Feature data processing method and device, storage medium and electronic device

Abstract

The application discloses a feature data processing method and device, a storage medium and an electronic device, and relates to the technical field of smart home / smart home, and the feature data processing method comprises the following steps: acquiring a first object feature of a target object, wherein the first object feature is a face recognition feature of the target object which is input by using a first device; saving the first object feature into a target feature library corresponding to the target object, and determining a second device to be distributed from a group of associated devices of the target object, wherein the second device is a device allowing input of the face recognition feature; and distributing the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the input face recognition feature of the target object. Through the application, the problem that the information processing efficiency is low due to the complicated processing operation in the related art device information processing method is solved.

Description

Technical Field

[0001] This application relates to the field of communications, and more specifically, to a method and apparatus for processing feature data, a storage medium, and an electronic device. Background Technology

[0002] In related technologies, visual devices in users' homes (i.e., devices with cameras, such as outdoor smart doorbells, indoor surveillance cameras, TV cameras, refrigerator screen cameras, etc.) are becoming increasingly abundant. Users can register family members through the cameras on these visual devices. When a user registers, the visual device can record the user's relevant information (e.g., age information, family identity information). After registration, the device can identify the user. When the user is identified as a family member, more personalized services can be provided based on the recorded user information corresponding to that user.

[0003] During the registration process, users need to follow the instructions of the vision device to capture images and register family members based on the captured user images. Because the information entry process is cumbersome and may require multiple entries due to insufficient image clarity, the registration process is inefficient.

[0004] It is evident that the equipment information processing methods in related technologies suffer from low efficiency due to cumbersome processing operations. Summary of the Invention

[0005] This application provides a method and apparatus for processing feature data, a storage medium, and an electronic device to at least solve the problem of low information processing efficiency caused by cumbersome processing operations in related technologies for processing device information.

[0006] According to one aspect of the embodiments of this application, a method for processing feature data is provided, comprising: acquiring a first object feature of a target object, wherein the first object feature is a facial recognition feature of the target object recorded using a first device; saving the first object feature to a target feature library corresponding to the target object, and determining a second device to be distributed from a set of associated devices of the target object, wherein the second device allows devices to record facial recognition features; distributing the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the recorded facial recognition feature of the target object.

[0007] According to another aspect of the embodiments of this application, a feature data processing apparatus is also provided, comprising: an acquisition unit, configured to acquire a first object feature of a target object, wherein the first object feature is a facial recognition feature of the target object recorded using a first device; an execution unit, configured to save the first object feature to a target feature library corresponding to the target object, and determine a second device to be distributed from a set of associated devices of the target object, wherein the second device allows devices to record facial recognition features; and a distribution unit, configured to distribute the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the recorded facial recognition feature of the target object.

[0008] In an exemplary embodiment, the acquisition unit includes: a recognition module, configured to perform face recognition on a first object image using a first processing device to obtain the features of the first object, wherein the first object image is an image of the target object recorded by the first device, and the first processing device is a device other than the first device whose remaining computing resources are greater than or equal to the computing resources required to perform face recognition on the first object image.

[0009] In one exemplary embodiment, the apparatus further includes: a selection unit, configured to select the first processing device from the plurality of candidate devices based on the remaining computing resources of each candidate device before performing face recognition on the first object image through the first processing device, wherein the plurality of candidate devices includes at least one of the following: smart devices under the target object group where the target object is located, and smart devices within the target area range where the first device is located.

[0010] In one exemplary embodiment, the apparatus further includes: a first determining unit, configured to determine the remaining computing resources of each candidate device based on device resource information of each candidate device before selecting the first processing device from the plurality of candidate devices according to the remaining computing resources of each candidate device, wherein the device resource information includes at least one of the following: available computing resources, usage percentage of available computing resources, network signal status, and available transmission bandwidth; and a second determining unit, configured to determine the computing resources required for face recognition of the first object image based on image attribute information of the first object image, wherein the image attribute information includes at least one of the following: image type and image size.

[0011] In an exemplary embodiment, the execution unit includes at least one of the following: a selection module, configured to select, from the set of associated devices, an associated device whose device permissions match the object category of the target object, to obtain the second device; a first determining module, configured to determine the recorded device in the set of associated devices as the second device, wherein the recorded device is a device that has stored the facial recognition features of the target object; a second determining module, configured to determine the authorized device in the set of associated devices as the second device, wherein the authorized device is a device authorized to share the facial recognition features of the target object; a third determining module, configured to determine, from the set of associated devices, the associated device to which the first object features are to be distributed, as indicated by the device indication information, as the second device; and a fourth determining module, configured to determine, from the set of associated devices, the device that requests to obtain the facial recognition features of the target object as the second device.

[0012] In one exemplary embodiment, the apparatus further includes: a first extraction unit, configured to, after distributing the target feature library to the second device, extract facial features from the second object image through a second processing device to obtain the second object feature of the object to be verified, wherein the second processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to perform facial recognition on the second object image; and a first sending unit, configured to send a target recognition result to the second device when the second object feature matches the first object feature in the target feature library, wherein the target recognition result is used to indicate that the object to be verified has been verified.

[0013] In an exemplary embodiment, the second device is a detection device for detecting target events; the apparatus further includes: a second extraction unit, configured to, after distributing the target feature library to the second device, extract target events from the target data to be detected by a third processing device when the second device obtains the target data to be detected, thereby obtaining a detection result of the target event, wherein the third processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to detect the target event from the target data to be detected; and a second sending unit, configured to send alarm information to the target device when the detection result of the target event indicates that the target event has occurred, wherein the alarm information is used to alarm for the occurrence of the target event.

[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the above-described feature data processing method when running.

[0015] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for processing feature data through the computer program.

[0016] In this embodiment, a method of sharing facial recognition features between different devices of the same user is adopted to obtain the first object feature of the target object, wherein the first object feature is the facial recognition feature of the target object recorded using the first device; the first object feature is saved in the target feature library corresponding to the target object, and a second device to be distributed is determined from a group of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features; the target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved as the recorded facial recognition feature of the target object on the second device. Since the facial recognition feature of the target object obtained by the first device is directly distributed to the second device, the second device can obtain the facial recognition feature of the target object without collecting the facial recognition feature of the target object, thereby reducing the operation of obtaining facial recognition features on multiple devices, achieving the technical effect of improving the efficiency of device information processing, and thus solving the problem of low information processing efficiency caused by cumbersome processing operations in the related technology of device information processing methods. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the hardware environment for an optional feature data processing method according to an embodiment of this application;

[0020] Figure 2 This is a flowchart illustrating an optional feature data processing method according to an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of an optional face registration process according to an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of an optional face recognition process according to an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of an optional home emergency detection process according to an embodiment of this application;

[0024] Figure 6 This is a structural block diagram of an optional feature data processing apparatus according to an embodiment of this application;

[0025] Figure 7 This is a structural block diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] According to one aspect of the embodiments of this application, a method for processing feature data is provided. This method for processing feature data is widely used in whole-house intelligent digital control application scenarios such as smart homes, smart home ecosystems, and intelligence house ecosystems. Optionally, in this embodiment, the above-mentioned method for processing feature data can be applied to, for example... Figure 1 The hardware environment shown consists of terminal 102 and server 104. Figure 1 As shown, server 104 is connected to terminal 102 via a network and can be used to provide services (such as application services) to the terminal or clients installed on the terminal. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data processing services to server 104.

[0029] The aforementioned network may include, but is not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. Terminal 102 may not be limited to PC, mobile phone, tablet computer, smart air conditioner, smart range hood, smart refrigerator, smart oven, smart stove, smart washing machine, smart water heater, smart washing equipment, smart dishwasher, smart projector, smart TV, smart clothes rack, smart curtains, smart audio-visual equipment, smart socket, smart speaker, smart speaker box, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart robot vacuum cleaner, smart window cleaning robot, smart mopping robot, smart air purifier, smart steam oven, smart microwave oven, smart water heater, smart air purifier, smart water dispenser, smart door lock, etc.

[0030] The feature data processing method of this application embodiment can be executed by server 104, terminal 102, or jointly by server 104 and terminal 102. Alternatively, the feature data processing method of this application embodiment can be executed by a client installed on terminal 102.

[0031] Taking the feature data processing method in this embodiment executed by server 104 as an example, Figure 2 This is a flowchart illustrating an optional feature data processing method according to an embodiment of this application, such as... Figure 2 As shown, the process of this method may include the following steps:

[0032] Step S202: Obtain the first object feature of the target object, wherein the first object feature is the facial recognition feature of the target object recorded using the first device.

[0033] The feature data processing method in this embodiment can be applied to scenarios where multiple devices perform collaborative transaction processing. These multiple devices can be distributed vision devices on the same network, and may include, but are not limited to, at least one of the following: outdoor smart doorbells / peepholes, indoor surveillance cameras, television cameras, refrigerator screen cameras, etc. They can also be devices that support other acquisition methods for object feature extraction. This embodiment uses vision devices as an example for explanation. The aforementioned collaborative transaction processing may include, but is not limited to, one of the following: multi-device shared identity feature registration, multi-device shared identity feature recognition, and emergency event monitoring. This embodiment does not limit this to any of these.

[0034] Taking identity registration as an example, in the field of smart homes, users' homes are increasingly equipped with a variety of visual devices, including cameras. A single household can have multiple distributed visual devices simultaneously. In related technologies, each of these distributed visual devices only has its own local functions. For instance, when users use devices with facial recognition capabilities, such as televisions or refrigerators, they need to complete cumbersome registration processes on different devices before facial recognition can be performed, resulting in low user adoption rates.

[0035] In this embodiment, user features acquired by one device can be distributed to other devices to simplify the user information entry process on those devices. For example, multiple distributed cameras on various smart home devices within a user's home can share a face registration method. After one camera registers, the registered face information is distributed to other devices, allowing feature vectors collected across devices to be shared and preventing duplicate registration across multiple devices within the home. Simultaneously, by sharing computing power with other hardware resources, ordinary cameras lacking computing resources can be upgraded to cameras with AI (Artificial Intelligence) capabilities, enabling seamless user recognition across various devices within the home, improving information processing efficiency, and ultimately increasing user engagement.

[0036] In this embodiment, the resource scheduling device can acquire a first object feature of the target object. The first object feature can be a facial recognition feature of the target object recorded using the first device. The first device can record object data of the target object, and the first object feature can be obtained by extracting features from the object data of the target object. The resource scheduling device can be a server 104 (e.g., a cloud server) or a terminal 102. It can be the first device or other devices connected to the server 104 besides the first device. This embodiment does not limit this.

[0037] The target object mentioned above can be an object that has established a binding or association relationship with the first device, or it can be an object within the collection range of the first device. The first object feature mentioned above can be a facial recognition feature recorded by facial capture through an image acquisition component. In addition, the first object feature can also be a fingerprint feature recorded by a fingerprint acquisition component, or an object feature recorded by a biometric feature acquisition component. This embodiment does not limit this.

[0038] The device that extracts features from the object data of the target object can be a first device, a resource scheduling device, or other devices besides the first device and the resource scheduling device, or at least two of the first device, the resource scheduling device, and other devices. For example, the first device, the resource scheduling device, or other devices can extract features from the object data of the target object to obtain the first object features. As another example, at least two of the first device, the resource scheduling device, and other devices can extract features from a portion of the object data of the target object to obtain the first object features.

[0039] Taking facial feature acquisition as an example, the process of obtaining the first object feature of the target object can be as follows: acquiring a facial image of the target object through a first device, performing image recognition on the acquired facial image locally to obtain the first object feature of the target object, and forwarding the first object feature to a resource scheduling device; or acquiring a facial image of the target object through a first device, forwarding the acquired facial image to a resource scheduling device, and having the resource scheduling device perform image recognition on the acquired facial image to obtain the first object feature of the target object; or acquiring a facial image of the target object through a first device, providing a portion of the acquired facial image to the first device, and forwarding the other portion to a resource scheduling device, with the resource scheduling device and the first device jointly performing image recognition on the acquired facial image to obtain the first object feature of the target object. This embodiment does not limit this approach.

[0040] Optionally, the resource scheduling device may determine the device that processes all or part of the face images, and the first device may interact with the determined device so that the determined device performs the image recognition process on all or part of the face images. Similar processing logic may be used for other object data, and this embodiment does not limit this.

[0041] Step S204: Save the first object features to the target feature library corresponding to the target object, and determine the second device to be distributed from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features.

[0042] After acquiring the first object's features, the resource scheduling device can save the first object's features to a target feature library corresponding to the target object, and determine a second device to be distributed from a set of associated devices of the target object. The second device is a device that allows the recording of facial recognition features, that is, a device that supports facial recognition. The aforementioned set of associated devices of the target object can be devices bound to the target object, and can include the first device and the second device.

[0043] Optionally, the first device and the second device can be the same type of device or different types of devices. For example, the first device can be a television set with a camera, and the second device can be a terminal device with a camera or the same type of device. This embodiment does not limit this.

[0044] In this embodiment, the resource scheduling device can record the feature types of object features supported by each associated device in a group of associated devices. Optionally, the first object feature may carry a feature type corresponding to the first object feature. After receiving the first object feature, the resource scheduling device can determine the device in the group of associated devices that supports the feature type corresponding to the first object feature as the second device based on the feature type carried in the first object feature and the recorded feature types of object features supported by each associated device.

[0045] It should be noted that the second device can be multiple devices or a single device. For example, when there are multiple devices in a group of associated devices that are allowed to record facial recognition features, some or all of the multiple devices can be designated as the second device, or only one of the devices can be designated as the second device. This embodiment does not impose any limitations on this.

[0046] Step S206: Distribute the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

[0047] In this embodiment, after identifying the second device, the resource scheduling device can distribute the target feature library to the second device. The first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object. Optionally, after receiving the target feature library, the second device can use the target feature library to verify whether the current user of the second device is the target object. For example, when a user passes by any device with a visual sensor, such as a mobile phone, television, refrigerator, smart doorbell, or smart indoor monitoring camera, the user's identity can be identified through the user's facial features.

[0048] Optionally, the resource scheduling device can synchronize the target feature library with the local feature library of the second device. The process of synchronizing the target feature library with the local feature library of the second device can be: adding the first object feature from the target feature library to the local feature library of the second device, or updating the facial recognition feature of the target object stored in the local feature library of the second device to the first object feature. When the first device has data storage capabilities, the resource scheduling device can also send the target feature library to the first device, and the first device can synchronize the target feature library with its local feature library. This embodiment does not limit this aspect.

[0049] Optionally, in order to reduce the consumption of storage resources by the first object feature, the resource scheduling device may also upload the first object feature to the server corresponding to the resource scheduling device, and the server may save the first object feature as the object feature of the target object, or send it to the corresponding database for storage. This embodiment does not limit this.

[0050] During data transmission, the amount of data transmitted is directly proportional to the transmission time; that is, the larger the amount of data transmitted, the longer the transmission time. To reduce the time spent during data transmission, the target feature library can be compressed to obtain a compressed target feature library, which can then be distributed to the second device. This embodiment does not limit this approach.

[0051] During the distribution of the target signature database to a second device, the database is vulnerable to theft. To better protect user privacy, the target signature database can be encrypted before distribution to the second device, resulting in an encrypted database. This encrypted database can then be distributed to the second device, reducing the risk of the database being stolen.

[0052] Through the above steps, a first object feature of the target object is obtained, wherein the first object feature is the facial recognition feature of the target object recorded using a first device; the first object feature is saved to a target feature library corresponding to the target object, and a second device to be distributed is determined from a group of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features; the target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved on the second device as the recorded facial recognition feature of the target object. This solves the problem of low information processing efficiency caused by cumbersome processing operations in the related technology of device information processing methods, and improves the efficiency of device information processing.

[0053] In one exemplary embodiment, obtaining a first object feature of the target object includes:

[0054] S11, the first object image is subjected to face recognition by the first processing device to obtain the features of the first object, wherein the first object image is the image of the target object recorded by the first device, and the first processing device is a device other than the first device whose remaining computing resources are greater than or equal to the computing resources required to perform face recognition on the first object image.

[0055] In this embodiment, an object image of the target object can be acquired through a first device, and the acquired object image can be used for recognition to obtain a first object feature. Here, the acquired object image is the first object image. Optionally, a first processing device can perform face recognition on the first object image to obtain the first object feature. The first object image is the image of the target object recorded by the first device, and the first processing device is a device other than the first device whose remaining computing resources are greater than or equal to the computing resources required to perform face recognition on the first object image.

[0056] Optionally, the first device can send the first object image to the resource scheduling device, which will then perform computational resource scheduling. The first object image can be sent by the first device when it determines that its remaining computational resources are insufficient to meet the computational requirements for face recognition of the first object image. The first device can send the first object image to the resource scheduling device, or it can send feature information obtained after processing the first object image, i.e., target feature information. The resource scheduling device can store the remaining computational resources of each of the multiple candidate devices, and select the first processing device from the multiple candidate devices based on the remaining computational resources of each candidate device.

[0057] The selection criteria for the processing device can be: the remaining computing resources can meet the requirements for face recognition of the first object image, for example, the remaining computing resources are greater than or equal to the computing resources required for face recognition of the first object image; or the remaining computing resources can meet the requirements for processing target feature information, for example, the remaining computing resources are greater than or equal to the computing resources required for face recognition based on target feature information. Optionally, the target feature information can also be obtained by processing the first object image using a resource scheduling device or other processing device, which is not limited in this embodiment. The following embodiments use the processing of the first object image as an example for illustration, and the same applies to target feature information unless otherwise specified.

[0058] For example, the process of performing face recognition on the first object image can be as follows: firstly, the first object image is initially recognized by the first device, and then the initially recognized first object image (i.e., target feature information) is sent to the first processing device, which then continues to perform face recognition on the first object image to obtain the first object features.

[0059] Optionally, the first processing device may be one device or multiple devices. For example, the device whose remaining computing resources among multiple candidate devices are greater than or equal to the computing resources required for face recognition of the first object image may be determined as the first processing device. If there are no candidate devices among multiple candidate devices whose remaining computing resources are greater than or equal to the computing resources required for face recognition of the first object image, at least two devices whose sum of remaining computing resources among multiple candidate devices is greater than or equal to the computing resources required for face recognition of the first object image may be determined as the first processing device.

[0060] When there are multiple first processing devices, the remaining computing resources of a single device may be less than the computing resources required for face recognition of the first object image. However, the sum of the remaining computing resources of all devices may be greater than or equal to the computing resources required for face recognition of the first object image. In this case, the first object image can be segmented into multiple object sub-images, and these sub-images can be distributed to multiple devices for face recognition to obtain multiple first object sub-features. Finally, these sub-features are concatenated to obtain the first object feature. This embodiment does not impose any limitations on this process.

[0061] In this embodiment, object feature extraction can be performed collaboratively by devices other than the device that acquired the object image, which can improve the efficiency of object feature extraction.

[0062] In one exemplary embodiment, before performing face recognition on the first object image via the first processing device, the method further includes:

[0063] S21, based on the remaining computing resources of each of the multiple candidate devices, select a first processing device from the multiple candidate devices, wherein the multiple candidate devices include at least one of the following: intelligent devices under the target object group where the target object is located, and intelligent devices within the target area range where the first device is located.

[0064] In this embodiment, a first processing device can be determined from a plurality of candidate devices, which include at least one of the following: smart devices under the target object group where the target object is located (e.g., devices in the user's home), and smart devices within the target area range where the first device is located (e.g., devices in the user's building).

[0065] As an optional implementation, a first processing device can be selected from the smart devices in the target object group based on the remaining computing resources of each smart device in the target object group.

[0066] The smart devices under the aforementioned target object group can be smart devices that have established a binding relationship with the terminal device of the target object, or smart devices that store user information corresponding to the user information of the target object. This embodiment does not limit this. For example, a device that has established a binding relationship with the mobile phone of the target object can be identified as a candidate device.

[0067] As another alternative implementation, a first processing device can be selected from the smart devices within the target area where the first device is located, based on the remaining computing resources of each smart device within the target area where the first device is located.

[0068] Optionally, the target area may be a smart device whose distance from the first device is less than or equal to a preset distance threshold, a smart device whose network is the same as the network connected to the first device, or a device located in the same building as the first device. This embodiment does not limit this.

[0069] This embodiment uses multiple methods to determine the device for facial recognition of the acquired object images, which can improve the success rate of object feature extraction.

[0070] In one exemplary embodiment, before selecting a first processing device from the plurality of candidate devices based on the remaining computing resources of each candidate device, the method further includes:

[0071] S31. Based on the device resource information of each candidate device, determine the remaining computing resources of each candidate device, wherein the device resource information includes at least one of the following: available computing resources, usage percentage of available computing resources, network signal status, and available transmission bandwidth.

[0072] S32, based on the image attribute information of the first object image, determine the computing resources required for face recognition of the first object image, wherein the image attribute information includes at least one of the following: image type, image size.

[0073] In this embodiment, before selecting the first processing device from multiple candidate devices based on the remaining computing resources of each candidate device, the resource scheduling device may first obtain the device resource information of each candidate device, and then determine the remaining computing resources of each candidate device based on the device resource information of each candidate device. The device resource information includes at least one of the following: available computing resources, the usage ratio of available computing resources, network signal status, and available transmission bandwidth.

[0074] Optionally, the resource scheduling device may obtain the device resource information of each candidate device in one or more ways. For example, each associated device may periodically (or non-periodically) report its remaining computing resources. The resource scheduling device may determine the remaining computing resources of each associated device based on the recorded remaining computing resources of each associated device. Alternatively, when the resource scheduling device needs to determine the remaining computing resources of each associated device, it may send an instruction message to each associated device to instruct each associated device to report its remaining computing resources.

[0075] For example, each end-side device can report its current available hardware computing resource usage percentage, current network signal level K, and available transmission bandwidth B to the resource scheduling device at time intervals. The resource scheduling device can obtain basic information about the computing resources of each end-side device, such as basic computing power (CPU, Central Processing Unit), GPU (Graphics Processing Unit) / NPU (Neural-Network Processing Unit), memory, and other basic configuration information.

[0076] The resource scheduling device can calculate the current available computing power G of each end-side device based on the hardware resource usage ratio and the corresponding basic hardware information of the device. It then comprehensively determines the current network status and the set of available devices, i.e., the available end-side devices, based on the network quality level K and the available transmission bandwidth B. From this set of available devices, devices with sufficient computing resources are selected for algorithmic information processing to extract facial features.

[0077] Since the resources consumed in performing face recognition on the first object image to obtain the features of the first object are related to the size and resolution of the image, the resource scheduling device may optionally determine the computing resources required for performing face recognition on the first object image based on the image attribute information of the first object image. The image attribute information includes at least one of the following: image type and image size.

[0078] In this embodiment, by scheduling among a group of associated devices, one or more devices can work together to perform object feature extraction, thereby improving the efficiency of object feature extraction.

[0079] In one exemplary embodiment, determining a second device to be distributed from a set of associated devices of a target object includes at least one of the following:

[0080] S41, From a group of associated devices, select the associated device whose device permissions match the object category of the target object to obtain the second device;

[0081] S42, identify the recorded device in a group of associated devices as the second device, wherein the recorded device is a device that has stored the facial recognition features of the target object;

[0082] S43, identify the authorized device in a group of associated devices as the second device, wherein the authorized device is a device authorized to share the facial recognition features of the target object;

[0083] S44, among a group of associated devices, the associated device to which the first object feature is to be distributed, as indicated by the device indication information, is identified as the second device;

[0084] S45, among a group of associated devices, the device that requests to obtain the facial recognition features of the target object is identified as the second device.

[0085] In this embodiment, the second device can be determined from a group of associated devices in various ways, including but not limited to at least one of the following methods:

[0086] Method 1: From a group of associated devices, select the associated device whose device permissions match the object category of the target object to obtain the second device.

[0087] Since some devices are dangerous to specific users—for example, turning on a water heater or other dangerous device may cause harm to children or the elderly—device permissions can be set for different devices. Device permissions can be used to indicate the types of objects that are allowed to use the device and the types of objects that are not allowed to use the device.

[0088] Method 2: The device that has been registered in a group of associated devices can be identified as the second device. The device that has been registered is the device that has stored the facial recognition features of the target object.

[0089] The resource scheduling device can identify a device that has been registered in a group of associated devices as a second device. The registered device is a device that has stored the facial recognition features of the target object. Here, the registered device can be a device that has registered the object features of the target object using facial recognition based on the detection of user operation, or it can be a device that stores the object features of the target object that have been registered using facial recognition and shared by other devices. This embodiment does not limit this.

[0090] Method 3: An authorized device from a group of associated devices can be identified as the second device, where the authorized device is a device authorized to share the facial recognition features of the target object.

[0091] To protect user privacy, devices that share user characteristics can be determined based on user authorization. Resource scheduling devices can identify authorized devices from a group of associated devices as secondary devices; these authorized devices are those authorized to share the facial recognition features of the target object.

[0092] For example, a user authorizes devices A and B to share the facial recognition features of a target object, but does not authorize device C. After obtaining the facial features recorded by device A using the facial recognition method, the user can share the recorded facial features with device B, but not with device C. Here, both device B and device C support the facial recognition method.

[0093] Method 4: In a group of associated devices, the first device can be identified as the second device based on the associated device to which the first object feature is to be distributed, as indicated by the device indication information.

[0094] After obtaining the first object feature of the target object, the first object feature can be sent to the device specified by the first device. Optionally, the first object feature can be sent to a group of associated devices, where the first device is assigned to the associated devices indicated by the device indication information.

[0095] For example, after device A collects the user's facial recognition features, it can send those features to device B.

[0096] Method 5: The device that requests to obtain the facial recognition features of the target object from a group of associated devices can be identified as the second device.

[0097] After obtaining the first object feature of the target object, the first object feature can be sent to the device that requests to obtain the facial recognition feature of the target object. Optionally, the first object feature can be sent to a group of associated devices that request to obtain the facial recognition feature of the target object.

[0098] For example, after device A collects a user's facial recognition features, if it receives an acquisition request from device B, it can specify that the facial recognition features be sent to device B.

[0099] It should be noted that the second device can be any of the devices mentioned above, or it can be any other device in a group of associated devices. For example, it can be any device in a group of associated devices that needs to use the first object feature. This embodiment does not limit this.

[0100] This embodiment allows for the sharing of recorded user characteristics with other devices, such as devices that have already recorded or authorized user characteristics, thereby improving the flexibility and security of user characteristic sharing.

[0101] The processing method for feature data in the embodiments of this application will be explained below with reference to optional examples. In this optional example, a group of associated devices is a group of vision devices in the same household, and the object feature of the target object is the facial feature of the target object.

[0102] Currently, devices equipped with visual sensors can perform computational processing based on their own hardware resources. For example, algorithms such as face registration and face recognition can be implemented offline on the device itself. However, face recognition using refrigerator cameras or television cameras requires registration of family members on different visual devices. Multiple devices (i.e., multiple smart home devices) require multiple registrations, which is cumbersome and results in low user adoption. Furthermore, performing face registration and recognition independently on each device incurs high computational resource costs.

[0103] Furthermore, edge network data raises concerns about family privacy, making offline algorithms for processing data locally a growing trend. Visual processing requires computational resources, but the computational resources of each edge device are limited. The hardware configurations and processing performance of each edge device are inconsistent, resulting in severe fragmentation and an inability to share hardware processing resources.

[0104] This optional example provides a scheme for collaborative processing of distributed visual devices under the same network, namely, a scheme for face registration by multiple distributed multi-camera devices sharing resources. It utilizes the available resources on each end to perform face detection, key point detection, correction, feature extraction, etc., and adds the extracted features to the family member feature database. The feature databases are then distributed to each device. By performing collaborative processing of the hardware-intensive parts in the cloud, the computing resources required for face registration and face recognition can be reduced.

[0105] Combination Figure 3 In the feature data processing method of this optional example, the face registration process may include the following steps:

[0106] Step S302: The user enters facial information on any terminal device equipped with a visual sensor.

[0107] Users can register their facial recognition from any device equipped with a visual sensor, such as a mobile phone, TV, refrigerator, smart doorbell, or smart indoor surveillance camera. Registration can be done by recording facial information on the device using single or multiple images or videos.

[0108] Step S304: Process the facial information entered by the user to obtain user feature information.

[0109] The resource scheduling device performs computational resource scheduling based on the current hardware resources of the acquisition equipment, the current network quality of the equipment, and the resource bandwidth usage, selects the best computing resources, and further calculates and processes the image or video information or the processed feature information.

[0110] Step S306: Save the user feature information.

[0111] After the resource scheduling device selects the appropriate computing resources to complete the algorithm feature extraction, it sends the feature information to the home user feature database. Based on the ID (Identity document) information entered by the user, the corresponding features are stored.

[0112] Step S308: Distribute the saved user feature information to each terminal device equipped with a visual sensing device.

[0113] The updated home user feature database is distributed to various terminals equipped with visual sensing devices that have face comparison algorithm capabilities.

[0114] Through this optional example, devices on the same home network can share feature vectors after a single registration, allowing users to perform seamless identification on various devices within the home, thus improving the efficiency of information processing.

[0115] In one exemplary embodiment, after distributing the target feature library to the second device, the method further includes:

[0116] S51, when the second device obtains the second object image of the object to be verified, the second processing device extracts facial features from the second object image to obtain the second object features of the object to be verified. The second processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to perform facial recognition on the second object image.

[0117] S52, if the second object feature matches the first object feature in the target feature library, the target recognition result is sent to the second device, wherein the target recognition result is used to indicate that the object to be verified has been verified.

[0118] In this embodiment, after the target feature library is distributed to the second device, the second device can use the first object feature to verify the facial recognition features of the collected target object. Optionally, when the second device acquires a second object image of the object to be verified, it can extract facial features from the second object image using a second processing device to obtain the second object feature of the object to be verified. After obtaining the second object feature, it can determine whether the object to be verified is the target object based on the first object feature and the second object feature. If it is, it can be determined that the object to be verified has passed verification; otherwise, it can be determined that the object to be verified has failed verification. The second processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to perform facial recognition on the second object image.

[0119] Determining whether the object to be verified is the target object based on the first object feature and the second object feature can be done by determining whether the object to be verified is the target object based on the similarity between the first object feature and the second object feature. The similarity can be cosine similarity or other types of similarity, which is not limited in this embodiment.

[0120] The above-mentioned operation of extracting object features from the second object image can be performed by the second device, or it can be performed by the resource scheduling device in a manner similar to that used in the previous embodiment for extracting object features from the first object image, or jointly by the second device and other devices.

[0121] If the resource scheduling device determines whether the object to be verified is the target object based on the first object feature and the second object feature, the resource scheduling device can send the verification result to the second device. Optionally, if the second object feature matches the first object feature in the target feature library, the target identification result is sent to the second device, wherein the target identification result is used to indicate that the object to be verified has been verified.

[0122] For example, after the resource scheduling device completes the algorithm feature extraction using the selected computing resources, it can perform cosine distance calculation between the extracted object features and the object features stored in the local feature base library, complete the comparison, obtain the face recognition result, and send the face recognition result to the user's device.

[0123] This embodiment demonstrates how scheduling among a group of associated devices allows one or more devices to collaborate in object identification, thereby improving the efficiency of object identification.

[0124] The processing method for feature data in the embodiments of this application will be explained below with reference to optional examples. In this optional example, a group of associated devices is a group of vision devices in the same household, and the object feature of the target object is the facial feature of the target object.

[0125] Combination Figure 4 As shown, in the feature data processing method of this optional example, the face recognition process may include the following steps:

[0126] Step S402: When a user passes by any device end of the vision sensor, the device is triggered to identify the user.

[0127] Users can undergo facial recognition by passing through any device equipped with a visual sensor, such as a mobile phone, TV, refrigerator, smart doorbell, or smart indoor surveillance camera.

[0128] Step S404: Select the best computing resources to further process the image, video, or processed feature information.

[0129] The resource scheduling device performs computational resource scheduling based on the current hardware resources of the acquisition equipment, the current network quality of the equipment, and the resource bandwidth usage, selects the best computing resources, and performs further computational processing on the images, videos, or processed feature information.

[0130] Step S406: Compare the processed user feature information with the feature information stored in the database to obtain the face recognition result.

[0131] After the resource scheduling equipment selects the appropriate computing resources to complete the algorithm feature extraction, it performs cosine distance calculation with the local feature base library, completes the comparison, and obtains the face recognition result.

[0132] Step S408: Send the face recognition device results to the user's device.

[0133] This optional example demonstrates how multiple devices can work together to extract facial features and then perform facial recognition, thereby improving the efficiency of information processing.

[0134] In one exemplary embodiment, the second device is a detection device for detecting target events; after distributing the target feature library to the second device, the method further includes:

[0135] The second device is a detection device for the target event, which can be a sudden event in the home, such as an elderly person falling or a fire. The actual application scenarios of smart homes are complex and varied. For example, recognizing falls in the elderly requires video stream analysis and processing, transmitting massive amounts of video data to the cloud for analysis. This can lead to increased network latency due to the increased bandwidth load, as well as increased hardware computing costs associated with centralized cloud computing resources.

[0136] To address this, a group of interconnected devices can collaborate on real-time media data processing, offloading the computational tasks to these devices. For example, home security cameras are small and inexpensive with limited processing power. In scenarios involving real-time video stream computation, the massive amounts of data cannot be processed directly on the limited hardware resources of the device itself. Instead, the computational tasks of the video stream can be offloaded to nearby hardware resources, allowing various devices in the home to collaborate on processing, thereby improving the efficiency of video stream processing.

[0137] Optionally, in this embodiment, after distributing the target feature library to the second device, the above method further includes:

[0138] S61, when the second device acquires the data to be detected, the third processing device extracts the target event from the data to be detected and obtains the detection result of the target event. The third processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to detect the target event in the data to be detected.

[0139] S62, if the detection result of the target event indicates that the target event has occurred, an alarm message is sent to the target device, wherein the alarm message is used to alarm for the occurrence of the target event.

[0140] In this embodiment, the third processing device can acquire the data to be detected and extract the target event from the data to obtain the detection result of the target event. After obtaining the detection result of the target event, it can determine whether the target event has occurred based on the detection result. If so, it can send an alarm message to the target device to alert that the target event has occurred; otherwise, it can refrain from performing the alarm operation.

[0141] The above-mentioned operation of extracting target events from the data to be detected can be performed by a third processing device, or it can be performed by a resource scheduling device in a manner similar to the object feature extraction of the first object image or the second object image in the foregoing embodiments, or jointly by the third processing device and other devices.

[0142] Optionally, the resource scheduling device may select, from a group of associated devices, a device whose remaining computing resources, excluding the second device, are greater than or equal to the computing resources required for target event detection of the data to be detected, in a manner similar to the aforementioned selection of the second processing device.

[0143] Optionally, after selecting the third processing device, the resource scheduling device can extract target events from the data to be detected through the third processing device, or through the third processing device and other devices, to obtain the detection results of the target events. The method of extracting target events from the data to be detected is similar to the method of feature extraction from the first object image or the second object image in the aforementioned embodiments, and will not be described in detail here.

[0144] In this embodiment, by scheduling among a group of associated devices, one or more devices can collaboratively perform feature extraction operations on the data to be detected, thereby improving the efficiency of feature extraction and the timeliness of event alarms.

[0145] The processing method for feature data in the embodiments of this application will be explained below with reference to optional examples. In this optional example, a group of associated devices is a group of vision devices in the same household, and the object feature of the target object is the facial feature of the target object.

[0146] In this optional example, during a home inspection emergency, in the case of no sensor trigger, a distributed vision device can be used to collaboratively analyze and process real-time video to improve the timeliness of emergency detection.

[0147] Combination Figure 5 In the feature data processing method of this optional example, the process for detecting family emergencies may include the following steps:

[0148] Step S502: Real-time acquisition of images and videos using acquisition devices in the user's home.

[0149] Step S504: Process the images and videos acquired in real time by the acquisition device.

[0150] The resource scheduling equipment performs computational resource scheduling based on the current hardware resources of the acquisition equipment, the current network quality of the equipment, and the resource bandwidth usage, and selects the best computing resources (e.g., the equipment with the largest available computing power) to further process the video information or the processed information.

[0151] Step S506: When an emergency is identified from the real-time acquired images and videos, the emergency information is pushed to the response device.

[0152] This optional example demonstrates how home image / video data can be collaboratively processed on the edge device, reporting only detected incidents, thus protecting user privacy and improving data computation efficiency.

[0153] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0155] According to another aspect of the embodiments of this application, a feature data processing apparatus for implementing the above-described feature data processing method is also provided. Figure 6 This is a structural block diagram of an optional feature data processing apparatus according to an embodiment of this application, such as... Figure 6 As shown, the device may include:

[0156] The acquisition unit 602 is used to acquire the first object feature of the target object, wherein the first object feature is the face recognition feature of the target object recorded using the first device;

[0157] The execution unit 604, connected to the acquisition unit 602, is used to save the first object features into the target feature library corresponding to the target object, and to determine the second device to be distributed from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features;

[0158] The distribution unit 606, connected to the execution unit 604, is used to distribute the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

[0159] It should be noted that the acquisition unit 602 in this embodiment can be used to execute the above step S202, the execution unit 604 in this embodiment can be used to execute the above step S204, and the distribution unit 606 in this embodiment can be used to execute the above step S206.

[0160] Through the above modules, a first object feature of the target object is obtained, wherein the first object feature is the facial recognition feature of the target object recorded using a first device; the first object feature is saved to a target feature library corresponding to the target object, and a second device to be distributed is determined from a group of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features; the target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved on the second device as the recorded facial recognition feature of the target object. This solves the problem of low information processing efficiency caused by cumbersome processing operations in the related technology of device information processing methods, and improves the efficiency of device information processing.

[0161] In one exemplary embodiment, the acquisition unit includes:

[0162] The recognition module is used to perform face recognition on a first object image through a first processing device to obtain the features of the first object. The first object image is an image of the target object recorded by the first device, and the first processing device is a device whose remaining computing resources, other than the first device, are greater than or equal to the computing resources required to perform face recognition on the first object image.

[0163] In one exemplary embodiment, the above-described apparatus further includes:

[0164] The selection unit is configured to select a first processing device from multiple candidate devices based on the remaining computing resources of each candidate device before performing face recognition on the first object image through the first processing device, wherein the multiple candidate devices include at least one of the following: intelligent devices under the target object group where the target object is located, and intelligent devices within the target area range where the first device is located.

[0165] In one exemplary embodiment, the above-described apparatus further includes:

[0166] The first determining unit is configured to determine the remaining computing resources of each candidate device based on the device resource information of each candidate device before selecting the first processing device from the multiple candidate devices according to the remaining computing resources of each candidate device. The device resource information includes at least one of the following: available computing resources, the percentage of available computing resources used, network signal status, and available transmission bandwidth.

[0167] The second determining unit is used to determine the computing resources required for face recognition of the first object image based on the image attribute information of the first object image, wherein the image attribute information includes at least one of the following: image type and image size.

[0168] In one exemplary embodiment, the execution unit includes at least one of the following:

[0169] The selection module is used to select the associated device whose device permissions match the object category of the target object from a set of associated devices, and obtain the second device;

[0170] The first determining module is used to determine the recorded device in a group of associated devices as the second device, wherein the recorded device is a device that has stored the facial recognition features of the target object;

[0171] The second determining module is used to determine the authorized device in a group of associated devices as the second device, wherein the authorized device is a device that is authorized to share the facial recognition features of the target object;

[0172] The third determining module is used to determine the associated device to which the first device is to be distributed according to the device indication information and the first object feature in a group of associated devices as the second device.

[0173] The fourth determination module is used to identify the device that requests to obtain the facial recognition features of the target object from a group of associated devices as the second device.

[0174] In one exemplary embodiment, the above-described apparatus further includes:

[0175] The first extraction unit is used to extract facial features from the second object image by the second processing device after the target feature library is distributed to the second device and the second device obtains the second object image of the object to be verified, thereby obtaining the second object features of the object to be verified. The second processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to perform face recognition on the second object image.

[0176] The first sending unit is used to send the target recognition result to the second device when the second object feature matches the first object feature in the target feature library, wherein the target recognition result is used to indicate that the object to be verified has been verified.

[0177] In one exemplary embodiment, the second device is a detection device for detecting a target event; the above apparatus further includes:

[0178] The second extraction unit is used to extract target events from the target feature library after the target feature library is distributed to the second device, and when the second device obtains the data to be detected, the third processing device extracts the target events from the data to be detected to obtain the detection results of the target events. The third processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to detect the target events in the data to be detected.

[0179] The second sending unit is used to send alarm information to the target device when the detection result of the target event indicates that the target event has occurred, wherein the alarm information is used to alarm for the occurrence of the target event.

[0180] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of a device, can operate in environments such as... Figure 1 The hardware environment shown can be implemented through software or hardware, and the hardware environment includes the network environment.

[0181] According to another aspect of the embodiments of this application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute program code for processing the feature data of any of the above-mentioned embodiments of this application.

[0182] Optionally, in this embodiment, the storage medium may be located on at least one of the multiple network devices in the network shown in the above embodiment.

[0183] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps:

[0184] S1, obtain the first object feature of the target object, wherein the first object feature is the facial recognition feature of the target object recorded using the first device;

[0185] S2, save the first object features to the target feature library corresponding to the target object, and determine the second device to be distributed from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features;

[0186] S3, the target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

[0187] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated in this embodiment.

[0188] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0189] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described method for processing feature data is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0190] Figure 7 This is a structural block diagram of an optional electronic device according to an embodiment of this application, such as... Figure 7 As shown, it includes a processor 702, a communication interface 704, a memory 706, and a communication bus 708. The processor 702, communication interface 704, and memory 706 communicate with each other via the communication bus 708.

[0191] Memory 706 is used to store computer programs;

[0192] When processor 702 executes a computer program stored in memory 706, it performs the following steps:

[0193] S1, obtain the first object feature of the target object, wherein the first object feature is the facial recognition feature of the target object recorded using the first device;

[0194] S2, save the first object features to the target feature library corresponding to the target object, and determine the second device to be distributed from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features;

[0195] S3, the target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

[0196] Optionally, in this embodiment, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 7 The symbol is represented by a single thick line, but this does not indicate that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned electronic device and other devices.

[0197] The memory may include RAM, or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0198] As an example, the memory 706 described above may include, but is not limited to, the acquisition unit 602, the execution unit 604, and the distribution unit 606 in the feature data processing device. Furthermore, it may include, but is not limited to, other module units in the feature data processing device, which will not be elaborated upon in this example.

[0199] The processor mentioned above can be a general-purpose processor, including but not limited to: CPU, NP (Network Processor), etc.; it can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0200] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0201] Those skilled in the art will understand that Figure 7 The structure shown is for illustrative purposes only. The device that implements the above-mentioned feature data processing method can be a terminal device, such as a smartphone (e.g., Android phone, iOS phone), tablet computer, handheld computer, mobile internet device (MID), PAD, etc. Figure 7 This does not limit the structure of the aforementioned electronic device. For example, the electronic device may also include components that are more... Figure 7 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 7 The different configurations shown.

[0202] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, ROM, RAM, disk or optical disk, etc.

[0203] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0204] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0205] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0206] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0207] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the solution provided in this embodiment, depending on actual needs.

[0208] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0209] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for processing feature data, characterized in that, include: A first object feature of a target object is obtained, wherein the first object feature is a facial recognition feature of the target object recorded using a first device, wherein the first object feature is obtained by concatenating multiple first object sub-features, wherein the multiple first object sub-features are obtained by performing facial recognition on multiple object sub-images by multiple devices, wherein the multiple object sub-images are obtained by segmenting the first object image when there are multiple first processing devices, and there is a case where the remaining computing resources of the first processing devices are less than the computing resources required for facial recognition of the first object image, but the remaining computing resources of the multiple first processing devices are greater than the computing resources required for facial recognition of the first object image, wherein the first processing devices are the multiple devices, and wherein the first object image is the image of the target object recorded by the first device; The first object features are saved to the target feature library corresponding to the target object, and a second device to be distributed is determined from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features; The target feature library is distributed to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

2. The method according to claim 1, characterized in that, The acquisition of the first object feature of the target object includes: The first object image is subjected to face recognition by a first processing device to obtain the features of the first object. The first processing device is a device whose remaining computing resources, other than the first device, are greater than or equal to the computing resources required to perform face recognition on the first object image.

3. The method according to claim 2, characterized in that, Before performing face recognition on the first object image using the first processing device, the method further includes: The first processing device is selected from the plurality of candidate devices based on the remaining computing resources of each candidate device. The plurality of candidate devices include at least one of the following: intelligent devices under the target object group where the target object is located, and intelligent devices within the target area range where the first device is located.

4. The method according to claim 3, characterized in that, Before selecting the first processing device from the plurality of candidate devices based on the remaining computing resources of each candidate device, the method further includes: Based on the device resource information of each candidate device, the remaining computing resources of each candidate device are determined, wherein the device resource information includes at least one of the following: available computing resources, the percentage of available computing resources used, network signal status, and available transmission bandwidth. Based on the image attribute information of the first object image, the computational resources required for face recognition of the first object image are determined, wherein the image attribute information includes at least one of the following: image type and image size.

5. The method according to claim 1, characterized in that, Determining the second device to be distributed from a set of associated devices of the target object includes at least one of the following: selecting an associated device whose device permissions match the object category of the target object from the set of associated devices to obtain the second device; The device that has been registered in the group of associated devices is identified as the second device, wherein the device that has been registered is a device that has stored the facial recognition features of the target object; The authorized device in the group of associated devices is identified as the second device, wherein the authorized device is a device authorized to share the facial recognition features of the target object; Among the group of associated devices, the first device is identified as the associated device to which the first object feature is to be distributed, as indicated by the device indication information; The device that requests to obtain the facial recognition features of the target object from the group of associated devices is identified as the second device.

6. The method according to claim 1, characterized in that, After distributing the target feature library to the second device, the method further includes: When the second device acquires a second object image of the object to be verified, the second processing device extracts facial features from the second object image to obtain the second object features of the object to be verified. The second processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to perform facial recognition on the second object image. If the second object feature matches the first object feature in the target feature library, the target recognition result is sent to the second device, wherein the target recognition result is used to indicate that the object to be verified has been verified.

7. The method according to any one of claims 1 to 6, characterized in that, The second device is a detection device for detecting target events; after distributing the target feature library to the second device, the method further includes: When the second device acquires the data to be detected, the third processing device extracts the target event from the data to be detected to obtain the detection result of the target event. The third processing device is a device other than the second device whose remaining computing resources are greater than or equal to the computing resources required to detect the target event in the data to be detected. If the detection result of the target event indicates that the target event has occurred, an alarm message is sent to the target device, wherein the alarm message is used to alert that the target event has occurred.

8. A device for processing feature data, characterized in that, include: An acquisition unit is configured to acquire a first object feature of a target object, wherein the first object feature is a facial recognition feature of the target object recorded using a first device, wherein the first object feature is obtained by concatenating multiple first object sub-features, wherein the multiple first object sub-features are obtained by performing facial recognition on multiple object sub-images by multiple devices, wherein the multiple object sub-images are obtained by segmenting the first object image when there are multiple first processing devices, and there is a case where the remaining computing resources of the first processing devices are less than the computing resources required for facial recognition of the first object image, but the remaining computing resources of the multiple first processing devices are greater than the computing resources required for facial recognition of the first object image, wherein the first processing devices are the multiple devices, and wherein the first object image is an image of the target object recorded by the first device; An execution unit is configured to save the first object features into a target feature library corresponding to the target object, and determine a second device to be distributed from a set of associated devices of the target object, wherein the second device is a device that allows the recording of facial recognition features; A distribution unit is used to distribute the target feature library to the second device, wherein the first object feature in the target feature library is saved on the second device as the facial recognition feature of the target object.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method of any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method of any one of claims 1 to 7 through the computer program.

Images