A method and device for determining personnel file cover

Abstract

The present application discloses a method and device for determining a personnel file cover, which relates to the field of artificial intelligence technology and is used to provide an accurate personnel file cover to implement personnel archiving and improve the precision and recall rate of the personnel archiving system. The method comprises: obtaining a target image set of a first person, wherein the target image set comprises a plurality of target images, and the target images are images representing the characteristics of the person; clustering the plurality of target images based on the type of the personnel characteristics to obtain a plurality of class centers; wherein one type of personnel characteristics corresponds to at least one class center; and determining the file cover of the first person according to the plurality of class centers.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and device for determining a personnel file cover. Background Art

[0002] In different scenarios, various captured personnel pictures need to be archived for subsequent management of personnel files. Currently, an important way to archive is to compare personnel pictures with the file covers of each person in the personnel file system. According to the similarity between the personnel picture and the file cover, the personnel file to which the personnel picture belongs is determined and archived.

[0003] Due to the influence of the clothing of the people in the captured pictures, the angle of capture, the environment of capture, etc., there may be a large deviation between the picture of the same person and the file cover of the person, which affects the accuracy of the personnel filing system. There may also be problems such as the inability to archive the pictures of the people, resulting in a low recall rate of the personnel filing system. Among them, the more images that cannot be archived, the lower the recall rate. Summary of the invention

[0004] The embodiments of the present application provide a method and device for determining a personnel file cover, which are used to provide an accurate personnel file cover to implement personnel archiving and improve the accuracy and recall rate of the personnel archiving system.

[0005] To achieve the above objectives, the present application provides the following technical solutions:

[0006] In a first aspect, a method for determining a personnel profile cover is provided, comprising: obtaining a target image set of a first person, wherein the target image set includes a plurality of target images, and the target images are images representing personnel characteristics; based on types of personnel characteristics, clustering the plurality of target images to obtain a plurality of class centers; wherein one type of personnel characteristics corresponds to at least one class center; and determining the profile cover of the first person based on the plurality of class centers.

[0007] Since the accuracy of historical archive covers is not high, the first personnel file that is archived based on the historical archive covers may contain images that are archived incorrectly (i.e., images that are not the first personnel). By using this technical solution, the first personnel file cover is determined according to the type of personnel characteristics, the representativeness of the personnel file cover is improved, and the first personnel is archived based on the first personnel file cover, which helps to improve the precision and recall rate of the personnel archiving system.

[0008] In a possible implementation, based on the type of personnel features, multiple target images are clustered to obtain multiple class centers, including: calculating the similarity between any two target images among the multiple target images; determining multiple true value images from the multiple target images according to the density distribution of the similarity between any two target images among the multiple target images; based on the type of personnel features, clustering the multiple true value images to obtain multiple class centers.

[0009] This possible implementation method is to obtain a true image by screening from the target image, and use the true image to determine the class center. If the computer device determines the class center according to this method, it will help improve the accuracy of the person image contained in the personnel file, and help improve the representativeness of the personnel file cover, thereby improving the precision and recall rate of the personnel file system.

[0010] In a possible implementation, the multiple target images include a first target image and a second target image, and multiple true value images are determined from the multiple target images based on the density distribution of the similarity between any two target images in the multiple target images, including: if the similarity between the first target image and the second target image is greater than or equal to a first threshold, it is determined that there is a connection relationship between the first target image and the second target image; if the number of target images in the target image set that have a connection relationship with the first target image is greater than or equal to a second threshold, the first target image is a true value image.

[0011] This possible implementation method provides a specific implementation method for determining the true value image based on the density distribution of the target image. If the computer device determines the true value image according to this method, it will help improve the correlation between the images of the person contained in the personnel file, thereby improving the accuracy of the personnel file cover.

[0012] In a possible implementation, the type of personnel features includes a first type, and the first type is any type of personnel features; based on the type of personnel features, multiple target images are clustered to obtain multiple class centers, including: if the number of target images of the first type meets a first preset condition, then based on the first type, multiple target images are clustered to obtain the class center of the first type.

[0013] This possible implementation method provides a specific implementation method for selecting the type of personnel features. By judging the number of target images contained in the type of personnel features, the type of clustering is determined, which helps to improve the representativeness of the personnel file cover determined by the class center, thereby improving the precision and recall rate of the personnel file system.

[0014] In one possible implementation, obtaining a target image set of a first person includes: obtaining a first image set, the first image set including a plurality of first images, the first image being an image representing characteristics of the person, and a similarity between the first image and a cover of a historical file of the first person being greater than or equal to a third threshold; obtaining a second image based on the first image, and a similarity between the second image and the first image meeting a second preset condition; and determining a target image set of the first person based on the plurality of first images and the second images.

[0015] This possible implementation method provides a method for determining the image of the first person that was omitted from the original file cover, which helps to ensure that the file cover has sufficient reference data when the file cover of the first person is subsequently determined, thereby making the file cover more representative.

[0016] In a possible implementation, multiple target images are clustered based on the type of personnel features to obtain multiple class centers, including: determining a first class center based on the type of personnel features in the target image set; wherein one type of personnel features corresponds to at least one first class center; clustering the target image set according to the first class centers to obtain a clustering result based on the first class centers; re-determining the Nth class center based on the clustering result, and clustering the target image set according to the Nth class center to obtain a clustering result based on the Nth class center; N is an integer greater than 1; wherein one type of personnel features corresponds to at least one Nth class center; when the Nth class center is the same as the N-1th class center, the Nth class center is determined as multiple class centers.

[0017] This possible implementation method provides a specific implementation method for determining the class center. The class center is determined by a clustering algorithm to determine the cover of the personnel file, which helps to improve the representativeness of the personnel file cover, thereby improving the precision and recall rate of the personnel file system.

[0018] In a second aspect, a computer device is provided, including: a functional unit for executing any one of the methods provided in the first aspect, wherein the actions executed by each functional unit are implemented by hardware or by executing corresponding software through hardware. For example, the computer device may include: an acquisition unit, a determination unit, and a processing unit; the acquisition unit is used to acquire a target image set of a first person, wherein the target image set includes a plurality of target images, and the target images are images representing the characteristics of the person; the processing unit is used to cluster the plurality of target images based on the type of the characteristics of the person, and obtain a plurality of class centers; wherein one type of the characteristics of the person corresponds to at least one class center; the determination unit is used to determine the profile cover of the first person based on the plurality of class centers.

[0019] In a third aspect, a computer device is provided, comprising: a processor and a memory. The processor is connected to the memory, the memory is used to store computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory, thereby implementing any one of the methods provided in the first aspect.

[0020] In a fourth aspect, a chip is provided, comprising: a processor and an interface circuit; the interface circuit is used to receive code instructions and transmit them to the processor; the processor is used to run the code instructions to execute any one of the methods provided in the first aspect.

[0021] In a fifth aspect, a computer-readable storage medium is provided, comprising computer execution instructions, which, when executed on a computer, enable the computer to execute any one of the methods provided in the first aspect.

[0022] In a sixth aspect, a computer program product is provided, comprising computer execution instructions, which, when executed on a computer, cause the computer to execute any one of the methods provided in the first aspect.

[0023] The technical effects brought about by any implementation method in the second to sixth aspects can refer to the technical effects brought about by the corresponding implementation method in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of the structure of a computer device provided in an embodiment of the present application;

[0025] Figure 2 A flowchart of a method for determining a personnel file cover provided in an embodiment of the present application;

[0026] Figure 3 A flowchart of a method for determining a personnel file cover provided in an embodiment of the present application;

[0027] Figure 4 A schematic diagram of a method for determining a target image provided in an embodiment of the present application;

[0028] Figure 5 A schematic diagram of the relationship between a true value image and a mixed value image provided in an embodiment of the present application;

[0029] Figure 6 A schematic diagram of a true value image divided into types based on personnel characteristics provided in an embodiment of the present application;

[0030] Figure 7 A schematic diagram of an initialization class center for determining the type of personnel characteristics provided in an embodiment of the present application;

[0031] Figure 8A schematic diagram of a true value image clustering result provided in an embodiment of the present application;

[0032] Fig. 9 A schematic diagram of the composition of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In the description of this application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" means one or more, and "plurality" means two or more. The words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not limit them to be different.

[0034] It should be noted that, in this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0035] The embodiment of the present application provides a method for determining the cover of a personnel file, which can be applied to a computer device. The embodiment of the present application does not impose any restrictions on the specific form of the computer device. For example, the computer device can be a terminal device or a network device. Among them, the terminal device can be referred to as: terminal, user equipment (UE), terminal equipment, access terminal, user unit, user station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication device, user agent or user device, etc. The terminal device can be a mobile phone, augmented reality (AR) device, virtual reality (VR) device, tablet computer, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), etc. The network device can be a server, base station, etc. Among them, the server can be a physical or logical server, or it can be two or more physical or logical servers that share different responsibilities and cooperate with each other to realize the functions of the server.

[0036] In terms of hardware implementation, the above-mentioned computer device can be implemented as follows: Figure 1 The computer device shown is implemented as shown. Figure 1 FIG. 1 is a schematic diagram of the hardware structure of a computer device 10 provided in an embodiment of the present application. The computer device 10 can be used to implement the functions of the above-mentioned computer devices.

[0037] Figure 1 The computer device 10 shown may include: a processor 101 , a memory 102 , a communication interface 103 , and a bus 104 . The processor 101 , the memory 102 , and the communication interface 103 may be connected via the bus 104 .

[0038] The processor 101 is the control center of the computer device 10, and may be a general-purpose central processing unit (CPU) or other general-purpose processors, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0039] As an example, the processor 101 may include one or more CPUs, such as Figure 1 CPU 0 and CPU 1 are shown in .

[0040] The memory 102 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0041] In a possible implementation, the memory 102 may exist independently of the processor 101. The memory 102 may be connected to the processor 101 via a bus 104 and used to store data, instructions, or program codes. When the processor 101 calls and executes the instructions or program codes stored in the memory 102, the method for determining the personnel file cover provided in the embodiment of the present application can be implemented.

[0042] In another possible implementation, the memory 102 may also be integrated with the processor 101 .

[0043] The communication interface 103 is used for connecting the computer device 10 to other devices through a communication network, which may be Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc. The communication interface 103 may include a receiving unit for receiving data and a sending unit for sending data.

[0044] The bus 104 may be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 1 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0045] It should be pointed out that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device 10, except Figure 1 In addition to the components shown, computer device 10 may include more or fewer components than shown, or combine certain components, or arrange components differently.

[0046] In order to make the embodiments of the present application clearer, the concepts and some contents related to the embodiments of the present application are briefly introduced below.

[0047] 1. Personnel file system

[0048] The personnel file system is used to manage personnel files. Personnel files include: personnel name, personnel age, personnel resume, personnel image and other information.

[0049] 2. Personnel file cover

[0050] Personnel profile cover, an image used to represent the personnel profile. Currently, one or more images can be used as the personnel profile cover.

[0051] 3. Personnel Archive

[0052] Personnel archiving refers to the process of grouping images of the same person together based on the personnel file cover. The personnel images or videos captured by the camera can generate model data, which is used by the computer device to identify the corresponding image. The computer device implements personnel archiving based on the model data of the personnel file cover.

[0053] like Figure 2 FIG. 1 is a flowchart of a method for determining a personnel profile cover provided by the present application. The method comprises:

[0054] S201. A computer device obtains a target image set of a first person, wherein the target image set includes a plurality of target images, and the target images are images representing characteristics of the person.

[0055] Personnel characteristics refer to the characteristics of individual persons, including facial features, body torso features and clothing features.

[0056] It is understandable that the target image representing the person's characteristics should have at least one of the above-mentioned person characteristics. The target image set is composed of a plurality of target images having at least one person characteristic.

[0057] Optionally, the computer device obtains the target image set of the first person by means of k-order nearest neighbor, wherein k is an integer greater than or equal to 1. Specifically, when k is 1, the target image set of the first person is the first image set of the first person. The first image set is the first image set after the person is archived based on the cover of the historical archive of the first person. The first image set includes a plurality of first images, and the first images are images that characterize the characteristics of the person. The similarity between the first image and the cover of the historical archive of the first person is greater than or equal to a third threshold.

[0058] Optionally, the computer device obtains the first image set of the first person through the ID of the first person. Exemplarily, in the personnel file system, each person after personnel filing has a personnel ID, such as an ID number, etc. The first image set of the first person is obtained through the ID of the first person.

[0059] When k is greater than 1, the computer device uses the first image set as the first-order neighboring points and obtains the k-order neighboring points as the second image, wherein the similarity between the second image and the first image satisfies the second preset condition. The first image set and the second image form a target image set.

[0060] Taking k as 2, using the second-order nearest neighbor method, and the second preset condition being greater than or equal to the fourth threshold as an example, each first image is taken as a first-order neighboring point, and images in the image set of other persons except the first person in the personnel archive system and / or unarchived images are taken as candidate images. The computer device takes an image that satisfies the similarity between the first image and the candidate image greater than or equal to the fourth threshold as a second image, and the second image may be an image of the first person that is omitted from the archive according to the historical archive cover. Multiple first images and at least one second image constitute a target image set.

[0061] It should be noted that the present application does not limit the threshold setting of the similarity between the above images. The third threshold and the fourth threshold can be the same or different. Exemplarily, the third threshold and the fourth threshold can be set to 90%.

[0062] Optionally, the fourth threshold is less than the third threshold, for example, the third threshold is 90% and the fourth threshold is 85%. It is understandable that the larger the threshold setting, the higher the similarity between the images, and the fewer images that meet the conditions. When the fourth threshold is less than the third threshold, the computer device may obtain a larger number of second images in the personnel file system, thereby expanding the target image set of the first person, which is helpful for the subsequent determination of the file cover of the first person, and the amount of data referenced by the file cover is sufficient, thereby making the file cover more representative.

[0063] S202. The computer device clusters the multiple target images based on the types of personnel features to obtain multiple class centers; wherein one type of personnel feature corresponds to at least one class center.

[0064] When the person feature includes a face feature, the type of the person feature may be classified based on the angle of the face. For example, the angle of the face may include left side face, right side face, front face, looking down and looking up.

[0065] When the person feature includes a human body feature, the type of the person feature may be obtained based on the classification of human body body parts. Exemplarily, the human body body parts include left arm, right arm, left leg, right leg, chest, left hand, right hand, head, etc.

[0066] When the personal feature includes a clothing feature, the type of the personal feature may be classified based on the style or color of the clothing.

[0067] It should be noted that the types of the above-mentioned personnel characteristics are obtained by the computer device according to the model data of the target image. The image is processed to obtain the model data. The model data of the image includes multi-dimensional data, and the multi-dimensional data is used by the computer device to identify the personnel characteristics. Exemplarily, the human skeleton points are identified by combining the image recognition technology with the deep learning technology to obtain multi-dimensional data, and the human skeleton points constitute the above-mentioned human torso parts. The computer device identifies the human torso characteristics according to the multi-dimensional data.

[0068] Optionally, the computer device calculates the similarity between any two target images among the multiple target images; determines multiple true value images from the multiple target images according to the density distribution of the similarity between any two target images among the multiple target images; clusters the multiple true value images based on the type of personnel characteristics to obtain multiple class centers. Due to the low accuracy of the personnel file cover or the low accuracy of the personnel filing method, the first personnel file may contain images with filing errors (i.e., images that are not the first person), that is, the target image set of the first person contains images that are not the first person, resulting in the target image being unable to represent the first person's characteristics. The above method helps to find the true value image that represents the first person's characteristics, determine the first person's file cover according to the true value image, and file the first person according to the first person's file cover, which helps to improve the precision and recall rate of the personnel filing system.

[0069] Specifically, the multiple target images include a first target image and a second target image, and the computer device calculates the similarity between the first target image and the second target image. If the similarity between the first target image and the second target image is greater than or equal to a first threshold, it is determined that there is a connection relationship between the first target image and the second target image; if the number of target images in the target image set that have a connection relationship with the first target image is greater than or equal to a second threshold, the first target image is a true value image. Exemplarily, the target images include image A, image B, and image C. Assuming that the similarity between image A and image B and image C all meet the first threshold, there is a connection relationship between image A and image B and image C, and the number of images that have a connection relationship with image A is 2. If the number of images 2 meets the second threshold, image A is the true value image.

[0070] It can be understood that using the above method to determine whether the target image is a true value image helps to improve the correlation between the images of the person included in the personnel file, exclude target images with low credibility, and thus improve the accuracy of the personnel file cover.

[0071] Optionally, the type of personnel feature includes a first type, and the first type is any type of personnel feature; if the number of target images of the first type meets the first preset condition, then based on the first type, multiple target images are clustered to obtain the class center of the first type. Exemplarily, assuming that the personnel feature includes a facial feature, the type of personnel feature includes a left face and a right face, and the first preset condition is greater than or equal to a fifth threshold. If the number of target images obtained by dividing based on the facial angle corresponding to the left face is greater than or equal to the fifth threshold, clustering is performed based on the left face, and the class center of this type is obtained; if the number of target images obtained by dividing based on the facial angle corresponding to the right face is less than the fifth threshold, clustering of this type is not adopted.

[0072] By judging the number of target images contained in the type of personnel characteristics and determining the type of clustering in the above manner, it is helpful to improve the representativeness of the personnel profile cover determined by the class center.

[0073] Optionally, the computer device determines multiple class centers in the following manner, including: the computer device determines a first class center based on the type of personnel characteristics in the target image set. One type of personnel characteristics corresponds to at least one first class center, and the first class center can be randomly selected from the types of personnel characteristics. The computer device clusters the target image set according to the first class center to obtain a clustering result based on the first class center; re-determines the Nth class center based on the clustering result, and clusters the target image set according to the Nth class center to obtain a clustering result based on the Nth class center; N is an integer greater than 1; one type of personnel characteristics corresponds to at least one Nth class center; when the Nth class center is the same as the N-1th class center, the Nth class center is determined as multiple class centers.

[0074] Exemplarily, when the person feature includes a face feature, the computer device can obtain the class centers of the left face, the right face, the front face, the top view, and the top view respectively, and each type can include one or more class centers. It can be understood that the one or more class centers are one or more target images.

[0075] S203. The computer device determines the file cover of the first person according to the multiple category centers.

[0076] The profile cover of the first person is determined according to the target images corresponding to the multiple class centers determined in the above step S203.

[0077] Optionally, the profile cover of the first person may include one image or multiple images. The target images corresponding to the multiple class centers are used as the profile cover set of the first person, and the subsequent images of the first person are archived through the profile cover set. Alternatively, the target images corresponding to the multiple class centers are fused, and the fused image is used as the profile cover of the first person.

[0078] Through the above steps S201-S203, the first personnel file cover is determined according to the type of personnel characteristics, the representative ability of the personnel file cover is improved, and the first personnel is archived according to the first personnel file cover, which helps to improve the accuracy and recall rate of the personnel archiving system.

[0079] The following is a method for determining a personnel profile cover using a facial feature as a specific example of a personnel feature. Figure 3 As shown, including:

[0080] S301. A computer device obtains a plurality of target images for representing facial features of a target image set of a first person.

[0081] like Figure 4 As shown, each circle represents an image. Among them, the blank circle is the first image of the first person, and the computer device can directly use it as the target image.

[0082] Alternatively, the second-order nearest neighbor method is used, and the computer device performs a global search in the personnel file system based on the first image. Exemplarily, the first image is used as a first-order neighbor point, and the similarity between the first image and the candidate image in the personnel file system is calculated, and the second image whose similarity is greater than or equal to the fourth threshold is used as a second-order neighbor point. Figure 4 The shaded circle shown in FIG. 1 is a block diagram of a computer device that uses the first image and the second image as the plurality of target images, namely Figure 4 All circles shown in .

[0083] Optionally, S302, the computer device determines multiple true value images from multiple target images.

[0084] like Figure 5 As shown in the figure, each point is a target image determined according to the above step S301. If the computer device determines that the similarity between any two target images is greater than or equal to the first threshold, then there is a connection relationship between the two target images. The computer device calculates the number of connection relationships between each target image, and takes the target image with a number greater than or equal to the second threshold as the true value image. A shown in the figure is the area where the true value image is located, and B is the area where the target image with a data amount of connection relationship less than the second threshold is located. The image with a small number of connection relationships can be called a mixed value image.

[0085] It should be noted that the second threshold is the number of connection relationships between the target images. Generally, the second threshold is preset in the computer device. Alternatively, the second threshold can be sorted in ascending order according to the number of connection relationships between each target image, and the number corresponding to the fifth position after sorting is used as the second threshold. Exemplarily, assuming that there are 10 target images, the number of connection relationships between the 10 target images and other images is 5, 6, 8, 9, 3, 4, 5, 7, 7, 2 respectively; the above numbers are sorted to obtain 2, 3, 4, 5, 5, 6, 7, 7, 8, 9; the number corresponding to the tenth position of the sorting is 3, and the target images corresponding to the number greater than or equal to 3 are used as true value images. The second threshold can also be other values, which are not limited in this application.

[0086] Optionally, the number of connection relationships in the target image is used as the density distribution of the target image. The more connection relationships there are, the greater the density of the target image; conversely, the smaller the density of the target image. The computer device determines whether the target image is a true value image by judging the density distribution of the target image.

[0087] S303: The computer device clusters the multiple true value images based on the angle of the face to obtain multiple class centers.

[0088] The specific implementation of step S303 may include the following steps S11-14.

[0089] S11. The computer device classifies the multiple true value images based on the types of personnel features.

[0090] Specifically, Figure 6 As shown, in the above steps S301-S302, the target image includes multi-dimensional data constituting facial features, and the multi-dimensional data includes the angle of the face. The computer device converts the face into Figure 5 The true value image in the area A determined in is divided into Figure 6 As shown, it is divided into five areas: left side face, right side face, looking down, looking up and front face, and the five areas correspond to the types of five facial features one by one.

[0091] It should be noted that the above-mentioned face angle division method is not limited to the above-mentioned five areas, and may also include more or fewer areas.

[0092] Optionally, after dividing the true value image, the computer device selects the type of personnel feature. If the number of true value images in the area corresponding to the overhead view does not meet the first preset condition, the computer device does not use the type of personnel feature for clustering. Specifically, the first preset condition may be greater than or equal to the fifth threshold. The fifth threshold is the minimum value of the number of true value images in the type of personnel feature, and the fifth threshold can be set by oneself, and this application does not limit it. For example, the fifth threshold is 5, Figure 6 Among the five areas shown, if the number of true value images contained in the area corresponding to the overhead view is less than 5, and the number of true value images contained in other areas is greater than or equal to 5, the computer device will not use the overhead view for clustering, but will use the facial angles corresponding to the other four areas for clustering.

[0093] Through the above-mentioned method of selecting the type of personnel features, after the computer device determines the true value image, it further selects the type of personnel features that can represent the first person, and excludes the type of personnel features with a small number of true value images, thereby avoiding the situation where the number of true value images is small and the personnel features represented by the true value images are not convenient for reference in subsequent personnel archiving.

[0094] S12. The computer device selects an initialization class center from each type of personnel characteristics after classification.

[0095] According to the type of the personnel feature selected in the above step S11, a true value image is randomly selected as the initialization class center of the type of the personnel feature. The initialization class center is the above first class center.

[0096] like Figure 7 As shown, the initialization class center is represented by a checkered circle. Figure 6 An initialization class center is randomly selected from each of the five divided areas, a is the initialization class center in the upward view, b is the initialization class center in the front face, c is the initialization class center in the right face, d is the initialization class center in the downward view, and e is the initialization class center in the left face.

[0097] S13. The computer device clusters the true value image according to the initialized class centers.

[0098] like Figure 7 As shown, based on the five initialized class centers, the similarities between other true value images in area A and the five initialized class centers are calculated, and the other true value images select the class centers with the highest similarity for clustering and are classified into the corresponding face angle areas.

[0099] S14. Repeat the above steps S12 and S13 until the areas to which all true value images belong no longer change.

[0100] Specifically, in the process of repeating the above step S12, the Nth class center is determined, where N is an integer greater than 1. Step S13 is executed to cluster the true value images according to the Nth class center until the regions to which all the true value images belong no longer change, that is, the class centers of each region no longer change. It can also be understood that the Nth class center is the same as the N-1th class center.

[0101] It should be noted that one type of personnel feature corresponds to at least one of the above-mentioned initialization type centers or Nth type centers.

[0102] like Figure 8 As shown, the computer device obtains the result of clustering the true value image in the above step S13, and obtains five classes and the class center that no longer changes in each class.

[0103] For example, the class center is represented by a checkered circle. a' is the class center when looking up, b' is the class center when looking straight, c' is the class center when looking right, d' is the class center when looking down, and e' is the class center when looking left.

[0104] Through the above steps S11-S14, the computer device completes the clustering process of the true value image and obtains multiple clustering results.

[0105] S304. The computer device determines the profile cover of the first person according to the multiple category centers.

[0106] This step is the same as the above step S203 and will not be described again.

[0107] Optionally, S305: The computer device deletes the image corresponding to the mixed-value image from the image set of the first person.

[0108] Optionally, the method for determining the personnel profile cover also includes the computer device deleting a mixed-value image from the image set of the first person, wherein the mixed-value image is an image other than the true-value image in a plurality of images. Exemplarily, in step S302, the true-value image is determined, and other images in the target image are deleted, that is, the mixed-value image is deleted. This method helps to reduce the management burden of the computer device. In addition, the mixed-value image can be marked to avoid including the image corresponding to the mixed-value image in the subsequent process of redetermining the cover of the first personnel profile, thereby increasing the computational burden.

[0109] It should be noted that step S305 can be executed at any step after the above step S302, and there is no limitation on this.

[0110] Through the above steps S301-S305, the computer device screens the acquired images and determines the true image that can represent the first person, thereby avoiding the situation in which, in the process of determining the personnel file cover, the image other than the first person is mixed in, which affects the ability of the file cover to represent the first person. The file cover of the first person determined based on the true image is more representative, which is conducive to improving the accuracy and recall rate of subsequent personnel archiving based on the file cover.

[0111] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the method. It is understandable that in order to realize the above functions, the computer device includes at least one of the hardware structure and software modules corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0112] The embodiment of the present application can divide the computer device into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0113] For example, Fig. 9 A possible structural diagram of a computer device (referred to as computer device 90) involved in the above embodiment is shown, and the computer device 90 includes an acquisition unit 901, a determination unit 902, and a processing unit 903. The acquisition unit 901 is used to acquire a target image set of a first person, wherein the target image set includes multiple target images, and the target images are images that represent the characteristics of the person. For example, Figure 2 Step S201 shown, and Figure 3 Step S301 is shown. Processing unit 902 is used to cluster multiple target images based on the type of personnel features to obtain multiple class centers; wherein one type of personnel feature corresponds to at least one class center. For example, Figure 2 Step S202 shown, and Figure 3 Steps S302, S303, and S305 are shown. A determination unit 903 is used to determine the profile cover of the first person according to the multiple category centers. For example, Figure 2 Step S203 shown, and Figure 3 Step S304 is shown.

[0114] Optionally, the processing unit 902 is specifically used to calculate the similarity between any two target images among the multiple target images; determine multiple true value images from the multiple target images according to the density distribution of the similarity between any two target images among the multiple target images; and cluster the multiple true value images based on the type of personnel features to obtain multiple class centers.

[0115] Optionally, the multiple target images include a first target image and a second target image, and the processing unit 902 is specifically used to determine that there is a connection relationship between the first target image and the second target image if the similarity between the first target image and the second target image is greater than or equal to a first threshold; if the number of target images in the target image set that have a connection relationship with the first target image is greater than or equal to a second threshold, then the first target image is a true value image.

[0116] Optionally, the type of personnel features includes a first type, which is any type of personnel features; the processing unit 902 is specifically used to cluster multiple target images based on the first type to obtain the class center of the first type if the number of target images of the first type meets a first preset condition.

[0117] Optionally, the acquisition unit 901 is specifically used to acquire a first image set, the first image set includes multiple first images, the first image is an image representing characteristics of a person, and the similarity between the first image and the cover of the historical file of the first person is greater than or equal to a third threshold; based on the first image, a second image is acquired, and the similarity between the second image and the first image meets a second preset condition; based on the multiple first images and the second images, a target image set of the first person is determined.

[0118] Optionally, the processing unit 902 is specifically used to determine the first-class center based on the type of personnel characteristics in the target image set; wherein one type of personnel characteristics corresponds to at least one first-class center; clustering the target image set according to the first-class centers to obtain a clustering result based on the first-class centers; re-determining the N-th class center based on the clustering result, and clustering the target image set according to the N-th class center to obtain a clustering result based on the N-th class center; N is an integer greater than 1; wherein one type of personnel characteristics corresponds to at least one N-th class center; when the N-th class center is the same as the N-1-th class center, the N-th class center is determined as multiple class centers.

[0119] Optionally, the computer device 90 further includes a storage unit 904. The storage unit 904 is used to store computer-executable instructions, and other units in the computer device can perform corresponding actions according to the computer-executable instructions stored in the storage unit 904.

[0120] For the detailed description of the above optional methods, please refer to the above method embodiments, which will not be repeated here. In addition, the explanation of any of the above computer devices 90 and the description of the beneficial effects can refer to the above corresponding method embodiments, which will not be repeated here.

[0121] As an example, combining Figure 1 , the functions partially or entirely implemented by the acquisition unit 901, the processing unit 902, the determination unit 903 and the storage unit 904 in the computer device 90 can be Figure 1 Processor 101 in the Figure 1 The acquisition unit 901 can also be implemented by Figure 1 The receiving unit in the communication interface 103 is implemented.

[0122] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed on a computer, the computer is enabled to execute any of the methods executed by the computer devices provided above.

[0123] For the explanation of the relevant contents and description of the beneficial effects of any of the above-mentioned computer-readable storage media, reference may be made to the above-mentioned corresponding embodiments, which will not be repeated here.

[0124] The embodiment of the present application also provides a chip. The chip integrates a control circuit and one or more ports for implementing the functions of the above-mentioned computer device 90. Optionally, the functions supported by the chip can be referred to above and will not be repeated here. A person of ordinary skill in the art can understand that all or part of the steps of implementing the above-mentioned embodiment can be completed by instructing the relevant hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium mentioned above can be a read-only memory, a random access memory, etc. The above-mentioned processing unit or processor can be a central processing unit, a general-purpose processor, an application specific integrated circuit (ASIC), a microprocessor (digital signal processor, DSP), a field programmable gate array (field programmable gatearray, FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof.

[0125] The embodiment of the present application also provides a computer program product including instructions, when the instructions are run on a computer, the computer executes any one of the methods in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means to another website site, computer, server, or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more servers that can be integrated with the medium. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., an SSD), etc.

[0126] It should be noted that the above-mentioned devices for storing computer instructions or computer programs provided in the embodiments of the present application, such as but not limited to the above-mentioned memories, computer-readable storage media and communication chips, etc., are all non-transitory.

[0127] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, data center, etc. that contains one or more servers that can be integrated with a medium. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0128] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other changes to the disclosed embodiments by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0129] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining a personnel file cover, characterized in that: include: Acquire a target image set of a first person, wherein the target image set includes a plurality of target images, and the target images are images representing characteristics of the person; Based on the types of the personnel features, clustering the multiple target images to obtain multiple class centers; wherein one type of personnel feature corresponds to at least one class center; The clustering of the plurality of target images based on the type of the personnel features to obtain a plurality of class centers includes: Calculating the similarity between any two target images among the multiple target images; Determine a plurality of true value images from the plurality of target images according to the density distribution of the similarity between any two target images among the plurality of target images; Determine a plurality of true value images from the plurality of target images according to the density distribution of the similarity between any two target images among the plurality of target images, comprising: if the number of target images in the target image set whose similarity with the first target image is greater than or equal to a first threshold is greater than or equal to a second threshold, then the first target image is a true value image; Based on the types of the personnel features, clustering the multiple true value images to obtain multiple class centers; Determine the profile cover of the first person based on the multiple class centers.

2. The method according to claim 1, characterized in that The plurality of target images include a first target image and a second target image, and determining a plurality of true value images from the plurality of target images according to a density distribution of similarities between any two target images among the plurality of target images includes: If the similarity between the first target image and the second target image is greater than or equal to a first threshold, determining that there is a connection relationship between the first target image and the second target image; If the number of target images in the target image set that are connected to the first target image is greater than or equal to a second threshold, the first target image is the true value image.

3. The method according to claim 1, characterized in that The type of the personnel feature includes a first type, which is any one of the types of the personnel feature; clustering the multiple target images based on the type of the personnel feature to obtain multiple class centers includes: If the number of target images of the first type meets a first preset condition, clustering the multiple target images is performed based on the first type to obtain the class center of the first type.

4. The method according to claim 1, characterized in that: The step of acquiring a target image set of a first person includes: Acquire a first image set, the first image set including a plurality of first images, the first images being images representing characteristics of the person, and the similarity between the first images and a cover of a historical file of the first person being greater than or equal to a third threshold; Based on the first image, acquiring a second image, wherein a similarity between the second image and the first image satisfies a second preset condition; Based on the plurality of first images and the second image, a target image set of the first person is determined.

5. The method according to any one of claims 1 to 4, characterized in that: The clustering of the plurality of target images based on the type of the personnel features to obtain a plurality of class centers includes: Determining a first type of center based on the type of the personnel feature in the target image set; wherein one type of personnel feature corresponds to at least one of the first type of centers; Clustering the target image set according to the first class centers to obtain a clustering result based on the first class centers; Re-determine the Nth class center based on the clustering result, and cluster the target image set according to the Nth class center to obtain a clustering result based on the Nth class center; N is an integer greater than 1; wherein one type of personnel feature corresponds to at least one Nth class center; When the Nth class center is the same as the N-1th class center, the Nth class center is determined as the multiple class centers.

6. A computer device, characterized in that: include: An acquisition unit, configured to acquire a target image set of a first person, wherein the target image set includes a plurality of target images, and the target images are images representing characteristics of the person; A processing unit, configured to cluster the plurality of target images based on the type of the personnel feature to obtain a plurality of cluster centers; wherein one type of personnel feature corresponds to at least one cluster center; The processing unit is specifically configured to calculate the similarity between any two target images among the multiple target images; Determine a plurality of true value images from the plurality of target images according to the density distribution of the similarity between any two target images among the plurality of target images; Determine a plurality of true value images from the plurality of target images according to the density distribution of the similarity between any two target images among the plurality of target images, comprising: if the number of target images in the target image set whose similarity with the first target image is greater than or equal to a first threshold is greater than or equal to a second threshold, then the first target image is a true value image; Based on the types of the personnel features, clustering the multiple true value images to obtain multiple class centers; A determination unit is used to determine the profile cover of the first person based on the multiple category centers.

7. The computer device according to claim 6, characterized in that The multiple target images include a first target image and a second target image, and the processing unit is specifically configured to determine that there is a connection relationship between the first target image and the second target image if the similarity between the first target image and the second target image is greater than or equal to a first threshold; If the number of target images in the target image set that are connected to the first target image is greater than or equal to a second threshold, the first target image is the true value image; The type of the personnel feature includes a first type, which is any type of the type of the personnel feature; the processing unit is specifically configured to cluster the plurality of target images based on the first type to obtain a cluster center of the first type if the number of target images of the first type meets a first preset condition; The acquisition unit is specifically used to acquire a first image set, the first image set includes a plurality of first images, the first images are images representing characteristics of the person, and the similarity between the first images and the cover of the historical file of the first person is greater than or equal to a third threshold; Based on the first image, acquiring a second image, wherein a similarity between the second image and the first image satisfies a second preset condition; Determining a target image set of the first person based on the plurality of first images and the second image; The processing unit is specifically configured to determine a first type of center based on the type of the personnel feature in the target image set; wherein one type of personnel feature corresponds to at least one first type of center; Clustering the target image set according to the first class centers to obtain a clustering result based on the first class centers; Re-determine the Nth class center based on the clustering result, and cluster the target image set according to the Nth class center to obtain a clustering result based on the Nth class center; N is an integer greater than 1; wherein one type of personnel feature corresponds to at least one Nth class center; When the Nth class center is the same as the N-1th class center, the Nth class center is determined as the multiple class centers.

8. A computer device, characterized in that: include: processor; The processor is connected to a memory, the memory is used to store computer-executable instructions, and the processor executes the computer-executable instructions stored in the memory, so that the computer device implements the method according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, when the computer instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 5.

Images