Character recognition method, device and system, electronic equipment and readable storage medium

By using image acquisition equipment to recognize and transmit facial region images, the problem of low accuracy in facial recognition by the central server in complex environments has been solved, achieving higher recognition accuracy and efficiency.

CN119027991APending Publication Date: 2024-11-26SMARTSENS TECH SHENZHEN CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202310554977.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-16
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

When the camera is far from the target, the target is in an uncooperative state of motion, or people are obscuring each other, the person recognition results of the central server are prone to missed detections and false detections, resulting in low accuracy of person recognition.

Method used

After face recognition is performed by the image acquisition device, the face region image is transmitted to the central server for feature extraction and recognition, avoiding the transmission of encoded images. The central server performs feature comparison based on the initial image.

Benefits of technology

It improves the accuracy of person recognition and reduces the bandwidth required for image data transmission, while maintaining high recognition efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027991A_ABST
    Figure CN119027991A_ABST
Patent Text Reader

Abstract

The invention is suitable for the technical field of image recognition, and provides a character recognition method and device, electronic equipment, a system and a readable storage medium. The figure recognition method specifically comprises the steps that image data transmitted by image acquisition equipment is acquired, the image data comprises a face area image obtained by performing face recognition on an initial image through the image acquisition equipment, and the initial image is an image obtained by performing acquisition on a preset area through the image acquisition equipment; according to the image data, extracting face features of each person in the preset area; and comparing the face features of the persons in the preset area with reference face features of persons in a face library to obtain recognition results of the persons in the preset area. According to the embodiment of the invention, the accuracy of character recognition can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of image recognition technology, and in particular, relates to a person recognition method, device, electronic device, system and readable storage medium. Background Art

[0002] Character recognition can be used to conduct intelligent monitoring of key personnel in crowded areas such as major events, major festivals, concerts, and stations. In related technologies, face and object recognition mostly encodes the images captured by the camera and transmits them to the central server, which then extracts and recognizes the features. However, when the camera is far away from the target, the target is in a non-cooperative motion state, or the characters block each other, the character recognition results of the central server are prone to problems such as missed detection and false detection, and the accuracy of character recognition is low. Summary of the invention

[0003] The embodiments of the present application provide a method, device, electronic device, system and storage medium for character recognition, which can solve the problem of low accuracy in character recognition in related technologies.

[0004] A first aspect of an embodiment of the present application provides a person recognition method, which is applied to a central server. The person recognition method includes: acquiring image data transmitted by an image acquisition device, the image data including a face area image obtained by the image acquisition device performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device acquiring a preset area; based on the image data, extracting facial features of each person in the preset area; and comparing the facial features of each person in the preset area with benchmark facial features of persons in a face library to obtain recognition results of each person in the preset area.

[0005] A second aspect of an embodiment of the present application provides a person recognition method, which is applied to an image acquisition device, and the person recognition method includes: acquiring a preset area to obtain an initial image; generating image data based on the initial image, and the image data includes a face area image obtained by the image acquisition device performing face recognition on the initial image; and sending the image data to a central server, wherein the central server is used to process the received image data according to the steps of the person recognition method described in the first aspect.

[0006] A third aspect of an embodiment of the present application provides a person recognition device, which is configured on a central server, and the person recognition device includes: an image acquisition unit, used to acquire image data transmitted by an image acquisition device, the image data including a face area image obtained by the image acquisition device performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device acquiring a preset area; a feature extraction unit, used to extract face features of each person in the preset area based on the image data; and a person recognition unit, used to compare the face features of each person in the preset area with the benchmark face features of people in a face library, to obtain recognition results of each person in the preset area.

[0007] A fourth aspect of an embodiment of the present application provides a person recognition device, which is configured on an image acquisition device, and the person recognition device includes: an image acquisition unit, which is used to acquire a preset area to obtain an initial image; an image processing unit, which is used to generate image data based on the initial image, and the image data includes a face area image obtained by the image acquisition device performing face recognition on the initial image; a sending unit, which is used to send the image data to a central server, wherein the central server is used to process the received image data according to the steps of the person recognition method described in the first aspect.

[0008] A fifth aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned person recognition method when executing the computer program.

[0009] A sixth aspect of an embodiment of the present application provides a person recognition system, comprising: an image acquisition device and a central server connected to the image acquisition device; the image acquisition device is used to send the image data to the central server according to the steps of the person recognition method described in the first aspect; the central server is used to process the received image data according to the steps of the person recognition method described in the second aspect.

[0010] A seventh aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned person recognition method are implemented.

[0011] An eighth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the person recognition method described in the first aspect or the second aspect above.

[0012] In the implementation of the present application, the image acquisition device acquires the preset area to obtain the initial image, and generates image data based on the initial image. The image data includes the face area image obtained by the image acquisition device performing face recognition on the initial image. Since the face area image is obtained based on the initial image recognition, rather than based on the encoded graphic recognition, when the central server extracts the face features of each person in the preset area according to the image data and compares the face features with the reference face features of the people in the face library, the identification result of the person obtained is more accurate. At the same time, since the face area image is part of the initial image, when the image acquisition device and the central server transmit the image data, the bandwidth occupied by the face area image is relatively small, so that the entire person recognition system can maintain a high recognition efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0014] Figure 1 It is a schematic diagram of the architecture of a person recognition system provided in an embodiment of the present application;

[0015] Figure 2 It is a schematic diagram of an implementation flow of a person recognition method applied to a central server provided in an embodiment of the present application;

[0016] Figure 3 is a schematic diagram of a specific implementation process of updating a reference image provided in an embodiment of the present application;

[0017] Figure 4 It is a schematic diagram of a specific implementation flow of step S302 provided in an embodiment of the present application;

[0018] Figure 5 is a schematic diagram of updating a face database provided in an embodiment of the present application;

[0019] Figure 6 is a schematic diagram of a specific architecture of a person recognition system provided in an embodiment of the present application;

[0020] Figure 7 It is a schematic diagram of the specific architecture of the central server provided in the embodiment of the present application;

[0021] Figure 8 It is a schematic diagram of the implementation process of the person recognition method applied to the image acquisition device provided in the embodiment of the present application;

[0022] Fig. 9 is a schematic diagram of a specific architecture of an image acquisition device provided in an embodiment of the present application;

[0023] Fig.10 It is a structural schematic diagram of a person recognition device configured on a central server provided in an embodiment of the present application;

[0024] Fig.11 It is a structural schematic diagram of a person recognition device configured in an image acquisition device provided in an embodiment of the present application;

[0025] Fig.12 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are protected by the present application.

[0027] In related technologies, face and object recognition mostly encodes the images captured by the camera and transmits them to the central server, which then extracts features and performs recognition. However, when the camera is far away from the target, the target is in an uncoordinated motion state, or the characters block each other, the character recognition results of the central server are prone to problems such as missed detection and false detection, and the accuracy of character recognition is low.

[0028] Based on this, the present application proposes a method for character recognition, which can enable an image acquisition device (such as a camera) to transmit a face area image after performing face recognition, so that a central server can perform feature extraction and recognition based on the face area image. Since the face area image is obtained by the image acquisition device through face recognition based on the initial image, character recognition can be completed more accurately compared to the central server using encoded images for recognition.

[0029] In order to illustrate the technical solution of the present application, a specific embodiment is provided below for illustration.

[0030] Figure 1 The schematic diagram of the structure of a person recognition system provided by an embodiment of the present application is shown. The person recognition system 10 may include an image acquisition device 101 and a central server 102 .

[0031] The image acquisition device 101 is a device with an image acquisition function, including but not limited to an optical camera, an industrial camera, a video camera, such as a built-in CMOS (Complementary Metal-Oxide-Semiconductor) image sensor. In the implementation of the present application, the image acquisition device 101 can be set in a specific position to capture images of a preset area to obtain image data. For example, the image acquisition device 101 can be set on a gantry at the entrance of a park to capture images of the entrance of the park and realize the function of character recognition for people entering and leaving the park.

[0032] In an embodiment of the present application, the image acquisition device 101 may be connected to the central server 102 via a wired or wireless manner to send image data to the central server 102 .

[0033] The central server 102 is a device with image processing capability, used to process received image data, perform character recognition, etc. The central server 102 may include but is not limited to a tower server, a rack server, a blade server, etc.

[0034] In some implementations, the person recognition system 10 may further include a monitoring device, which may be connected to the central server 102 by wire or wirelessly to obtain the person recognition result output by the central server 102. The monitoring device may be configured with components such as a display screen and an audio system, which may be used to display the information output by the central server 102 in real time, and to perform voice broadcast when the person recognition result meets the broadcast condition (e.g., a stranger is recognized).

[0035] It is understandable that the above-mentioned person recognition system 10 may also include more or fewer devices. For example, the person recognition system 10 may also include a network terminal for forwarding data between the image acquisition device 101 and the central server 102, or between the central server 102 and the monitoring device, etc., which is not limited in this application.

[0036] For ease of understanding, the character recognition method provided by the present application is described below from the perspective of the central server 102. Figure 2 A schematic diagram of an implementation flow of a person recognition method provided in an embodiment of the present application is shown. The method can be applied to the central server 102 and is suitable for situations where the accuracy of person recognition needs to be improved.

[0037] Specifically, the above-mentioned person recognition method may include the following steps S201 to S203.

[0038] Step S201, acquiring image data transmitted by an image acquisition device.

[0039] The image data may include a face region image obtained by performing face recognition on an initial image by the image acquisition device 101, and the initial image is an image obtained by the image acquisition device 101 acquiring a preset region. The preset region is the image acquisition range of the image acquisition device 101, which is set according to actual needs.

[0040] That is, the image acquisition device 101 can obtain an initial image by performing image acquisition on a preset area. After obtaining the initial image, the image acquisition device 101 can perform face recognition on the initial image to obtain a face area image.

[0041] It should be understood that the face region image is the pixel region where the face is located in the initial image. The image acquisition device 101 can identify the pixel region where the face is located in the initial image, and cut out each identified pixel region from the initial image to obtain one or more face region images.

[0042] It should be noted that the face recognition algorithm used by the image acquisition device 101 can be a recognition algorithm based on facial feature points (feature-based recognition algorithms), a detection / recognition algorithm based on the entire face image (appearance-based detection / recognition algorithms), an algorithm using a neural network for recognition, or other face recognition algorithms, and this application does not impose any restrictions on this.

[0043] In the implementation of the present application, the image acquisition device 101 can acquire the initial image in real time, and send the image data corresponding to each frame of the initial image to the central server 102 in sequence, or send the image data corresponding to multiple consecutive frames of initial images to the central server 102. The present application does not limit this. Based on the method of the present application, the transmission bandwidth requirement can also be improved.

[0044] Step S202: extracting facial features of each person in a preset area based on the image data.

[0045] In the embodiment of the present application, based on the face region images in the image data, the central server 102 can extract the face features in each face region image, and then obtain the face features of each person in the preset region. The face features can represent the characteristics of the corresponding person or the face in the corresponding face region image, and can be used to distinguish different people.

[0046] Among them, the facial features can be extracted by using histograms of oriented gradients (HOG), local binary patterns (LBP), scale-invariant feature transform (SIFT), accelerated robust features (SURF), neural network-based face recognition algorithm or other feature extraction algorithms, which are not limited in this application.

[0047] Step S203, comparing the facial features of each person in the preset area with the reference facial features of the persons in the face library to obtain the recognition results of each person in the preset area.

[0048] In the embodiments of the present application, the face library may be referred to as a face database, which is a collection of face data of people in the library. People in the library are people whose face data is stored in the face library. Specifically, the user may input images of people in the library to the central server 102 in advance. The central server 102 may perform feature extraction on the image input by the user, obtain reference face features, and store the reference face features in the face library. In other embodiments, the image input by the user may be further stored in the face library.

[0049] The reference facial features are features used to compare with facial features of people in the preset area. It should be understood that the method of extracting the reference facial features can be the same as or different from the method of extracting facial features in step S202.

[0050] After the central server 102 extracts the facial features of each person in the preset area, the facial features of each person in the preset area can be compared with the benchmark facial features of the people in the face library to obtain the recognition results of each person in the preset area. Specifically, when there are multiple people in the preset area, the facial features of the first person can be compared with the benchmark facial features of each person in the library to determine the recognition result of the first person, and then the facial features of the second person can be compared with the benchmark facial features of each person in the library to determine the recognition result of the second person, and so on, until the recognition results of each person are obtained.

[0051] Among them, the above recognition result can indicate whether the corresponding person belongs to the person in the database.

[0052] For example, the administrator of the company can input the images of all employees into the central server 102 to extract the reference facial features of each employee (i.e., the personnel in the database) and save them in the face database. After the image acquisition device 101 acquires the initial image of the company entrance, it can output the image data to the central server 102, which compares the facial features of each person in the image with the reference facial features of each employee to determine whether the person entering or leaving the company entrance is a company employee.

[0053] In an embodiment of the present application, the image acquisition device 101 acquires an image of a preset area to obtain an initial image, and generates image data based on the initial image. The image data includes a face area image obtained by the image acquisition device 101 performing face recognition on the initial image. Since the face area image is obtained based on the initial image recognition rather than the encoded graphic recognition, when the central server 102 extracts the facial features of each person in the preset area according to the image data and compares the facial features with the benchmark facial features of the people in the face library, the recognition result of the person obtained is more accurate.

[0054] At the same time, since the face area image is usually part of the initial image, compared with transmitting the initial image, when the image acquisition device 101 and the central server 102 transmit image data, the bandwidth required for the face area image is smaller, so that the entire character recognition system 10 can maintain a high recognition efficiency.

[0055] In some embodiments of the present application, in order to ensure the accuracy of the recognition result, the face library can be iteratively updated. Specifically, the face library stores a reference image of each person in the library, and the reference face features of each person in the library are the face features obtained by face recognition based on the corresponding reference image.

[0056] like Figure 3 As shown, after obtaining the recognition results of each person in the preset area, the central server 102 can also perform the following steps S301 to S302.

[0057] Step S301, if the recognition result of the target person is that the target person matches a person in the library, the face area image of the target person is stored in the pre-stored library of the target person.

[0058] The target person is any one or more persons in the preset area.

[0059] In the implementation of the present application, if the recognition result of the target person is that the target person matches the person in the library, it means that the target person belongs to the person in the library, and the face area image used in the recognition result is an image that can meet the person recognition requirements. At this time, the face area image of the target person can be temporarily stored in the pre-stored library of the target person. The pre-stored library is a data set of face area images. The central server 102 can create a pre-stored library for each person, and then the face area image of each person can be temporarily stored in the corresponding pre-stored library.

[0060] Step S302: When the face database meets the update condition, the reference image of the target person in the face database is updated based on the pre-stored database of the target person.

[0061] Among them, the face library meeting the update condition may refer to the face library reaching the update time, the face library receiving the update instruction issued by the user, etc. Taking the face library reaching the update time as an example, the user can configure the update cycle for the face library (e.g., once every seven days). When the current time is away from the last update time and the update cycle is required, it can be confirmed that the face library has reached the update time and the face library meets the update condition. Taking the face library receiving the update instruction issued by the user as an example, when the user needs to add or delete characters in the library or update the baseline facial features of a character in a certain library, the user can input the update instruction to the central server 102. After the central server 102 receives the update instruction, it can be confirmed that the face library meets the update condition. This application does not impose any restrictions on this.

[0062] In the implementation of the present application, when the face library meets the update condition, the reference image of the target person in the face library can be updated based on the pre-stored library of the target person. Specifically, the central server 102 can use one or more face area images stored in the pre-stored library as the reference image of the target person. At the same time, one or more reference images can be deleted from the face library.

[0063] Accordingly, after the update, the central server 102 may re-determine the benchmark facial features of the target person based on the updated image of the target person in the face library.

[0064] The updated image includes the face region image used as the reference image in the pre-stored library and the face region image retained in the face library.

[0065] It is understandable that if there is only one updated image, the central server 102 can extract facial features from the updated image to obtain new reference facial features. If there are multiple updated images, the central server 102 can extract facial features from each updated image, and then perform mean processing on the facial features extracted from each updated image to obtain new reference facial features.

[0066] In an embodiment of the present application, by updating the reference image of the target person in the face library, the reference image features can be made close to the current face condition of the target person, so that the recognition result will be more accurate when the person is recognized next time.

[0067] In order to further improve the accuracy of next person recognition, Figure 4 As shown, the above step S302 may specifically include the following steps S401 to S403.

[0068] Step S401, when the face database meets the update condition, the selection probability of each face region image in the pre-stored database is calculated.

[0069] In the implementation of the present application, each time a target person is identified, the central server 102 can store a new face region image in the pre-stored library. When the face library meets the update conditions, multiple face region images can be stored in the pre-stored library. In order to filter out the face region images that can better represent the people in the library, the central server 102 can calculate the selection probability of each face region image.

[0070] The selection probability refers to the probability of being selected as the reference image, which is related to the shooting time of the face area image and / or the image confidence of the face area image. The image confidence is also the image quality, which is related to at least one of the brightness, image clarity and face orientation of the face area image.

[0071] Step S402: Filter out at least one target image from a pre-stored library of target persons according to the selection probability.

[0072] The target image is a face region image used to update the face library.

[0073] In the implementation of the present application, based on the selection probability, N face region images with the highest selection probability can be screened out from the pre-stored library of the target person, and each image can be used as a target image, where N is a positive integer greater than or equal to 1. Alternatively, all face region images with a selection probability higher than a probability threshold can be used as a target image. This application does not impose any restrictions on this.

[0074] For example, please refer to Figure 5 The central server 102 can sort the face area images in the pre-stored library in the order of shooting time from near to far and image confidence from high to low. The closer the shooting time and the higher the image confidence, the greater the selection probability. According to the selection probability, at least one target image can be selected.

[0075] When there are multiple target images, the central server 102 may also calculate the feature center of the facial features corresponding to all target images (for example, using the mean of the facial features corresponding to all target images), and calculate the cosine distance between the facial features corresponding to each target image and the feature center. Then, the target images whose cosine distance is greater than the distance threshold are eliminated.

[0076] Among them, the feature center can be the mean of all facial features. When the cosine distance between the facial features corresponding to a certain target image and the feature center is greater than or equal to the distance threshold, it means that the facial features in the target image are close to the real facial features of the target person. Correspondingly, when the cosine distance between the facial features corresponding to a certain target image and the feature center is less than the distance threshold, it means that the facial features in the target image deviate from the real facial features of the target person. At this time, the target image can be eliminated to avoid affecting the accuracy of the next person recognition after updating the benchmark image based on the target image.

[0077] Step S403: updating the reference image of the target person in the face database according to the target image.

[0078] Specifically, the target image may be used as a newly added reference image in the face library, or the target image may be used to replace the original reference image in the face library, thereby updating the reference image of the target person in the face library.

[0079] In an implementation of the present application, based on the selection probability, at least one target image is screened out from a pre-stored library of the target person to update the baseline image. The baseline image can be made into a face area image with a more recent shooting time and a higher image confidence level. The baseline face features are extracted based on the updated baseline image, which helps to improve the accuracy of the next person recognition.

[0080] In step S202 , in order to ensure the accuracy of facial feature extraction, the central server 102 may detect the quality score of the image quality of the facial region image after acquiring the image data transmitted by the image acquisition device 101 .

[0081] The quality score is related to at least one of the brightness of the face area image, the image clarity, and the face orientation in the face area image. It should be understood that the closer the brightness is to the preset optimal brightness, the closer the image clarity is to the optimal clarity, and the closer the face orientation in the face area image is to the front face, the higher the quality score.

[0082] Correspondingly, if the quality score is higher than the score threshold, it means that the face region image can extract the face features well at this time, and the central server 102 can extract the face features of each person in the preset region according to the image data. If the quality score is lower than the score threshold, it means that the face region image cannot extract the face features well at this time, and the central server 102 can re-acquire the image data, or improve the image quality of the face region image to extract the face features of each person in the preset region.

[0083] After obtaining the facial features, in step S203, the central server 102 may compare the facial features of each person in the preset area with the benchmark facial features of the persons in the face library to obtain the recognition results of each person in the preset area.

[0084] In some embodiments, when the recognition result of the target person is that the target person does not match the person in the library, it means that the target person may be a "stranger" other than the person in the library. In order to ensure the accuracy of the recognition result, the central server 102 can perform image quality improvement processing on the face area image to obtain an image with improved image quality, and then re-determine the recognition result of the target person based on the image with improved image quality; or, the central server 102 can perform quality score screening on the face area image to screen out a target quality image, and determine whether it is a stranger through the recognition result corresponding to the screened target quality image; wherein the screened target quality image can be an image with a higher image quality score, for example, a quality score greater than the score threshold used in step S202.

[0085] Specifically, the above-mentioned image quality improvement can be achieved through interpolation algorithms, brightness and color adjustment, super-resolution enhancement, or other existing image quality improvement algorithms. Based on the image after image quality improvement, the central server 102 can re-extract facial features, and then use the re-extracted facial features to compare with the reference facial features to re-determine the recognition result of the target person.

[0086] It should be understood that if the recognition result is re-determined to still not match the person in the library after the image quality is improved or the target quality image is screened out after quality score screening, it can be confirmed that the target person is a stranger.

[0087] To facilitate character tracking, the image data may have multiple frames. After extracting facial features of each character in a preset area based on the image data, the central server 102 may associate facial features of the same character in multiple frames of image data to obtain an information sequence of each character in the preset area.

[0088] That is to say, the information sequence of a certain person can be composed of the facial features of the person in multiple frames of image data.

[0089] In order to improve the accuracy of person recognition, the image data may include, in addition to the face region image, a human body region image obtained by the image acquisition device 101 performing human body recognition on the initial image. Human body recognition may be achieved by human body feature re-recognition and the like. The human body region image is the pixel region where the human body is located in the initial image.

[0090] Correspondingly, after step S201, the central server 102 may extract the body features of each person in the preset area according to the image data.

[0091] Among them, the human body features can be realized by a human body weight recognition algorithm or other existing feature extraction algorithms. The human body features can be used to match the facial features of each person in the preset area with the benchmark facial features of the people in the face library, and obtain the recognition results of each person in the preset area. Specifically, in some implementations, the human body features are compared with the benchmark human body features of the people in the library, and the facial features are compared with the benchmark facial features of the people in the library, and the human body comparison results of each person in the preset area and the facial comparison results of each person in the preset area can be obtained respectively. The human body comparison results and the facial comparison results are merged to obtain the recognition results of each person in the preset area. In other implementations, the human body features and facial features can be combined according to the matching habits of the human body features and facial features, so as to improve the accuracy of the image acquisition of a certain person's facial area, and finally improve the recognition results. For the face and human body pictures, the features extracted by the corresponding existing algorithms can be associated, and reliable tracking can be performed when the face is not aligned with the surveillance camera, while improving the recall rate and accuracy of tracking.

[0092] Similarly, when the number of frames of image data is multiple frames, the central server 102 can associate the facial features and body features belonging to the same person in the multiple frames of image data to obtain the information sequence of each person in the preset area. In one embodiment, the personal information sequence adopts an incremental update method, and the latest information is directly inserted into the end of the queue. For example, the facial feature comparison result time 1 to the facial feature comparison result time n can be obtained, where n is an integer greater than 1. In addition, as an example, it can be based on the above results to first determine whether the person meets the database condition, and then determine whether the stranger condition is met.

[0093] At this time, the obtained information sequence can be used to track people in multiple frames of image data, that is, to track the position changes, movement changes, etc. of the same person in different frames of image data, and then determine the movement trend of the person in a preset area.

[0094] In some embodiments of the present application, after obtaining the recognition result, the central server 102 may also send the recognition result to the monitoring device, and the monitoring device processes the recognition result. The number of the monitoring device may be one or more, and the present application does not limit this.

[0095] Furthermore, the image data may also include a coded image (for example, a compressed image or video) obtained by encoding one or more frames of the initial image by the image acquisition device 101. After obtaining the recognition result, the central server 102 may also send the recognition result and the coded image to the monitoring device, and the monitoring device processes the coded image and the recognition results of each person in the coded image.

[0096] For example, see Figure 6 The monitoring device can send a monitoring request to the central server 102 through the forwarding of the network terminal 2. In response to the monitoring request, the central server 102 can package the coded image and the recognition result into a monitoring data stream, and send the monitoring data stream to the monitoring device through the forwarding of the network terminal 2.

[0097] As an example, the monitoring device can display the coded image and the recognition result, and output a video of a preset area marked with the output result, for example, outputting a video of a preset area marked with a stranger.

[0098] As another example, the monitoring device may store the coded image and the recognition result. Based on the recognition result, if the number of strangers recognized is greater than the number threshold, or the number of times the strangers are recognized is greater than the number threshold, the coded image and the recognition result are output to the user's handheld terminal. For example, the coded image and the recognition result are output to the administrator's mobile phone to implement an alarm.

[0099] In other embodiments of the present application, the central server may include a background management module, which may refer to application software integrated on the central server 102, or may refer to a physical chip or processor of the central server 102. After obtaining the recognition result, the central server 102 may also send the recognition result to the background management module, which processes the recognition result.

[0100] Similarly, the background management module can process the recognition results and the encoded images at the same time.

[0101] For example, see Figure 6 The central server 102 can obtain image data through the network terminal 1 through the internal algorithm module, and the algorithm module processes the image data to obtain a recognition result, and then transmits the recognition result to the background management module for processing.

[0102] The specific processing process of the background management module can refer to the example of the aforementioned monitoring device, that is, the recognition results can be displayed, and alarms can be issued. In addition, data collation can be performed, and information processing and authority management such as new employee registration and resignation can be supported. This application will not go into details.

[0103] In some other embodiments, the image data may include one or more of a face region image obtained by performing face alignment on the initial image, face coordinate information obtained by performing face coordinate detection on the initial image, and body coordinate information obtained by performing body coordinate detection on the initial image. Similarly, the face coordinate information and body coordinate information may be used to track a person in multiple frames of image data.

[0104] For ease of understanding, Figure 7 The schematic diagram of the structure inside the above-mentioned central server 102 is shown. The central server 102 can obtain a face region image, a human body region image, face coordinate information, and human body coordinate information. Among them, the face region image can calculate the quality score through the evaluation module. When the quality score is greater than the score threshold, the face recognition module performs feature recognition, so that the face features are compared with the reference face features in the face library in the feature comparison module, and the face comparison result is output. Correspondingly, after the human body region image obtains the human body features through the human body weight recognition module, it can be transmitted to the tracking module together with the face coordinate information and the human body coordinate information. The tracking module performs character tracking according to the face comparison result, human body features, face coordinate information, and human body coordinate information, generates an information sequence, and the recognition module outputs the recognition result. For example, the tracking module generates a metric matrix based on the above information, and performs matching based on the Hungarian matching algorithm, thereby generating an information sequence.

[0105] Accordingly, the person recognition method provided by the present application is described below from the perspective of the image acquisition device 101. Figure 8 A schematic diagram of an implementation flow of a person recognition method provided in an embodiment of the present application is shown. The method can be applied to the image acquisition device 101 and is suitable for situations where the accuracy of person recognition needs to be improved.

[0106] Specifically, the above-mentioned person recognition method may include the following steps S801 to S803.

[0107] Step S801, collecting data from a preset area to obtain an initial image.

[0108] Step S802: Generate image data based on the initial image.

[0109] The image data may include a face area image obtained by performing face recognition on an initial image by the image acquisition device 101 .

[0110] Step S803: sending the image data to the central server.

[0111] The central server 102 may be used according to Figures 1 to 7 The described person recognition method processes the received image data.

[0112] It should be understood that the implementation of steps S801 to S803 can refer to Figures 1 to 7 The description of this application will not be repeated here.

[0113] In the process of generating image data based on the initial image, the image acquisition device 101 can perform face alignment on the initial image, such as alignment of facial key points, to obtain a face alignment image, and accordingly, the face region image can be a face alignment image obtained after face alignment. The image acquisition device 101 can also perform face coordinate detection on the initial image, and in this case, the image data can also include face coordinate information obtained after face coordinate detection.

[0114] For ease of understanding, Fig. 9 The schematic diagram of the structure inside the above-mentioned image acquisition device 101 is shown. The image acquisition device 101 can shoot a preset area through the image acquisition module to obtain a high-definition initial image. Then, the face detection module is used to perform face recognition on the initial image, and the face alignment module is used to perform face alignment to output the face area image. Correspondingly, the human body detection module can be used to perform human body detection on the initial image and output the human body area image. In order to reduce the transmission burden, the image acquisition device 101 can encode the initial image through the encoding module to obtain an encoded image. Then, the image acquisition device 101 can package the face area image, the human body area image and the encoded image through the image packaging module and output them to the central server 102.

[0115] It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders. Among them, the camera has a built-in AI algorithm to perform basic operations related to faces and bodies (such as face detection, alignment of face key points, and pedestrian detection). By running the algorithm directly on high-definition images, the accuracy of the algorithm is improved. At the same time, the computing pressure on the server is reduced. Only the aligned face and body images with a very small amount of additional data need to be transmitted, which avoids the direct transmission of high-definition images to the server and greatly reduces bandwidth requirements.

[0116] like Fig.10FIG. 1 is a schematic diagram of the structure of a person recognition device 1000 provided in an embodiment of the present application. The person recognition device 1000 is configured on a central server 102 .

[0117] Specifically, the person recognition device 1000 may include:

[0118] The image acquisition unit 1001 is used to acquire image data transmitted by the image acquisition device 101, wherein the image data includes a face region image obtained by the image acquisition device 101 performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device 101 acquiring a preset region;

[0119] A feature extraction unit 1002 is used to extract facial features of each person in the preset area according to the image data;

[0120] The person recognition unit 1003 is used to compare the facial features of each person in the preset area with the reference facial features of the people in the face library to obtain the recognition results of each person in the preset area.

[0121] In some embodiments of the present application, a face library stores a baseline image of each person in the library, and the baseline facial features of each person in the library are facial features obtained by performing facial recognition based on the corresponding baseline image; the above-mentioned person recognition device 1000 may include an updating unit, which is used to: if the recognition result of the target person is that the target person matches the person in the library, then store the facial area image of the target person in the pre-stored library of the target person, wherein the target person is any one or more persons in the preset area; when the face library meets the update condition, update the baseline image of the target person in the face library based on the pre-stored library of the target person.

[0122] In some embodiments of the present application, the above-mentioned update unit can be specifically used to: when the face library meets the update condition, calculate the selection probability of each face area image in the pre-stored library, and the selection probability is related to the shooting time of the face area image, and / or the image confidence of the face area image; according to the selection probability, filter out at least one target image from the pre-stored library of the target person; according to the target image, update the baseline image of the target person in the face library.

[0123] In some embodiments of the present application, the above-mentioned updating unit can be specifically used to: re-determine the benchmark facial features of the target person based on the updated image of the target person in the face library.

[0124] In some embodiments of the present application, the above-mentioned update unit can be specifically used to: calculate the feature center of the facial features corresponding to all the target images; calculate the cosine distance between the facial features corresponding to each of the target images and the feature center; and eliminate the target images whose cosine distance is greater than a distance threshold.

[0125] In some embodiments of the present application, the above-mentioned person recognition unit 1003 can be specifically used for: when the recognition result of the target person is that the target person does not match the person in the library, performing image quality improvement processing on the face area image to obtain an image with improved image quality, wherein the target person is any one or more persons in the preset area; based on the image with improved image quality, re-determining the recognition result of the target person.

[0126] In some embodiments of the present application, the person identification unit 1003 may be specifically used to: if the re-determined identification result still shows that the target person does not match the person in the database, confirm that the target person is a stranger.

[0127] In some embodiments of the present application, the above-mentioned person recognition device 1000 may include a scoring unit, which is used to: detect a quality score of the image quality of the face area image, and the quality score is related to at least one of the brightness of the face area image, the image clarity, and the orientation of the face in the face area image; the above-mentioned person recognition unit 1003 can be specifically used to: if the quality score is higher than the score threshold, then extract the facial features of each person in the preset area according to the image data.

[0128] In some implementations of the present application, the image data may also include an encoded image obtained after the image acquisition device 101 encodes the initial image.

[0129] In some embodiments of the present application, the above-mentioned person recognition device 1000 may include an output unit, which is used to: send the recognition result to a monitoring device, and the monitoring device processes the recognition result.

[0130] In some embodiments of the present application, the central server 102 may include a background management module, and the person recognition device 1000 may include a background management unit for sending the recognition result to the background management module, and the background management module processes the recognition result.

[0131] In some embodiments of the present application, the number of frames of image data is multiple frames; the above-mentioned person recognition device 1000 may include an association unit, which is used to: associate facial features belonging to the same person in the multiple frames of image data to obtain an information sequence of each person in the preset area.

[0132] In some embodiments of the present application, the image data may also include a human body area image obtained by the image acquisition device 101 performing human body recognition on the initial image; the above-mentioned feature extraction unit 1002 may be specifically used to: extract the human body features of each person in the preset area according to the image data; wherein the human body features are used to cooperate with the facial features of each person in the preset area to be compared with the benchmark facial features of the persons in the face library, so as to obtain the recognition results of each person in the preset area; the above-mentioned association unit may be specifically used to: associate the facial features and human body features belonging to the same person in the multiple frames of the image data, so as to obtain the information sequence of each person in the preset area.

[0133] It should be noted that, for the convenience and simplicity of description, the specific working process of the above-mentioned person recognition device 1000 can refer to Figures 1 to 7 The corresponding process of the method will not be repeated here.

[0134] like Fig.11 The figure is a schematic structural diagram of a person recognition device 1100 provided in an embodiment of the present application, wherein the person recognition device 1100 is configured on an image acquisition device 101 .

[0135] Specifically, the person recognition device 1100 may include:

[0136] The image acquisition unit 1101 is used to acquire an image of a preset area to obtain an initial image;

[0137] An image processing unit 1102 is used to generate image data based on the initial image, wherein the image data includes a face region image obtained by the image acquisition device 101 performing face recognition on the initial image;

[0138] The sending unit 1103 is used to send the image data to the central server 102, wherein the central server 102 is used to Figures 8 to 9 The steps of the described person recognition method are to process the received image data.

[0139] It should be noted that, for the convenience and simplicity of description, the specific working process of the above-mentioned person recognition device 1100 can refer to Figures 8 to 9 The corresponding process of the method will not be repeated here.

[0140] like Fig.12 , which is a schematic diagram of an electronic device provided in an embodiment of the present application, wherein the electronic device may refer to the aforementioned image acquisition device 101 or the central server 102.

[0141] Specifically, the electronic device 120 may include: a processor 1200, a memory 1201, and a computer program 1202 stored in the memory 1201 and executable on the processor 1200, such as a person recognition program. When the processor 1200 executes the computer program 1202, the steps in the above-mentioned various person recognition method embodiments are implemented, such as Figure 2 Alternatively, when the processor 1200 executes the computer program 1202, the steps in the above-mentioned various person recognition method embodiments are implemented, for example, Figure 8 Steps S801 to S803 are shown.

[0142] When the processor 1200 executes the computer program 1202, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Fig.10 The functions of the image acquisition unit 1001, the feature extraction unit 1002, and the person recognition unit 1003 are shown. Alternatively, when the processor 1200 executes the computer program 1202, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Fig.11 The functions of the image acquisition unit 1101, the processing unit 1102, and the sending unit 1103 are shown.

[0143] The computer program may be divided into one or more modules / units, which are stored in the memory 1201 and executed by the processor 1200 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0144] For example, the computer program can be divided into: an image acquisition unit, a feature extraction unit, and a person recognition unit. The specific functions of each unit are as follows: an image acquisition unit, used to acquire image data transmitted by the image acquisition device 101, wherein the image data includes a face area image obtained by the image acquisition device 101 performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device 101 acquiring a preset area; a feature extraction unit, used to extract the face features of each person in the preset area according to the image data; a person recognition unit, used to compare the face features of each person in the preset area with the reference face features of the people in the face library, respectively, to obtain the recognition results of each person in the preset area.

[0145] For another example, the computer program can be divided into: an image acquisition unit, an image processing unit, and a sending unit. The specific functions of each unit are as follows: the image acquisition unit is used to acquire a preset area to obtain an initial image; the image processing unit is used to generate image data based on the initial image, and the image data includes a face area image obtained by the image acquisition device 101 performing face recognition on the initial image; the sending unit is used to send the image data to the central server 102, wherein the central server 102 is used to Figures 8 to 9 The steps of the described person recognition method are to process the received image data.

[0146] The electronic device may include, but is not limited to, a processor 1200 and a memory 1201. Those skilled in the art will appreciate that Fig.12 These are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0147] The processor 1200 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0148] The memory 1201 may be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device. The memory 1201 may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 1201 may also include both an internal storage unit of the electronic device and an external storage device. The memory 1201 is used to store the computer program and other programs and data required by the electronic device. The memory 1201 may also be used to temporarily store data that has been output or is to be output.

[0149] It should be noted that, for the convenience and brevity of description, the structure of the above electronic device can also refer to the specific description of the structure in the method embodiment, which will not be repeated here.

[0150] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0151] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software 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 this application.

[0153] In the embodiments provided in the present application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0154] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0155] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0156] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0157] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A person recognition method, characterized in that: Applied to a central server, the person recognition method includes: Acquire image data transmitted by an image acquisition device, the image data comprising a face region image obtained by the image acquisition device performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device acquiring a preset region; Extracting facial features of each person in the preset area according to the image data; The facial features of each person in the preset area are compared with the benchmark facial features of the people in the face library to obtain the recognition results of each person in the preset area.

2. The person recognition method according to claim 1, characterized in that: The face database stores a reference image of each person in the database, and the reference face feature of each person in the database is a face feature obtained by performing face recognition based on the corresponding reference image; After obtaining the recognition results of each person in the preset area, the method further includes: If the recognition result of the target person is that the target person matches the person in the library, the face area image of the target person is stored in the pre-stored library of the target person, wherein the target person is any one or more persons in the preset area; When the face library meets the update condition, the reference image of the target person in the face library is updated based on the pre-stored library of the target person.

3. The person recognition method according to claim 2, characterized in that: When the face library meets the update condition, the reference image of the target person in the face library is updated based on the pre-stored library of the target person, including: When the face database meets the update condition, calculating the selection probability of each face region image in the pre-stored database, wherein the selection probability is related to the shooting time of the face region image and / or the image confidence of the face region image; According to the selection probability, at least one target image is selected from a pre-stored library of the target person; The reference image of the target person in the face library is updated according to the target image.

4. The person recognition method according to claim 3, characterized in that: After the reference image of the target person in the face library is updated according to the target image, the method further includes: re-determining the reference facial features of the target person based on the updated image of the target person in the face library; and / or, There are multiple target images, and according to the selection probability, at least one target image is screened out from the pre-stored library of the target person, which also includes: calculating the feature center of the facial features corresponding to all the target images; calculating the cosine distance between the facial features corresponding to each of the target images and the feature center; and eliminating the target images whose cosine distance is greater than a distance threshold.

5. The person recognition method according to claim 1, characterized in that: After obtaining the recognition results of each person in the preset area, the method further includes: When the recognition result of the target person is that the target person does not match the person in the database, the image of the face area is processed to improve the image quality to obtain an image with improved image quality, wherein the target person is any one or more persons in the preset area; Based on the image with improved image quality, the recognition result of the target person is re-determined.

6. The person recognition method according to claim 5, characterized in that: After re-determining the recognition result of the target person, the method further includes: If the re-determined recognition result still shows that the target person does not match the person in the database, it is confirmed that the target person is a stranger.

7. The person recognition method according to claim 1, characterized in that: After acquiring the image data transmitted by the image acquisition device, the method further includes: Detecting a quality score of the image quality of the face region image, wherein the quality score is related to at least one of the brightness of the face region image, the image clarity, and the orientation of the face in the face region image; Extracting facial features of each person in the preset area according to the image data includes: If the quality score is higher than the score threshold, extracting facial features of each person in the preset area based on the image data; And / or, after obtaining the recognition results of each person in the preset area, the method further includes: When the recognition result of the target person is that the target person does not match the person in the database, the face area image is screened by quality score to screen out the target quality image, and whether the person is a stranger is determined by the recognition result corresponding to the screened target quality image.

8. The person recognition method according to claim 1, characterized in that: The image data also includes a coded image obtained after the image acquisition device encodes the initial image; and / or, After obtaining the recognition results of each person in the preset area, the method further includes: sending the recognition results to a monitoring device, the number of which is at least one; and / or, The central server also includes a background management module. After obtaining the recognition results of each person in the preset area, it also includes: sending the recognition results to the background management module, and the background management module processes the recognition results.

9. The person recognition method according to claim 1, characterized in that: The number of frames of the image data is multiple frames; After extracting the facial features of each person in the preset area according to the image data, the method further includes: The facial features of the same person in the multiple frames of image data are associated to obtain an information sequence of each person in the preset area.

10. The person recognition method according to any one of claims 1 to 9, characterized in that: The image data also includes a human body region image obtained by the image acquisition device performing human body recognition on the initial image; After acquiring the image data transmitted by the image acquisition device, the method further includes: Extracting the body features of each person in the preset area according to the image data; wherein the body features are used to match the facial features of each person in the preset area with the reference facial features of the persons in the face library to obtain the recognition results of each person in the preset area; Wherein, when the number of frames of the image data is multiple, the facial features of the same person in the multiple frames of the image data are associated to obtain the information sequence of each person in the preset area, and further includes: The facial features and body features belonging to the same person in the multiple frames of image data are associated to obtain the information sequence of each person in the preset area.

11. A method for identifying a person, characterized in that: Applied to an image acquisition device, the person recognition method comprises: Capture the preset area to obtain an initial image; Based on the initial image, generating image data, the image data comprising a face area image obtained by performing face recognition on the initial image by the image acquisition device; The image data is sent to a central server, wherein the central server is used to process the received image data according to the steps of the person recognition method according to any one of claims 1 to 10.

12. The person recognition method according to claim 11, characterized in that: The process of generating image data based on the initial image includes: Performing face alignment on the initial image, wherein the initial image is a face alignment image obtained after performing the face alignment; and / or, Face coordinate detection is performed on the initial image, and the image data also includes face coordinate information obtained after the face coordinate detection.

13. A person recognition device, applicable to the person recognition method according to any one of claims 1 to 10, characterized in that: Configured on a central server, the person recognition device includes: An image acquisition unit, used to acquire image data transmitted by an image acquisition device, wherein the image data includes a face region image obtained by the image acquisition device performing face recognition on an initial image, wherein the initial image is an image obtained by the image acquisition device acquiring a preset region; A feature extraction unit, used to extract facial features of each person in the preset area according to the image data; The person recognition unit is used to compare the facial features of each person in the preset area with the reference facial features of the people in the face library to obtain the recognition results of each person in the preset area.

14. A person recognition device, applicable to the person recognition method according to any one of claims 11 to 12, characterized in that: Configured in an image acquisition device, the person recognition device comprises: An image acquisition unit, used to acquire images of a preset area to obtain an initial image; An image processing unit, configured to generate image data based on the initial image, wherein the image data includes a face region image obtained by performing face recognition on the initial image by the image acquisition device; A sending unit is used to send the image data to a central server, wherein the central server is used to process the received image data according to the steps of the person recognition method according to any one of claims 1 to 9.

15. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the person recognition method according to any one of claims 1 to 10 are implemented, or when the processor executes the computer program, the steps of the person recognition method according to claim 11 or 12 are implemented.

16. A person recognition system, characterized in that: include: An image acquisition device and a central server connected to the image acquisition device; The image acquisition device is used to send the image data to the central server according to the steps of the person recognition method according to claim 11 or 12; The central server is used to process the received image data according to the steps of the person recognition method according to any one of claims 1 to 10.

17. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the person recognition method according to any one of claims 1 to 10 are implemented; or, when the computer program is executed by a processor, the steps of the person recognition method according to claim 11 or 12 are implemented.

Citation Information

Cited By

  • Recognition processing method for large mine materials and related equipment

    CN120411650A