CN115474000A

CN115474000A - Data processing method and device

Info

Publication number: CN115474000A
Application number: CN202210981202.3A
Authority: CN
Inventors: 郑丹丹; 何晓光
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2022-08-16
Filing date: 2022-08-16
Publication date: 2022-12-13

Abstract

The embodiment of the specification provides a data processing method and a device, wherein the data processing method is applied to a terminal and comprises the following steps: determining a data processing object to be acquired for the Internet of things equipment based on equipment parameters of the Internet of things equipment, wherein the Internet of things equipment is associated with the terminal; acquiring the data processing object from an object management terminal based on the object information of the data processing object to be acquired; and processing the data to be processed generated by the equipment of the Internet of things on the basis of the data processing object. Therefore, the problem that the terminal is incompatible with a larger data processing object package due to the fact that all data processing objects are packaged and sent to the terminal is solved, the corresponding data processing objects are deployed on the terminal for the Internet of things equipment, and the data processing objects are compatible with the terminal.

Description

Data processing method and device

Technical Field

The embodiment of the specification relates to the technical field of computers, in particular to a data processing method.

Background

With the rapid development of the technology of the internet of things, the types of the internet of things equipment are more and more abundant, and different types of data processing logics are required to be configured for the different types of the internet of things equipment, so that data generated by the internet of things equipment are processed; for example, when the internet of things device is a camera, a corresponding image processing algorithm needs to be configured for the camera, and when the internet of things device is a microphone, a corresponding voice processing algorithm needs to be configured for the microphone.

Because the types of the internet of things devices deployed on different terminals are different, in the prior art, various types of data processing logics are packaged into logic packets and are distributed to different terminals; and then selecting a data processing logic corresponding to the Internet of things equipment deployed on the terminal from the multiple types of data processing logic, and processing the data generated by the Internet of things equipment.

However, in the prior art, the logic package obtained after packaging is large, and many terminals cannot be compatible with the large logic package, so it is necessary to provide a method for deploying corresponding data processing logic for the internet of things device on the terminal, where the data processing logic is compatible with the terminal.

Disclosure of Invention

In view of this, the embodiments of the present specification provide a data processing method. One or more embodiments of the present specification also relate to a data processing apparatus, a computing device, a computer-readable storage medium, and a computer program, so as to solve the technical problems in the prior art.

According to a first aspect of the embodiments of the present specification, there is provided a data processing method applied to a terminal, including:

determining a data processing object to be acquired for the Internet of things equipment based on equipment parameters of the Internet of things equipment, wherein the Internet of things equipment is associated with the terminal;

acquiring the data processing object from an object management terminal based on the object information of the data processing object to be acquired;

and processing the data to be processed generated by the equipment of the Internet of things on the basis of the data processing object.

According to a second aspect of embodiments of the present specification, there is provided a data processing apparatus applied to a terminal, including:

the determining module is configured to determine a data processing object to be acquired for the Internet of things device based on device parameters of the Internet of things device, wherein the Internet of things device is associated with the terminal;

the acquisition module is configured to acquire the data processing object from an object management terminal based on the object information of the data processing object to be acquired;

and the processing module is configured to process the data to be processed generated by the Internet of things equipment based on the data processing object.

According to a third aspect of embodiments herein, there is provided a computing device comprising:

a memory and a processor;

the memory is used for storing computer-executable instructions, and the processor is used for executing the computer-executable instructions, and the computer-executable instructions realize the steps of the data processing method when being executed by the processor.

According to a fourth aspect of embodiments herein, there is provided a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the data processing method.

According to a fifth aspect of embodiments herein, there is provided a computer program, wherein the computer program, when executed in a computer, causes the computer to perform the steps of the data processing method of claim.

A data processing method provided in an embodiment of the present specification, applied to a terminal, includes: determining a data processing object to be acquired for the Internet of things equipment based on equipment parameters of the Internet of things equipment, wherein the Internet of things equipment is associated with the terminal; acquiring the data processing object from an object management terminal based on the object information of the data processing object to be acquired; and processing the data to be processed generated by the equipment of the Internet of things on the basis of the data processing object.

Specifically, the method accurately determines a data processing object to be acquired for the internet of things equipment based on equipment parameters of the internet of things equipment, acquires the data processing object from an object management end, and then processes data to be processed generated by the internet of things equipment by using the data processing object. Therefore, the problem that the terminal is incompatible with a larger data processing object package due to the fact that all data processing objects are packaged and sent to the terminal is solved, the corresponding data processing objects are deployed on the terminal for the Internet of things equipment, and the data processing objects are compatible with the terminal.

Drawings

Fig. 1 is a schematic application diagram of a data processing method provided in an embodiment of the present specification;

FIG. 2 is a flow chart of a data processing method provided by an embodiment of the present specification;

FIG. 3 is a flowchart illustrating a data processing method according to an embodiment of the present disclosure;

fig. 4 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present specification;

fig. 5 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, as those skilled in the art will be able to make and use the present disclosure without departing from the spirit and scope of the present disclosure.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if," as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination," depending on the context.

First, the noun terms to which one or more embodiments of the present specification relate are explained.

NPU: the NPU adopts a data-driven parallel computing architecture and is particularly good at processing massive multimedia data of videos and images and deep learning models.

3D: three-dimensional refers to a space system formed by adding a direction vector in a planar two-dimensional system. In the embodiment of the specification, the 3D refers to an image which is acquired by a 3D camera and has a depth; it should be noted that the image may be any type of image, for example, a depth image of a face captured by the 3D camera.

IR (Infrared Radiation): the infrared ray is short, in the embodiment of the present specification, the IR refers to a near infrared image, that is, an image acquired by an IR camera; it should be noted that the image may be any type of image, and in practical applications, the near-infrared image may be a near-infrared facial image.

Mode: the system refers to a multi-mode camera, and different images of RGB/IR/3D are acquired; it should be noted that the image may be any type of image, and in practical applications, the image may be a face image.

Size nucleus: a large Core/performance Core (P-Core) and a small Core/energy efficiency Core (E-Core) blending scheme.

GPU (graphics processing uni): a graphics processor.

2D camera: a camera capable of acquiring 2D images.

3D camera: a camera capable of acquiring 2D images.

An IR camera: an infrared camera refers to an infrared night vision camera device recorded by an infrared light technology.

MD5 (MD 5 Message-Digest Algorithm): message digest algorithm, a widely used cryptographic hash function.

With the rapid development of the technology of the internet of things, the types of the internet of things equipment are more and more abundant, and different types of data processing logics are required to be configured for the different types of the internet of things equipment, so that data generated by the internet of things equipment are processed; for example, taking internet of things (IoT) as an example, with the development of IoT, the types of image capturing devices supported by face-brushing payment/body-checking become more and more abundant;

the system comprises 2D cameras, 3D cameras, 2D + IR cameras, 2D cameras and 2D + IR +3D cameras. It should be noted that the image capturing device is a device which is disposed on the terminal and can capture an image, such as a camera; in practical application, one or more types of image acquisition devices can be deployed on one terminal.

Different cameras acquire images and correspond to different modal detection/identification algorithms of the full link, such as a 2D detection/key point/quality/identification algorithm, an IR detection/key point/quality/identification algorithm and a 3D detection/key point/quality/identification algorithm; on the other hand, different image acquisition devices have large computing force difference, including 8-core devices, 4-core devices, large-core devices, and low-computing-force devices with only 4 small cores, and including GPU devices and NPU devices, and different computing forces have different requirements on the size and the performance of the model. Therefore, only a small model is used as far as possible on a hardware device with only a small core, and a large model is used as far as possible on a large core or an NPU high-computation-power device; therefore, the calculation force is fully exerted, and meanwhile, the image recognition performance with higher cost performance is pursued.

Because different types of image acquisition devices require different models, many mechanisms provide models for image acquisition devices; the algorithm models corresponding to all types of image acquisition equipment are packaged, and the algorithm model package is used as an external delivery object and is sent to the terminal where the image acquisition equipment is deployed, so that the image processing requirements of the terminal where different image acquisition equipment is deployed are met.

However, for the requirement of the terminal (for example, android device) on the packet size, the terminal cannot be compatible with a larger algorithm model packet, so that all models cannot be put in one packet, and different models cannot be started according to the scene. To address this problem, one solution provided in this specification is to perform sub-channel packing according to differences between different hardware and models; that is, corresponding models are respectively determined for different manufacturers or different types of terminals, and the algorithm models are packaged and sent to different manufacturers or terminals. Thereby reducing the size of the algorithmic model package.

However, this solution is determined in that the deliverables are too fragmented for the ecological partners (different manufacturers or terminals); the configuration complexity of the ecological partners in the aspects of scene understanding, hardware understanding, package selection, link docking and the like is obviously improved.

Based on this, in the present specification, a data processing method is provided, and the present specification relates to a data processing apparatus, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

Fig. 1 is a schematic application diagram of a data processing method according to an embodiment of the present disclosure, and referring to fig. 1, a terminal in fig. 1 is capable of determining a corresponding model to be acquired for a camera based on device parameters of the camera, and requesting a cloud model pool for a model based on a model identifier of the model to be acquired; the cloud model pool issues a model to the terminal based on the request of the terminal, so that the terminal can obtain an image processing model for processing the image acquired by the camera; and then, the terminal inputs the image acquired by the terminal into an image processing model for processing.

The camera can be understood as internet of things equipment; the model to be acquired can be understood as a data processing object to be acquired; the image processing model can be understood as a data processing object; the cloud model pool can be understood as an object management side; the model identification can be understood as object information; the image collected by the camera can be understood as the data to be processed.

The data processing method provided by the embodiment of the specification provides a dynamic algorithm loading scheme with adaptive computing power and modal, scene hardware perception can be achieved, dynamic model loading and configuration can be automatically completed, the efficiency of ecological introduction is obviously improved, the difficulty of model configuration is reduced, a terminal can rapidly acquire a required model, and the data processing efficiency is improved.

Fig. 2 is a flowchart illustrating a data processing method according to an embodiment of the present specification, where the method is applied to a terminal, and specifically includes the following steps.

Step 202: and determining a data processing object to be acquired for the Internet of things equipment based on equipment parameters of the Internet of things equipment, wherein the Internet of things equipment is associated with the terminal.

The terminal may be understood as a user terminal or a server. The internet of things device includes, but is not limited to, an image capturing device, an audio capturing device, a detection device, and the like, which is not specifically limited in this specification.

The image acquisition device may be understood as a device for acquiring image data, and it should be noted that the audio acquisition device in the embodiment of the present specification may be any type of audio acquisition device, which is not specifically limited in the present specification; for example, the audio acquisition device may be a camera, a scanner, or the like; wherein, this camera can be any type of camera, for example, 2D camera, 3D camera, IR camera etc..

The audio acquisition device may be understood as a device for acquiring audio data, and it should be noted that the audio acquisition device in the embodiment of the present specification may be any type of audio acquisition device, which is not specifically limited in the present specification; for example, the audio capture device may be a microphone, sound pick-up, voice pen, or the like. The microphone may be any type of microphone, such as a condenser microphone, a moving coil microphone, and the like.

The detection device can be understood as a sensor device; it should be noted that the sensor device in the embodiments of the present specification may be any type of sensor device, and the type of the sensor device is not particularly limited in the present specification, and for example, the sensor device may be a temperature sensor, a humidity sensor, or a pressure sensor.

It should be noted that, in an embodiment of the present specification, associating the device of the internet of things with the terminal may be understood that the device of the internet of things is disposed on the terminal, for example, the terminal is a mobile phone, and the device of the internet of things and a camera disposed on the mobile phone. Or, in another embodiment of the present specification, the association between the internet of things device and the terminal may be understood as that the internet of things device and the terminal are connected to each other, and the internet of things device exists independently of the terminal, and data transmission can be performed between the connected internet of things device and the terminal; for example, the terminal is a computer, the internet of things device is a camera connected with the computer, and the camera can transmit collected images to the computer connected with the camera. It should be noted that the connection method between the internet of things device and the terminal may be wired connection or wireless connection.

The device parameter may be understood as a parameter characterizing a device condition of the internet of things device, including but not limited to a type parameter, a calculation force parameter, a performance parameter, and the like of the internet of things device. The type parameter may be understood as a parameter representing a type of the internet of things device, for example, a model of a camera, or a type of the camera (such as a 2D camera type, a 3D camera type, and the like), a type of a microphone, and the like.

The computing power parameter can be understood as a parameter representing the computing power of the internet of things equipment; including but not limited to computing unit information that the internet of things is configured with (e.g., information such as the number, model, etc. of the cameras configured as the CPU/GPU/NPU); the terminal is the information of a computing unit allocated to the internet of things device (for example, the mobile phone configures the number, model and the like of the CPU/GPU/NPU for the camera from the computing resources of the CPU/GPU/NPU and the like).

The performance parameter can be understood as a performance parameter that a data processing object matched with the camera needs to have. In practical application, different cameras have different collection efficiency, collected images and other factors; therefore, the performance of the data processing object configured for each camera also needs to be different. For example, the data processing object is an image processing model, based on which, when the quality of the image collected by the camera is high and the number of images is large, a model capable of rapidly processing the high-quality image needs to be configured for the camera, so that the data processing object is compatible with the camera and can rapidly process the image collected by the camera with high quality and large number.

In addition, in different application scenarios, the model matched with the camera also needs to have different performance parameters, for example, in a face-brushing payment scenario, since the accuracy of face recognition is required to be higher in the scenario, a model with higher accuracy of face image recognition needs to be matched for the camera. For another example, in an entrance guard scene, since the scene has a low requirement on accuracy of face recognition, but has a high requirement on efficiency of face recognition, a model with high efficiency of face image needs to be matched for the camera, so that the face image of the user acquired by the camera is quickly recognized.

The data processing object may be understood as an object capable of processing data to be processed generated by the internet of things device, and the data processing object includes, but is not limited to, an algorithm model, a script, a program, a driver or a device, and other tools capable of processing the data to be processed. That is, the data processing object may be a data processing model, a data processing program, a data processing script, a data processing driver, or a data processing device.

The data to be processed can be understood as data generated by the internet of things equipment; under the condition that the Internet of things equipment is different, the data to be processed are different; for example, in the case that the internet of things device is an image acquisition device, the data to be processed is image data (such as a face image, a two-dimensional code image, and the like) acquired by the image acquisition device; under the condition that the Internet of things equipment is audio acquisition equipment, the data to be processed is audio data acquired by the audio acquisition equipment; under the condition that the Internet of things equipment is a temperature sensor, the data to be processed is temperature data of the current environment acquired by the temperature sensor. That is, the data to be processed may be an image to be processed, audio to be processed, environmental data to be processed, or the like.

In addition, in the case that the data processing method provided by the present specification is applied to a face recognition scene, the internet of things device may be an image acquisition device; the data to be processed can be a face image acquired by an image acquisition device; the data processing object may be a face recognition model. The face recognition model can be used for carrying out face recognition on the face image acquired by the image acquisition equipment; thereby obtaining a face recognition result.

Taking the application of the data processing method provided in this specification in a face recognition scenario as an example, the determination of a data processing object to be acquired for an internet of things device is further described below. The Internet of things equipment is a camera, and the data processing object is a face recognition model; based on the above, the data processing method provided in this specification can determine a matching facial recognition model for the camera based on the camera type parameter, the computational power parameter, and the performance parameter of the camera.

In the embodiments provided in this specification, in order to quickly determine the device parameters, the internet of things device provides an interface for quickly acquiring the device parameters; in order to quickly determine the data processing object to be acquired, the terminal is locally configured with an object configuration unit, and can quickly determine the data processing object to be acquired for the internet of things device based on the object configuration unit, specifically, determining the data processing object to be acquired for the internet of things device based on the device parameter of the internet of things device includes:

acquiring equipment parameters of the Internet of things equipment based on a parameter acquisition interface associated with the Internet of things equipment;

and matching a data processing object to be acquired for the equipment of the Internet of things by using an object configuration unit based on the equipment parameters.

The parameter obtaining interface may be understood as an interface capable of obtaining the device parameter. It should be noted that when the internet of things device has multiple types of device parameters, the parameter obtaining interface also has multiple types, for example, the parameter obtaining interface includes a system interface and a camera interface; performance parameters configured for the camera in advance according to different scenes can be obtained through the system interface; the computational power parameter, the camera type parameter and the like of the camera can be obtained through the camera interface.

The object configuration unit may be understood as a program or script capable of determining a data processing object to be acquired for the internet of things device based on the device parameter; that is, the object configuration unit may be an object configuration program, an object configuration script.

Specifically, the data processing method provided in this specification can obtain the device parameters of the internet of things device through a parameter acquisition interface associated with the internet of things device, and match the data processing object to be obtained for the internet of things device by using an object configuration unit based on the device parameters.

In an embodiment provided in this specification, in view of the above-mentioned problem that terminals cannot be compatible with a larger algorithm model package due to packing algorithm models corresponding to all types of image acquisition devices, and the problems that deliveries are too fragmented and configuration complexity is high due to packing different hardware and models in different channels, the data processing method provided in this specification can provide a unified program package (i.e., a delivery) for all terminals, so that scene hardware perception is achieved through the unified delivery, dynamic loading and configuration are automatically completed, and efficiency of ecological introduction is significantly improved. The delivered program package may not contain an algorithm file (algorithm model), and only contains a file supporting dynamic configuration; also, the package size is uniform and small. The file supporting dynamic configuration can be understood as a module capable of automatically completing dynamic loading and configuration of the model according to different scene hardware (internet of things devices) in the program package. That is, the configuration module is dynamically configured. Specifically, in an embodiment provided in this specification, the obtaining, based on a parameter acquisition interface associated with an internet of things device, a device parameter of the internet of things device includes:

and determining a parameter acquisition interface associated with the Internet of things equipment through an object management module running on the terminal, and acquiring the equipment parameters of the Internet of things equipment through the parameter acquisition interface.

The object management module can be understood as a module which can determine a data processing object to be acquired for the internet of things equipment according to the equipment parameters of the internet of things equipment and can acquire the data processing object from an object management end; such as the above-mentioned program package.

That is to say, the data processing method provided in this specification may determine, by using the object management module, a data processing object to be acquired for the internet of things device based on the device parameter of the internet of things device, and acquire, by using the object management module, the data processing object from the object management side based on object information of the data processing object to be acquired.

It should be noted that the object management module may download the data processing object from the object management terminal to the terminal. And then, the object management module processes the data to be processed generated by the equipment of the Internet of things based on the data processing object. That is, the data processing method provided in this specification may be applied to an object management module running on a terminal.

Specifically, a parameter acquisition interface associated with the internet of things device is determined through an object management module running on the terminal, and device parameters of the internet of things device are acquired through the parameter acquisition interface.

Further, in order to implement scene hardware perception through the unified program package and automatically complete dynamic loading and configuration of the model, the terminal needs to acquire and operate the program package, and specifically, before determining a data processing object to be acquired for the internet of things device, the method further includes:

and acquiring an object management module provided by the object management terminal for the terminal, and operating the object management module.

The object management module provided by the object management terminal for the terminal is obtained, and the object management module sent by the object management terminal is received by the terminal; or, the terminal requests the object management module from the object management terminal, specifically, the terminal sends a module acquisition request to the object management terminal, and the object management terminal determines, in response to the module acquisition request, an object management module corresponding to the module identifier carried in the module acquisition request from the terminal and returns the object management module to the terminal. The module identifier may be understood as information uniquely identifying one object management module, and in the case where the object management module is a package, the module identifier may be understood as a name of the package.

The object management side may be understood as a server side that performs issuing management on the data processing object or the object management module, that is, the object management side may be a server side, for example, the object management side may be a model resource pool.

Specifically, the terminal can be obtained from an object management terminal, and the object management terminal is an object management module provided for terminals including different types of internet of things devices. Then, the terminal can run the object management module after acquiring the object management module.

In the above example, the terminal requests the model resource pool to acquire the program package, and after acquiring the program package issued by the model resource pool, locally runs the program package.

Further, in an embodiment provided in this specification, the terminal locally has a model configuration library file, and model configuration information matched with all types of internet of things devices is stored in the file, so that, based on the device information of the internet of things devices, a corresponding model can be quickly matched for the internet of things devices from the model configuration library file. Specifically, the matching of the data processing object to be acquired for the internet of things device from the object configuration unit based on the device parameter includes:

determining target object configuration parameters matched with the equipment parameters from object configuration parameters contained in an object configuration unit;

and determining the data processing object corresponding to the target object configuration parameter as the data processing object to be acquired corresponding to the Internet of things equipment.

The object configuration unit may also be understood as a unit storing object configuration parameters, and it should be noted that the unit may be a file, for example, a model configuration library file, and the file may be stored in a cache or a register of the terminal; or in a storage area partitioned by the terminal from memory or a local disk.

It should be noted that the object configuration unit may belong to an object management module, that is, in the embodiment provided in this specification, the object configuration unit is included in the object management module.

The object configuration parameters may be understood as parameters characterizing the configuration of different types of data processing objects. That is, in the case where the data processing object is a data processing model, the object configuration parameters may be model configuration parameters, such as performance parameters of the data processing object, types of data that the data processing object can process, and the like. The data type that the data processing object can process can be understood as that the model A can process a 2D image collected by a 2D camera; the model B can process 3D images acquired by the 3D camera, and so on.

According to the above example, the internet of things device may be a 3D camera, the data processing object is an image processing model, and the object configuration unit is a model configuration library file, based on which, the terminal or a program running on the terminal can dynamically configure a model for the 3D camera based on the device information of the 3D camera in a model dynamic configuration manner. Specifically, the terminal or a program running on the terminal can match corresponding target object configuration parameters for the 3D camera from a model configuration library file based on the device information of the 3D camera, and determine a model corresponding to the target object configuration parameters as a model matched with the 3D camera.

To facilitate further explanation of the dynamic configuration of the model, the manner in which the model is matched for a camera based on the camera's device information is shown in table 1 below. The table 1 may be a model configuration file. The specific definition is as follows.

TABLE 1

The "camera" row in table 1 is used to indicate a type of camera that can be deployed on one terminal; the row of "calculation type and calculation size" in table 1 is used to indicate the type of the calculation unit that the camera is deployed on one terminal, and the number of the calculation units; the "performance parameters" row in table 1 is used to indicate the performance parameters of the model that is matched to each camera; the "model file" row in table 1 is used to indicate the model type that matches each camera. It should be noted that one camera may be matched to one or more models; the images acquired by the camera are then identified by one or more models. For example, "2D detection/recognition model combination 1" in table 1 is used to indicate that the model matched for the 2D camera includes a 2D detection model and a 2D recognition model.

Step 204: and acquiring the data processing object from an object management terminal based on the object information of the data processing object to be acquired.

The object information may be understood as identification information of the data processing object to be obtained, and in the case that the data processing object is a data processing model, model identification information of the identification information, for example, a name of the model, a number of the model, and the like.

Alternatively, the object information may be an object configuration parameter of the data processing object to be acquired.

Subsequently, the object management terminal can send the data processing object corresponding to the identification information to the terminal; alternatively, the object management module may send the data processing object whose object configuration parameter is consistent to the terminal.

Specifically, the acquiring the data processing object from an object management side based on the object information of the data processing object to be acquired includes:

generating an object acquisition request based on the object information of the data processing object to be acquired, and sending the object acquisition request to an object management terminal;

and receiving a data processing object which is sent by the object management end in response to the object acquisition request and corresponds to the object information.

Along with the above example, the object information is model identification information. Based on the method, the terminal or a program running on the terminal can generate a model acquisition request based on the model identification information A1 of the facial recognition model A to be acquired, and the model acquisition request is sent to the cloud model pool; the cloud model pool responds to the received model acquisition request, determines a face recognition model A corresponding to the model identification information A1 from a plurality of models stored in the cloud model pool, and sends the face recognition model A to the terminal, so that the terminal acquires the face recognition model A.

Step 206: and processing the data to be processed generated by the equipment of the Internet of things on the basis of the data processing object.

Specifically, the terminal can process the data to be processed generated by the internet of things equipment based on the data processing object; and obtaining a processing result.

Along with the above example, the terminal or a program running on the terminal can process the face image collected by the camera through the face recognition model to obtain the face recognition result.

Further, in the embodiments provided in this specification, in order to ensure the accuracy of the received data processing object and avoid using an incorrect data processing object to process the data to be processed, the terminal needs to verify the received data processing object. Specifically, the processing the to-be-processed data generated by the internet of things device based on the data processing object includes:

verifying the data processing object based on object verification information of the data processing object, wherein the object verification information is information provided by the object management terminal;

and under the condition that the verification is determined to pass, processing the to-be-processed data generated by the equipment of the Internet of things on the basis of the data processing object.

The object verification information may be understood as information for verifying correctness of the data processing object, for example, summary information of the data processing object.

Specifically, after the data processing object is acquired from the object management side, the object verification information corresponding to the data processing object issued by the object management side can be determined. And then, checking the correctness of the data processing object based on the object checking information, and processing the to-be-processed data generated by the Internet of things equipment based on the data processing object under the condition that the checking is passed.

In an embodiment provided in this specification, in order to ensure correctness of the data processing object, the object verification information of the data processing object may be compared with locally stored local object verification information, and if the two pieces of object verification information are consistent, it is determined that the data processing object is correct. Specifically, the verifying the data processing object based on the object verification information of the data processing object includes:

determining object verification information of the data processing object and local object verification information corresponding to the data processing object to be acquired;

and carrying out correctness verification on the object verification information based on the local object verification information.

The local object verification information may be understood as information that is locally stored and is verified with respect to the data processing object. For example, the local object verification information is locally stored summary information. In practical application, in the process of determining the data processing object to be acquired for the internet of things device, the summary information corresponding to the data processing object to be acquired is determined, and the summary information is used as the information for performing correctness check on the acquired data processing object subsequently. It should be noted that the summary information may be obtained by any summary calculation method, and this specification is not limited in this respect. The digest is computed, for example, by the MD5 digest algorithm.

Specifically, in order to ensure the correctness of the data processing object, the terminal or the object management model needs to determine the object verification information of the data processing object and the local object verification information corresponding to the data processing object to be acquired, perform consistency matching based on the local object verification information and the object verification information, and determine that the data processing object is correct when the matching result is consistent.

For example, the object verification information is summary information, and based on the summary information, the summary information corresponding to the face recognition model and sent by the cloud resource pool can be received while the face recognition model is acquired from the cloud model pool; and then, determining local abstract information corresponding to the face recognition model to be acquired from the local, then, carrying out consistency comparison on the two abstract information, and determining that the face recognition model issued by the cloud resource pool is a correct model under the condition that the comparison result is consistent.

In implementations provided herein, the data processing objects are at least two;

correspondingly, the processing the data to be processed generated by the internet of things device based on the data processing object comprises:

receiving a data processing request, wherein the data processing request carries to-be-processed data generated by the Internet of things equipment and a data type corresponding to the to-be-processed data;

arranging at least two data processing objects based on the object processing strategy corresponding to the data type to obtain a data processing object link;

and processing the data to be processed based on the data processing object link.

The data processing link may be understood as a link formed by a plurality of data processing objects and used for processing the data to be processed.

The data type may be an application scene corresponding to the to-be-processed data, for example, the to-be-processed image is applied to a face payment scene, an entrance guard scene, or the like. Different models are required to process the image to be processed in different combination modes in different application scenes.

The object processing policy may be understood as a policy for arranging the data processing object, and it should be noted that the object processing policy is a preset limit, and each data type (i.e. different scenes) may correspond to a different object processing policy.

The data processing request may be understood as a request for processing the data to be processed, for example, a face recognition request for performing a recognition process on a face image, or a voice conversion request for converting audio to be processed into text.

For example, the object processing policy is a preset arrangement policy for the models, and the data processing objects are a face detection model, a face recognition model a, and a face recognition model B. Based on this, in the case where a face recognition request is received, a face image carried in the face recognition request and a scene to which the face image is applied can be acquired.

In the case that the scene is a face payment scene, considering that the scene has high accuracy in face recognition, the face detection model, the face recognition model a and the face recognition model B are arranged as the face detection model → the face recognition model a → the face recognition model B according to a preset arrangement strategy corresponding to the scene; the "face detection model → face recognition model a → face recognition model B" can be understood as an algorithmic link. The facial image is processed according to the sequence based on the three models, and the accuracy of facial recognition is ensured through the facial recognition model A and the facial recognition model B.

Under the condition that the scene is an entrance guard scene, considering that the scene has high efficiency of face identification, selecting a face detection model and a face identification model A from the three models according to a preset arrangement strategy corresponding to the scene, and arranging the face detection model and the face identification model A into a face detection model → a face identification model A; the "face detection model → face recognition model a" can be understood as an algorithmic link. The facial images are processed according to the sequence based on the two models, and the facial images are processed only through the facial recognition model A, so that the facial recognition efficiency is improved.

In an embodiment provided in this specification, the internet of things device is an image acquisition device, and the data processing object is an image processing model;

and processing the image to be processed acquired by the image acquisition equipment based on the image processing model.

For the explanation of this embodiment, reference may be made to the corresponding or corresponding contents in the above-described embodiments.

In a first embodiment provided in this specification, the number of the image capturing devices is at least two, and the at least two image capturing devices correspond to different image processing models respectively;

and processing the images to be processed acquired by the at least two image acquisition devices based on the image processing models corresponding to the at least two image acquisition devices.

The data processing method provided by the specification accurately determines a data processing object to be acquired for the internet of things device based on the device parameter of the specific internet of things device, acquires the data processing object from an object management terminal, and then processes the data to be processed generated by the internet of things device by using the data processing object. Therefore, the problem that the terminal is incompatible with a larger data processing object package due to the fact that all data processing objects are packaged and sent to the terminal is solved, the corresponding data processing objects are deployed on the terminal for the Internet of things equipment, and the data processing objects are compatible with the terminal.

The following description will further describe the data processing method provided in this specification by taking an application of the data processing method in a face recognition scenario as an example with reference to fig. 3. Fig. 3 shows a processing procedure flowchart of a data processing method provided in an embodiment of the present specification, which specifically includes the following steps.

Step 302: device parameter sensing.

Specifically, a dynamic configuration module in the image processing program has a device sensing capability, and the image processing program (i.e., the client) can acquire the calculation parameters of the camera device and the modalities supported by the camera, that is, the camera parameters, through the system interface and the camera interface.

The image processing program may be understood as the object management model described above. The image processing program runs on the terminal, is uniform and small in type, and can be provided for all types of terminals, so that the work of automatically completing model configuration and loading is realized based on the image processing program.

Step 304: matching the model configuration library file.

Specifically, after obtaining the device parameters, the dynamic configuration module in the image processing program can match the device parameters, such as the camera parameters, the computational power parameters, and the performance parameters of the combined scene, with the model configuration parameters contained in the model configuration library file, so as to maximally match an excellent model combination for the camera.

It should be noted that the model may be a face recognition model.

Step 306: and generating a model configuration file.

The model configuration file may refer to table 1 above.

Specifically, in the process of matching the model for the camera based on the device parameter, the dynamic configuration module in the image processing program generates a model configuration file, and the model configuration file records the device parameter of the camera and the corresponding relationship between the model matched with the camera.

And then, the dynamic configuration module generates a model acquisition request based on the model identification of the model matched with the camera, and uploads the model acquisition request to the cloud model pool.

Step 308: and (5) issuing the model.

Specifically, the cloud model pool responds to a model acquisition request sent by the dynamic configuration module, and determines a model corresponding to a model identifier carried in the model acquisition request from a locally held model; and the model is sent to an image processing program through a big data channel.

Step 310: and pulling the local model.

Specifically, the image processing program receives the model issued by the cloud model pool. The local model can be understood as a model which is issued to the local terminal by the cloud model pool.

Step 312: and (5) checking the correctness of the model.

Specifically, after the image processing program acquires the model, in order to ensure the correctness of the model, the correctness of the model is checked. Specifically, the image processing program determines an abstract which is sent by a cloud resource pool and corresponds to the received model; then, the abstract is compared with the locally stored local abstract in a consistent manner; in the case where the two are identical, the model is determined to be the correct model.

It should be noted that the local digest is a digest of a model matched by the camera from the model configuration library file by the image processing program.

Step 314: and loading the local model.

Specifically, in the case where the model correctness check passes, the image processing program loads the local model.

Step 316: and constructing an algorithm link.

And selecting and executing a correct model construction algorithm link according to the model configuration and the camera input.

Specifically, after acquiring a model matched with a camera, an image processing program can receive an image processing request, where the image processing request carries a user face image acquired by the camera and an application scene (such as an access control scene, a face-brushing payment scene, and the like) corresponding to the face image. The number of models matched with the camera can be multiple.

Then, the image processing program selects a corresponding model from the plurality of face recognition models according to different scenes, and constructs the model as a model algorithm link. And then, inputting the face image acquired by the camera into a model algorithm link to obtain a face recognition result.

The data processing method in the specification provides a dynamic algorithm loading scheme with adaptive computing power and modal, deliveries can be effectively unified, scene hardware perception is achieved through the algorithm, dynamic loading and configuration are automatically completed, and efficiency of ecological introduction is remarkably improved. The scheme uses a uniform delivery package (i.e. program) and configuration on different hardware and devices, and after the program is started, different models are downloaded in an internet mode according to the difference of the hardware.

Corresponding to the above method embodiment, this specification further provides a data processing apparatus embodiment, and fig. 3 shows a schematic structural diagram of a data processing apparatus provided in an embodiment of this specification. As shown in fig. 3, the apparatus is applied to a terminal, and includes:

a determining module 402, configured to determine a data processing object to be acquired for an internet of things device based on a device parameter of the internet of things device, where the internet of things device is associated with the terminal;

an obtaining module 404, configured to obtain the data processing object from an object management side based on the object information of the data processing object to be obtained;

the processing module 406 is configured to process the to-be-processed data generated by the internet of things device based on the data processing object.

Optionally, the determining module 402 is further configured to:

Optionally, the data processing apparatus further includes an execution module configured to:

Optionally, the determining module 402 is further configured to:

Optionally, the obtaining module 404 is further configured to:

Optionally, the processing module 406 is further configured to:

and under the condition that the verification is determined to pass, processing the to-be-processed data generated by the Internet of things equipment based on the data processing object.

Optionally, the processing module 406 is further configured to:

Optionally, the number of the data processing objects is at least two;

accordingly, optionally, the processing module 406 is further configured to:

Optionally, the internet of things device is an image acquisition device, and the data processing object is an image processing model;

accordingly, optionally, the processing module 406 is further configured to:

Optionally, the number of the image acquisition devices is at least two, and the at least two image acquisition devices respectively correspond to different image processing models;

accordingly, optionally, the processing module 406 is further configured to:

The data processing apparatus provided in this specification accurately determines a data processing object to be acquired for an internet of things device based on device parameters of the specific internet of things device, acquires the data processing object from an object management terminal, and then processes data to be processed generated by the internet of things device using the data processing object. Therefore, the problem that the terminal is incompatible with a larger data processing object package due to the fact that all data processing objects are packaged and sent to the terminal is solved, the corresponding data processing objects are deployed on the terminal for the Internet of things equipment, and the data processing objects are compatible with the terminal.

The above is a schematic configuration of a data processing apparatus of the present embodiment. It should be noted that the technical solution of the data processing apparatus and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the data processing apparatus can be referred to the description of the technical solution of the data processing method.

FIG. 5 illustrates a block diagram of a computing device 500 provided in accordance with one embodiment of the present description. The components of the computing device 500 include, but are not limited to, a memory 510 and a processor 520. Processor 520 is coupled to memory 510 via bus 530, and database 550 is used to store data.

Computing device 500 also includes access device 540, access device 540 enabling computing device 500 to communicate via one or more networks 560. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. The access device 540 may include one or more of any type of network interface, e.g., a Network Interface Card (NIC), wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above-described components of computing device 500, as well as other components not shown in FIG. 5, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device architecture shown in FIG. 5 is for purposes of example only and is not limiting as to the scope of the present description. Those skilled in the art may add or replace other components as desired.

Computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), mobile phone (e.g., smartphone), wearable computing device (e.g., smartwatch, smartglasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 500 may also be a mobile or stationary server.

Wherein the processor 520 is configured to execute computer-executable instructions that, when executed by the processor, implement the steps of the data processing method described above.

The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the data processing method.

An embodiment of the present specification also provides a computer-readable storage medium storing computer-executable instructions, which when executed by a processor, implement the steps of the data processing method described above.

The above is an illustrative scheme of a computer-readable storage medium of the embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the data processing method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the data processing method.

An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the data processing method.

The above is a schematic scheme of a computer program of the present embodiment. It should be noted that the technical solution of the computer program and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the computer program can be referred to the description of the technical solution of the data processing method.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of acts, but those skilled in the art should understand that the present embodiment is not limited by the described acts, because some steps may be performed in other sequences or simultaneously according to the present embodiment. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.

Claims

1. A data processing method is applied to a terminal and comprises the following steps:

2. The data processing method of claim 1, wherein the determining a data processing object to be acquired for the internet of things device based on the device parameter of the internet of things device comprises:

3. The data processing method of claim 2, wherein obtaining the device parameters of the internet of things device based on a parameter acquisition interface associated with the internet of things device comprises:

and determining a parameter acquisition interface associated with the equipment of the Internet of things through an object management module running on the terminal, and acquiring the equipment parameters of the equipment of the Internet of things through the parameter acquisition interface.

4. The data processing method of claim 3, wherein before determining the data processing object to be acquired for the internet of things device based on the device parameter of the internet of things device, the method further comprises:

5. The data processing method according to claim 2, wherein the matching of the data processing object to be acquired for the internet of things device by using an object configuration unit based on the device parameter comprises:

6. The data processing method according to claim 1, wherein the acquiring the data processing object from an object management side based on the object information of the data processing object to be acquired comprises:

7. The data processing method of claim 1, wherein the processing the to-be-processed data generated by the internet of things device based on the data processing object comprises:

8. The data processing method of claim 7, wherein the verifying the data processing object based on the object verification information of the data processing object comprises:

9. The data processing method according to claim 1 or 7, the data processing objects being at least two;

10. The data processing method according to claim 1 or 7, wherein the internet of things device is an image acquisition device, and the data processing object is an image processing model;

11. The data processing method according to claim 10, wherein the image capturing devices are at least two types, and the at least two types of image capturing devices respectively correspond to different image processing models;

12. A data processing device is applied to a terminal and comprises:

13. A computing device, comprising:

a memory and a processor;

the memory is for storing computer-executable instructions, and the processor is for executing the computer-executable instructions, which when executed by the processor, implement the steps of the data processing method of any one of claims 1 to 11.

14. A computer-readable storage medium storing computer-executable instructions which, when executed by a processor, implement the steps of the data processing method of any one of claims 1 to 11.