acquire the resident's face image sent by the terminal and blur the resident's face image appearing in the recorded video stored on the server. Simultaneously, the server is also used to store the recorded video of the community entrance and exit captured by the cameras. Specifically, the server acquires the resident's face image sent by the terminal and calculates the resident's face proportion data corresponding to the resident's face image. When it is necessary to screen outsiders at the community entrance and exit, the server blurs the face image of the target person that meets the resident's face proportion data in the recorded video of the cameras at the community entrance and exit.

[0067] The following will describe the specific implementation methods. Figure 2 The processing flow shown is explained in detail below:

[0068] Step 201: The server obtains the homeowner's face image uploaded by the terminal.

[0069] In this embodiment, the terminal can be a mobile phone. The homeowner can take an electronic frontal photo (i.e., a photo of their face) using their phone and then send the photo to the property management server. The property management server can be simply referred to as the server. The server can then obtain the homeowner's facial image uploaded by the terminal.

[0070] Step 202: The server measures the length and width of the owner's face image, as well as the location of the intersection of the lines containing the length and width of the face.

[0071] In this embodiment, after receiving the homeowner's face image, the server measures the length of the widest line segment in the image and then the length of the longest line segment. This allows the server to obtain the face length and width. The server then obtains the intersection point of the line containing the widest and longest points of the face. This intersection point can be simplified to a point location later.

[0072] Step 203: Based on the face length, face width, and intersection point, the server calculates the owner's face proportion data and stores the owner's face proportion data in the preset owner face database.

[0073] In this embodiment, after obtaining the face length, face width, and intersection point, the server calculates the face aspect ratio based on the face length and width. Simultaneously, based on the distances from the intersection point to the two endpoints of the straight line containing the longest part of the face, the server calculates the length ratio of the upper half to the lower half of the face. For example, the widest part of the face is typically the cheekbone; therefore, the ratio of the distance from the cheekbone to the top of the head and the distance from the cheekbone to the jawline reflects the length ratio of the upper half to the lower half of the face. The server compiles and summarizes the face aspect ratio and cheekbone height ratio to generate the homeowner's face proportion data. The server then stores the homeowner's face proportion data in a preset homeowner face database.

[0074] Optionally, if the value of the homeowner's facial proportion data meets the preset range for children's facial proportions, the server adds a periodic update flag to the homeowner's facial proportion data. The periodic update flag is used to send information to the terminal that the homeowner's facial proportion data needs to be updated when a preset period is reached.

[0075] In this embodiment, considering that children's facial features change rapidly with age, the server needs to update children's faces regularly. Since children's faces are rounder than adult faces, and the cheekbone height of children's faces is greater than that of adult faces, the server presets a range of child face proportions to enable it to distinguish between adult and child face images.

[0076] Based on this mechanism, after the server calculates the facial proportion data of the homeowners, and when the value of the homeowner's facial proportion data obtained by the server meets the range of a child's facial proportion, the server adds a periodic update flag to the homeowner's facial proportion data. The periodic update flag is used to send information to the terminal indicating that the homeowner's facial proportion data needs updating when a preset period is reached. The preset period can be one year or two years.

[0077] Step 204: Before playing the target recorded video, the server calculates the proportion data of the face to be tested in all the face images to be tested based on the face length, face width and intersection position of all the face images to be tested in the target recorded video.

[0078] In this embodiment, the entrances and exits of the residential community are equipped with cameras that can record residents' faces directly. The server periodically acquires the recorded videos from these cameras, either daily or weekly. Before the property management or a resident needs to play a specific recorded video (i.e., the target recorded video), the server filters and extracts all the face images to be tested from the recorded video, and measures the face length, face width, and intersection point of all the face images. The server can then calculate the proportion data of the faces to be tested from all the face images.

[0079] Step 205: The server blurs the face image of the target person in the recorded video, where the proportion data of the face to be tested meets the requirements of the owner's face database.

[0080] In this embodiment, after the server calculates all the face proportion data to be tested, the server blurs the face image in the target recorded video whose face proportion data matches that of the owner. In this embodiment, blurring can be done by adding a mosaic effect.

[0081] Optionally, the server obtains the terminal's family account, which includes at least a family identifier. The server adds the family identifier to the resident's facial proportion data and stores the resident's facial proportion data in a preset resident facial database.

[0082] In this embodiment, after the server calculates the homeowner's facial proportion data, the homeowner's terminal logs into the corresponding family account. The family account includes a family identifier, which is used to distinguish homeowners from different families. The server adds the family identifier to the homeowner's facial proportion data and stores the homeowner's facial proportion data in a preset homeowner facial database.

[0083] Optionally, when the server recognizes that the ratio data of the face to be tested in the first face image to be tested is consistent with the ratio data of the owner's face in the first owner's face image, the server obtains the first family identifier corresponding to the first owner's face image, and at the same time obtains the ratio data of the relatives corresponding to the first family identifier, and blurs the face image to be tested that meets the ratio data of the relatives in the target recorded video.

[0084] In this embodiment, to improve the efficiency of the server in screening the target recorded video of the face image to be tested, after the server calculates the face ratio data of all the face images to be tested, and when the server identifies that the face ratio data of a certain face image to be tested (i.e., the first face image to be tested) matches the face ratio data of a certain homeowner's face image (i.e., the first homeowner's face image), the server obtains the corresponding family identifier (i.e., the first family identifier) ​​stored in the face ratio data of the first homeowner. Since the probability of multiple homeowners of the same household entering and leaving the community at the same time is relatively high, the server searches for the face ratio data of relatives with the first family identifier. The server prioritizes comparing the face ratio data of each relative with the face data of all the face images to be tested. The server can then blur the face images of the face that meet the face ratio data of relatives.

[0085] Optionally, in each recorded video, if the spatial distance between multiple homeowner face images is less than a preset intimate distance threshold, a suspected intimate relationship is established between the multiple homeowner face images. In each recorded video, if the number of times multiple homeowner face images with suspected intimate relationships appear simultaneously exceeds a preset close neighbor relationship threshold, a close neighbor relationship identifier is added to the face ratio data corresponding to the multiple homeowner face images.

[0086] In this embodiment, the server captures image frames from different recorded videos. When multiple face images to be tested exist within an image frame, the server can measure the spatial distance between the mapped face images. The property management can pre-set an intimate distance threshold within the server based on the spatial distance between the face images to be tested; the threshold range can be 60cm to 80cm. The server then calculates the actual spatial distance corresponding to the image frame based on the ratio of the actual size to the electronic photo size. When the proportion data of the face images to be tested meet the requirements of the resident face database, the multiple face images to be tested can correspond to multiple residents of the community. If the actual spatial distance between the faces of multiple residents is less than the pre-set intimate distance threshold, the server establishes a suspected intimate relationship between the multiple resident face images. The server is equipped with storage space to store the number of times the spatial distance between multiple residents falls within the intimate distance threshold. For the value of the above number of times, technicians can pre-set a close neighbor relationship threshold within the server; this threshold can be 5 times or 10 times. In each recorded video, if the number of times multiple homeowners' facial images, suggesting a close relationship, appear simultaneously exceeds the threshold for close neighborly relations, a close neighborly relationship identifier is added to the facial proportion data corresponding to these multiple homeowners' facial images. This identifier is used to associate and store the facial proportion data corresponding to multiple homeowners' facial images. When the server identifies a homeowner's facial image with this identifier, it can prioritize screening other homeowners' facial images with the same identifier to improve the service screening speed. After adding the close neighborly relationship identifier to multiple homeowners' facial images, the server removes the statistics on the number of times the spatial distance between the multiple homeowners falls within the close neighborly distance threshold to reduce the storage space occupied on the server.

[0087] Optionally, when the server identifies that the proportion data of the test face in the second test face image matches the proportion data of the owner's face in the second owner's face image, the server obtains the second close neighbor relationship identifier corresponding to the second owner's face image. The server obtains the close neighbor relationship proportion data corresponding to the second close neighbor relationship identifier and blurs the test face image in the target recorded video where the test face proportion data meets the close neighbor relationship proportion data.

[0088] In this embodiment, after the server calculates the proportion data of all test face images, when the server identifies that the proportion data of a test face image (i.e., the second test face image) matches the proportion data of a homeowner face image (i.e., the second homeowner face image), the server obtains the second close neighbor relationship identifier corresponding to the proportion data of the second homeowner face image. The server obtains the close neighbor relationship proportion data corresponding to the second close neighbor relationship identifier. The server obtains test face images whose spatial distance to the second homeowner face image is less than the close neighbor distance threshold, calculates the proportion data of the test face images, and compares the close neighbor relationship proportion data with the proportion data of the test face images one by one. The server then blurs the test face images that match the comparison.

[0089] Optionally, the server identifies the existence of identical homeowner face proportion data. The server extracts facial features from the homeowner's face image, determines the location of the facial features in the homeowner's face image, and generates corresponding facial feature data. The server then adds the facial feature data to the homeowner's face proportion data.

[0090] In this embodiment, after the server calculates the proportion data of the homeowner's face, it identifies the existence of other homeowners with the same proportion data. The server then extracts facial features from the homeowner's face image; these features can be the homeowner's facial features. Based on pre-defined feature points, the server determines the location of the facial features in the homeowner's face image and generates corresponding facial feature data. For example, the server pre-defines the tip of the nose as a feature point. When the server identifies a homeowner's face image that matches the homeowner's face database, it obtains the location of the tip of the nose in the homeowner's face image and generates corresponding facial feature data. The server then adds the facial feature data to the homeowner's face proportion data.

[0091] Optionally, if the target resident's face data in the face ratio database does not contain corresponding facial feature data, and if the target recorded video contains a target face image that matches the target face image in the resident face database, then the server obtains the target resident's face ratio data corresponding to the target face image. The server selects a pre-stored target terminal identifier corresponding to the target resident's face ratio data, and the target terminal identifier carries the resident's building unit information. The server blurs the target face image and adds tags of highly similar individuals and the resident's building unit information to the target face image.

[0092] In this embodiment, when the server identifies that a certain face proportion data (i.e., the target face proportion data) in the target recorded video matches the resident's face data, and a certain resident face proportion data (i.e., the target resident face proportion data) in the face proportion database does not contain the facial feature data of that resident's face image, the server obtains the target resident face proportion data corresponding to the target face image. The server pre-stores the terminal identifiers (i.e., the target terminal identifiers) corresponding to each resident's terminal, and the target terminal identifiers also carry the resident's building unit information. The server first blurs the target face image, then adds a tag for highly similar individuals to the blank space near the target face image in the target recorded video, and points the arrow of the tag to the target face image. At the same time, the server adds the resident's building unit information to the blank space to facilitate property management personnel to check whether the target face image belongs to a resident.

[0093] Based on the same technical concept, embodiments of this application also disclose a face image processing system, which includes a server and a terminal, such as... Figure 3 As shown, the server includes:

[0094] The acquisition module 301 is used to acquire the owner's face image uploaded by the terminal;

[0095] The measurement module 302 is used to measure the length of the face, the width of the face, and the position of the intersection of the straight lines containing the length and width of the face in the owner's face image.

[0096] The first calculation module 303 is used to calculate the owner's face ratio data based on the face length, face width and intersection position, and store the owner's face ratio data in a preset owner face database.

[0097] The second calculation module 304 is used to calculate the proportion data of the face to be tested in all face images to be tested based on the face length, face width and intersection position of all face images to be tested in the target recorded video before playing the target recorded video.

[0098] The processing module 305 is used to blur the face image of the target recorded video that meets the proportion data of the face in the owner's face database.

[0099] Optionally, the acquisition module 301 is also used to acquire the terminal's family account, which includes at least a family identifier;

[0100] Add a module to add family identifiers to the homeowner's face proportion data;

[0101] The selection module is used to store the proportion data of the homeowner's face into a preset homeowner face database.

[0102] Optionally, when the proportion data of the face to be tested in the first face image to be tested is found to be consistent with the proportion data of the owner's face in the first owner's face image, the acquisition module 301 is further used to acquire the first family identifier corresponding to the first owner's face image, and acquire the proportion data of the relatives' faces corresponding to the first family identifier. The processing module 305 is further used to blur the face image to be tested in the target recorded video that meets the proportion data of the relatives' faces.

[0103] Optionally, in each recorded video, if the spatial distance between multiple homeowner face images is less than a preset intimate distance threshold, the module establishes a suspected intimate relationship between the multiple homeowner face images.

[0104] If the number of times multiple homeowners' facial images that appear to be in close relationships appear simultaneously in each recorded video exceeds the preset threshold for close neighborly relations, the module will add a close neighborly relationship identifier to the facial proportion data corresponding to the multiple homeowners' facial images.

[0105] Optionally, when the proportion data of the face to be tested in the second face image to be tested is found to be consistent with the proportion data of the owner's face in the second owner's face image, the acquisition module 301 is further used to acquire the second close neighbor relationship identifier corresponding to the second owner's face image, and acquire the close neighbor relationship proportion data corresponding to the second close neighbor relationship identifier. The processing module 305 is further used to blur the face image of the target recorded video in which the proportion data of the face to be tested meets the close neighbor relationship proportion data.

[0106] Optionally, when identical homeowner face proportion data is detected, the advance module is used to extract facial features from the homeowner's face image; the determination module is used to determine the location of the facial features in the homeowner's face image and generate corresponding facial feature data; the addition module is also used to add facial feature data to the homeowner's face proportion data.

[0107] Optionally, when the target owner's face data in the face ratio database does not contain corresponding facial feature data, if the target test face data in the target recorded video matches the target test face image in the owner's face database, then the acquisition module 301 is used to acquire the target owner's face ratio data corresponding to the target test face image; the selection module is also used to select a pre-stored target terminal identifier corresponding to the target owner's face ratio data, the target terminal identifier carrying the owner's building unit information; the processing module 305 is also used to blur the target test face image and add a label of highly similar person and the owner's building unit information to the target test face image.

[0108] Optionally, if the value of the homeowner's face proportion data is greater than the preset value for a child's face proportion data, the module adds a periodic update flag to the homeowner's face proportion data. The periodic update flag is used to send information to the terminal indicating that the homeowner's face proportion data needs updating when a preset period is reached.

[0109] This application also discloses a face image processing system including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described above for face image processing.

[0110] This application also discloses a computer-readable storage medium that stores a computer program that can be loaded by a processor and executed as described above for processing a face image. The computer-readable storage medium includes, for example, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0111] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit the scope of protection of the application. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

Claims

1. A method for processing facial images, characterized in that, The method includes: Obtain the facial image of the homeowner uploaded by the terminal; The length and width of the face in the owner's face image, as well as the location of the intersection of the lines containing the length and width of the face, are measured. Based on the face length, face width, and intersection point position, the owner's face ratio data is calculated and stored in a preset owner face database; the owner's face ratio data includes the face length-to-width ratio, i.e., face length / face width, and the upper and lower face length ratio, i.e., the distance from the intersection point to the top of the head / the distance from the intersection point to the lower jaw. The system identifies whether the homeowner's facial proportion data meets the preset range of child facial proportions. If it does, a periodically updated identifier is added to the homeowner's facial proportion data. After calculating the face proportion data based on the face length, face width, and intersection point positions, the method further includes: When identical homeowner face proportions are identified, the facial features of the homeowner's face image are extracted. Determine the location of the facial features in the owner's face image and generate the corresponding facial feature data; Add the facial feature data to the owner's face proportion data; The step of storing the homeowner's facial proportion data into a preset homeowner facial database includes: Obtain the family account of the terminal, wherein the family account includes at least a family identifier; Add the family identifier to the homeowner's face proportion data; The property owner's facial proportion data is stored in a preset property owner facial database; Before playing the target recorded video, the proportion data of the face to be tested in all the face images to be tested are calculated based on the face length, face width and intersection position of all the face images to be tested in the target recorded video. In the target recorded video, the face image of the subject whose proportion data satisfies the owner's face database is blurred; including: When the proportion data of the face to be tested in the first face image to be tested is found to be consistent with the proportion data of the owner's face in the first owner's face image, the first family identifier corresponding to the first owner's face image is obtained; the proportion data of the relatives' faces corresponding to the first family identifier is obtained. The test face image that meets the relative face ratio data in the target recorded video is blurred.

2. The method for processing facial images according to claim 1, characterized in that, The method further includes: In each recorded video, if the spatial distance between multiple homeowner face images is less than a preset intimate distance threshold, a suspected intimate relationship is established between the multiple homeowner face images. If the number of times multiple homeowners' facial images that are suspected of having a close relationship appear simultaneously in each recorded video exceeds a preset threshold for close neighborly relations, then a close neighborly relationship identifier is added to the facial proportion data corresponding to the multiple homeowners' facial images.

3. The method for processing facial images according to claim 2, characterized in that, The step of blurring the face image in the recorded video where the proportions of the face to be tested match those in the owner's face database includes: When the proportion data of the face to be tested in the second face image is identified to be consistent with the proportion data of the owner's face in the second owner's face image, the second close neighbor relationship identifier corresponding to the second owner's face image is obtained. Obtain the proportion of close contact faces corresponding to the second close neighbor relationship identifier; In the target video recording, the face image of the subject whose proportion data meets the closely spaced face proportion data is blurred.

4. The method for processing facial images according to claim 1, characterized in that, The step of blurring the face image in the recorded video where the proportions of the face to be tested match those in the owner's face database includes: When the target owner's face data in the face ratio database does not contain corresponding facial feature data, if there is a target test face data in the target recorded video that matches the target test face image in the owner's face database, then the target owner's face ratio data corresponding to the target test face image is obtained. Select a pre-stored target terminal identifier corresponding to the facial proportion data of the target owner, wherein the target terminal identifier carries information about the building unit where the owner is located; The target face image is blurred, and tags of highly similar individuals and the building information of the owner are added to the target face image.

5. A device for processing facial images, characterized in that, The device includes: The acquisition module is used to acquire the facial images of the homeowners uploaded by the terminal; The measurement module is used to measure the face length, face width, and the intersection point of the straight lines containing the face length and face width in the owner's face image. The first calculation module is used to calculate the owner's face ratio data based on the face length, face width and intersection position, and store the owner's face ratio data in a preset owner face database. The second calculation module is used to calculate the proportion data of the face to be tested in all face images to be tested before playing the target recorded video, based on the face length, face width and intersection position of all face images to be tested in the target recorded video. The processing module is used to blur the face image of the target recorded video that meets the proportion data of the owner's face database.

6. A system for processing human face images, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed according to any one of claims 1 to 4.

Images