CN111221803B - Feature library management method and coprocessor - Google Patents
Feature library management method and coprocessor Download PDFInfo
- Publication number
- CN111221803B CN111221803B CN201911381948.5A CN201911381948A CN111221803B CN 111221803 B CN111221803 B CN 111221803B CN 201911381948 A CN201911381948 A CN 201911381948A CN 111221803 B CN111221803 B CN 111221803B
- Authority
- CN
- China
- Prior art keywords
- feature
- characteristic value
- network data
- feature library
- library
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007726 management method Methods 0.000 title claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000004590 computer program Methods 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000013478 data encryption standard Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 210000001508 eye Anatomy 0.000 description 1
- 210000004709 eyebrow Anatomy 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000001331 nose Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000003800 pharynx Anatomy 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The application is applicable to the technical field of data processing, and provides a feature library management method and a coprocessor, wherein the method comprises the following steps: receiving a binding instruction sent by a main processor, identifying information in the binding instruction, releasing the original binding relation of the network data source according to the binding instruction, establishing a binding relation between a designated network data source and at least one feature library or feature table to be bound, receiving network data from the network data source, and analyzing the network data in the feature library or feature table establishing the binding relation with the network data source to obtain a corresponding feature value. According to the method and the device, the feature library is stored in the storage medium of the coprocessor, the time required for reading the feature library is shortened, the upgrading cost of equipment is reduced, and meanwhile, the coprocessor can identify the feature values of various different scenes simultaneously by dynamically binding the appointed network data source with the appointed feature library or the appointed feature tables, so that the identification efficiency is improved, and the storage requirement is reduced.
Description
Technical Field
The application belongs to the technical field of data processing, and particularly relates to a feature library management method and a coprocessor.
Background
With the rapid development of artificial intelligence technology, the field of artificial intelligence application is also increasing, wherein AI chips are the key points in the field of artificial intelligence technology.
In the existing data management method based on artificial intelligence co-processing, a feature library is mainly stored in a permanent storage medium of a main processor or obtained from a network.
In practice, obtaining a feature library from a network wastes a lot of time. And when the residual capacity of the permanent storage medium of the main processor is too small, the storage requirement of the large-capacity feature library cannot be met, and the hardware of the main processor needs to be replaced, so that the equipment cost is increased.
Disclosure of Invention
The embodiment of the application provides a feature library management method and device, which can solve the problems of waste of a large amount of time and higher equipment cost when a feature library is acquired in the prior art.
In a first aspect, an embodiment of the present application provides a feature library management method, including:
receiving a binding instruction sent by a main processor;
identifying at least one feature library to be bound or at least one feature table to be bound and a designated network data source contained in the binding instruction; the feature library is stored in a collaborative data processing library, and at least one feature table is stored in each feature library;
according to the binding instruction, the original binding relation of the network data source appointed by the main processor is released;
establishing a binding relation between the designated network data source and at least one feature library to be bound or at least one feature table to be bound;
receiving network data from the designated network data source sent by the main processor;
and analyzing the network data in a feature library or a feature table which establishes a binding relation with the appointed network data source to obtain a feature value corresponding to the network data.
In a possible implementation manner of the first aspect, the method further includes:
encrypting the return result, and sending the encrypted return result to the main processor; the returned result comprises a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction, an updating result, a deleting result, a first query result and a second query result.
In a second aspect, embodiments of the present application provide a coprocessor, including:
the first receiving module is used for receiving the binding instruction sent by the main processor.
The identification module is used for identifying at least one feature library to be bound or at least one feature table to be bound and a designated network data source contained in the binding instruction; the feature library is stored in a collaborative data processing library, and at least one feature table is stored in each feature library;
the releasing module is used for releasing the original binding relation of the network data source appointed by the main processor according to the binding instruction;
the first establishing module is used for establishing a binding relation between the appointed network data source and at least one feature library to be bound or at least one feature table to be bound;
a second receiving module, configured to receive network data from the specified network data source sent by the main processor;
and the analysis module is used for analyzing the network data in a feature library or a feature table which establishes a binding relation with the appointed network data source to obtain a feature value corresponding to the network data.
In a third aspect, an embodiment of the present application provides a co-data processor system, including a main processor and at least one coprocessor, where each coprocessor is correspondingly mounted with a storage medium, where a computer program capable of running on the corresponding coprocessor is stored in the storage medium, and when the coprocessor executes the computer program, the feature library management method according to any one of the first aspect is implemented.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium storing a computer program which, when executed by a processor, implements the feature library management method according to any one of the first aspects.
In a fifth aspect, embodiments of the present application provide a computer program product, which when run on a terminal device, causes the terminal device to perform the feature library management method of any one of the first aspects above.
According to the embodiment of the application, the feature library is stored in the storage medium of the coprocessor, the time required for reading the feature library is shortened, the upgrading cost of equipment is reduced, and meanwhile, the coprocessor can identify the feature values of various different scenes simultaneously by dynamically binding the appointed network data source with the appointed feature libraries or the appointed feature tables, so that the identification efficiency is improved, and the storage requirement is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required for the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a feature library management method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a co-processing system according to an embodiment of the present application;
FIG. 3 is a schematic diagram of an application scenario of multi-library multi-table dynamic binding based on a feature library management method according to an embodiment of the present application;
FIG. 4 is a flowchart of a feature library management method according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a coprocessor according to one embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As used in this specification and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
The feature library management method provided by the embodiment of the application can be applied to terminal devices such as coprocessors, mobile phones including coprocessors, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (augmented reality, AR)/Virtual Reality (VR) devices, notebook computers, ultra-mobile personal computer (UMPC), netbooks, personal digital assistants (personal digital assistant, PDA) and the like, and the specific types of the terminal devices are not limited.
For example, the terminal device may be a Station (ST) in a WLAN, which may be a cellular telephone, a cordless telephone, a session initiation protocol (Session Initiation Protocol, SIP) phone, a wireless local loop (Wireless Local Loop, WLL) station, a personal digital assistant (Personal Digital Assistant, PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an on-board device, an internet of vehicle terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite radio device, a wireless modem card, a television Set Top Box (STB), a customer premise equipment (customer premise equipment, CPE) and/or other devices for communicating over a wireless system as well as next generation communication systems, such as a mobile terminal in a 5G network or a mobile terminal in a future evolved public land mobile network (Public Land Mobile Network, PLMN) network, etc.
By way of example, but not limitation, when the terminal device is a wearable device, the wearable device may also be a generic name for applying wearable technology to intelligently design daily wear, developing wearable devices, such as glasses, gloves, watches, apparel, shoes, and the like. The wearable device is a portable device that is worn directly on the body or integrated into the clothing or accessories of the user. The wearable device is not only a hardware device, but also can realize a powerful function through software support, data interaction and cloud interaction. The generalized wearable intelligent device comprises full functions, large size, and complete or partial functions which can be realized independent of a smart phone, such as a smart watch or a smart glasses, and is only focused on certain application functions, and needs to be matched with other devices such as the smart phone for use, such as various smart bracelets, smart jewelry and the like for physical sign monitoring.
The embodiment of the invention provides a feature library management method which can be applied to coprocessors in a coprocessors data processing system, and can be realized by the coprocessors when a computer program is run.
Referring to FIG. 1, a feature library management method according to the present embodiment may be applied to a coprocessor 12 in a data processing system 1 in FIG. 2;
FIG. 2 is a schematic diagram schematically showing a configuration of a cooperative data processing system 1; the co-data processing system 1 comprises a main processor 11, a co-processor 12 and a storage medium 13 on which the co-processor 12 is mounted.
A main processor 11 for performing data processing (mainly referred to as artificial intelligence acceleration operation); wherein the data comprises a characteristic value.
A coprocessor 12, which may be a chip for assisting the host processor in performing artificial intelligence (Artificial Intelligence, AI) acceleration operations, for assisting the host processor in data processing.
And the storage medium is used for establishing and storing a coprocessing database in the coprocessor.
The feature library management method comprises the following steps:
s101, receiving a binding instruction sent by a main processor.
In a specific application, the binding instruction refers to an instruction for controlling the coprocessor to bind a specified network data source with a specified one or more feature libraries (or with a specified one or more feature tables). Feature library refers to a collection of target features (i.e., feature values) used to annotate in the identification of network data, which is built into a co-processing database. The co-processing database is a data processing database established in a storage medium on which the co-processor is mounted.
The network data source refers to a source address from which network data is acquired. The characteristic value can be obtained by acquiring network data in the network data source and analyzing the network data. For example, if a certain IP address in the network is designated as a network data source by the binding instruction, when the network data from the IP address sent by the host processor is acquired, the network data is transmitted to a binding database or a data table for identification, and a characteristic value is obtained.
By way of example, and not limitation, the above-described host processor may be a central processing unit (Central Processing Unit, CPU), which may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
S102, identifying at least one feature library to be bound or at least one feature table to be bound and a designated network data source contained in the binding instruction; the feature library is stored in a co-data processing library, and at least one feature table is stored in each feature library.
In a specific application, the binding instruction includes, but is not limited to, an address of a specified network data source and a name of a specified at least one feature library to be bound, or a name of a specified at least one feature table to be bound. By identifying the binding instruction, the specified network data source, specified feature library or feature table may be obtained. The binding instruction is used for commanding the coprocessor to establish a binding relationship between the address of the designated network data source and the designated feature library or feature table.
S103, according to the binding instruction, the original binding relation of the network data source appointed by the main processor is released.
In a specific application, if the original binding relation between the network data source appointed by the main processor and different feature libraries, feature library combinations, feature tables or feature table combinations is established, the original binding relation of the appointed network data source is released according to the binding instruction, and then the binding relation between the appointed network data source and at least one feature library to be bound or at least one feature table to be bound is established;
if the network data source specified by the main processor does not have a binding relationship with any feature library or feature table, step S103 is not required to be executed, and the binding relationship between the specified network data source and at least one feature library to be bound or at least one feature table to be bound is directly established.
S104, establishing a binding relation between the designated network data source and at least one feature library to be bound or at least one feature table to be bound.
FIG. 3 is a schematic diagram of an application scenario for multi-library multi-table dynamic binding based on a feature library management method;
wherein the feature library A, B, C includes 3 feature tables, respectively;
when the identification scene X is identified, a binding instruction sent by a main processor is received, and a dynamic binding relation between a designated network data source and the characteristic tables A1, B2 and C3 is established;
when the identification scene Y is identified, a binding instruction sent by a main processor is received, the dynamic binding relation between the appointed network data source and the characteristic tables A1, B2 and C3 is released according to the binding instruction, and the dynamic binding relation between the appointed network data source and the characteristic tables A2, C2 and C33 is established;
from the above, it can be seen that the network data identifying scene X and identifying scene Y are each identified only in the bound feature library.
According to the binding instruction or the binding instruction sent by the main processor, the dynamic binding relation between the identification scene X and the identification scene Y and the target feature library C3 can be realized, and one target feature library C3 does not need to be respectively created according to the identification scene X and the identification scene Y, so that the storage space is saved.
The method and the device have the advantages that under different application scenes, any specified network data source sent by the main processor is received and combined with different feature tables or an instruction for establishing a binding relation with different feature library combinations is executed, the purpose that different feature libraries, feature library combinations, feature tables or feature table combinations are used under different application scenes can be processed through the same coprocessor, the identification accuracy is improved, meanwhile, the fact that a specific feature library needs to be regenerated under each scene is avoided, the storage requirement is reduced, and meanwhile, the maintenance operation of equipment is simplified.
S105, receiving the network data which is sent by the main processor and originates from the appointed network data source.
S106, analyzing the network data in a feature library or a feature table which establishes a binding relation with the appointed network data source to obtain a feature value corresponding to the network data.
In particular applications, network data derived from a network data source includes, but is not limited to, eigenvalues and raw data.
If the network data is the original data, identifying and analyzing the original data in a feature library or a feature table bound with the network data source through a preset feature extraction algorithm to obtain a feature value corresponding to the original data.
In this embodiment, if the network data is a feature value, the network data is directly acquired.
The method for identifying the network data can be set to a specific implementation mode according to actual conditions. For example, if the feature value is key point data of a face, image data (i.e., network data) including the face in the network data source may be obtained first, and then the image data including the face is identified by a face key point identification algorithm, so as to obtain position data of the key point of each face in each image data including the face, that is, the obtained feature value.
The feature library may be classified according to the type of network data. For example, if the type of network data is a picture, a feature library of type a may be obtained. If the type of the network data is video, a feature library with the type B can be obtained.
If the type of the network data is text, a feature library with the type of C can be obtained; if the type of the network data is an array, a feature library with the type D can be obtained.
It should be noted that, the feature table can classify the types of algorithms for identifying the network data; for example, if a picture is identified by a face recognition algorithm based on a convolutional neural network model, a feature value obtained by the identification may be saved to a feature table with a type of a. If the image is identified by a human body identification algorithm based on a convolutional neural network model, the characteristic value obtained by identification can be stored in a characteristic table with the type b.
In one possible implementation, the method further includes:
receiving a feature library query instruction sent by the main processor;
inquiring the feature library information appointed by the main processor in the collaborative data processing library according to the feature library inquiry instruction, and returning a second inquiry result; the feature library information comprises at least one of the number of feature libraries, the number of feature tables in any feature library, the number of feature values in any feature table and the number of feature values in any feature library.
In a specific application, when a query instruction sent by the main processor is received, the feature library information specified by the main processor can be queried according to the query instruction to obtain a second query result, and the second query result is returned to the main processor. The feature library information includes, but is not limited to, at least one of the number of feature libraries, the number of feature tables in any feature library, the number of feature values in any feature table, and the number of feature values in any feature library.
The step can improve the management efficiency of the feature library by counting the feature library information in real time. And returning the feature library information to the main processor, so that a control instruction of the feature library, the feature table or the feature value, which is sent by the main processor according to the feature library information, is conveniently received. Meanwhile, as the feature library is stored in the storage medium of the coprocessor, the time required for reading the feature library is shortened, and the upgrading cost of the equipment is reduced.
As shown in fig. 4, before step S101, the method includes:
s201, establishing a co-data processing library in a storage medium on which the coprocessor is mounted;
s202, establishing a feature library in a co-processing database;
s203, generating a feature table in the feature library;
s204, storing the characteristic value into the characteristic table.
In a specific application, when the encrypted data sent by the main processor is received, decrypting the encrypted data to obtain a characteristic value; among other things, encryption algorithms include, but are not limited to, symmetric algorithms (Data Encryption Standard, DES), digital signature algorithms (Digital Signature Algorithm, DSA) or digest algorithms (i.e., MD5 algorithms).
When the original data sent by the main processor is received, extracting the characteristic value in the original data through a preset characteristic extraction algorithm;
if the feature library corresponding to the extracted feature value does not exist in the co-processing database, establishing a corresponding feature library in the co-processing database according to the type of the original data, generating a corresponding feature table in the feature library according to the type of a preset feature extraction algorithm, and storing the feature value into the feature table.
The steps are that the data processing library is built in the storage medium hung on the coprocessor, and the feature library is built in the data processing library, so that the time for reading and processing the feature value is shortened, and meanwhile, the cost for equipment upgrading is reduced.
In one embodiment, the method further comprises:
receiving a characteristic value or original data sent by a main processor; wherein the original data comprises multimedia data;
and extracting the characteristic value in the original data according to a preset characteristic extraction algorithm when the original data is received.
In particular applications, the manner in which the feature values are obtained includes, but is not limited to: 1. analyzing the network data to obtain a characteristic value; 2. directly acquiring a characteristic value sent by a main processor; and receiving the original data sent by the main processor, and extracting the characteristic value in the original data through a preset characteristic extraction algorithm. The raw data includes, but is not limited to, multimedia data; for example: picture, video or voice data. The preset feature extraction algorithm comprises, but is not limited to, a face recognition algorithm based on a convolutional neural network model, a face recognition algorithm based on a depth residual error network model or a human body recognition algorithm based on a convolutional neural network model. For example, when any picture is identified by a face recognition algorithm based on a neural network model, positioning data of eyebrows, ears, nose, throat and eyes of a person are obtained, namely the characteristic values.
In one possible implementation, after step S201, the method includes:
when the characteristic value sent by the main processor is received, performing validity check on the characteristic value;
detecting whether the characteristic value is encrypted or not when the validity check result of the characteristic value is legal;
decrypting the feature value when the feature value is encrypted.
In a specific application, when the feature value sent by the main processor is received, the feature value is subjected to validity check, and when the result of the validity check of the feature value is illegal, the feature value identification failure is judged, the feature value identification failure cannot be stored in any feature library or feature table, and a notification that the feature value identification failure cannot be stored is sent to the main processor. Encryption algorithms include, but are not limited to, symmetric algorithms (Data Encryption Standard, DES), digital signature algorithms (Digital Signature Algorithm, DSA) or digest algorithms (MD 5).
The validity of the characteristic value is specifically set according to the type of the original data or the type of the characteristic value. For example, if the original data is a picture, detecting whether the feature value accords with the data obtained after feature extraction of the picture; if the feature value is human body identification data, whether the feature value accords with the human body identification data is detected, if the feature value belongs to the human body identification data, the validity check result of the feature value is judged to be legal, and if the feature value does not belong to the human body identification data, for example, the feature value is key point positioning data of a human face, the validity check result of the feature value is judged to be illegal.
The method can filter out some unnecessary, demand-independent or illegal characteristic values through the validity check of the characteristic values, thereby improving the accuracy and the effectiveness of characteristic value identification.
In one possible implementation, the method further includes:
when a characteristic value inserting instruction sent by the main processor is received, inserting a characteristic value carried by the characteristic value inserting instruction into a characteristic table appointed by the main processor, generating a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction, and sending the characteristic value ID to the main processor;
when a characteristic value updating instruction sent by the main processor is received, updating a characteristic value corresponding to the characteristic value ID to be updated according to the characteristic value ID to be updated and new characteristic value or original data carried by the characteristic value updating instruction, and returning an updating result to the main processor;
deleting the characteristic value corresponding to the characteristic value ID to be deleted carried by the characteristic value deleting instruction when the characteristic value deleting instruction sent by the main processor is received, and returning a deleting result to the main processor;
and inquiring a characteristic value corresponding to a characteristic value ID to be inquired carried by the characteristic value inquiry instruction when the characteristic value inquiry instruction sent by the main processor is received, and returning a first inquiry result to the main processor.
The above steps manage the characteristic values according to the received instructions sent by the main processor, so that the efficiency of managing the characteristic values is improved, and meanwhile, the purpose of managing the corresponding characteristic values according to different requirements of a plurality of different application scenes is realized.
In one possible implementation, the method further includes:
encrypting the return result, and sending the encrypted return result to the main processor; the returned result comprises a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction, an updating result, a deleting result, a first query result and a second query result.
The steps ensure the data security in the data transmission process by encrypting the returned result.
By way of example and not limitation, the encryption algorithm used in this step may be the same as that used to decrypt encrypted data sent by the host processor, including but not limited to a symmetric algorithm (Data Encryption Standard, DES), a digital signature algorithm (Digital Signature Algorithm, DSA), or a digest algorithm (MD 5).
According to the embodiment, the feature library, the feature table or the feature value is managed through the control instruction sent by the main processor by providing different feature libraries or feature tables in various different application scenes, so that the management efficiency of the feature library is improved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present application in any way.
Corresponding to the feature library management method described in the above embodiments, fig. 5 shows a block diagram of the feature library management apparatus 100 provided in the embodiment of the present application, and for convenience of explanation, only the portions related to the embodiments of the present application are shown.
Referring to fig. 5, the apparatus 100 includes:
the first receiving module 101 is configured to receive a binding instruction sent by the main processor.
The identifying module 102 is configured to identify at least one feature library to be bound or at least one feature table to be bound and a specified network data source, which are included in the binding instruction; the feature library is stored in a collaborative data processing library, and at least one feature table is stored in each feature library;
a releasing module 103, configured to release the original binding relationship of the network data source specified by the host processor according to the binding instruction;
a first establishing module 104, configured to establish a binding relationship between the specified network data source and at least one feature library to be bound or at least one feature table to be bound;
a second receiving module 105, configured to receive network data from the specified network data source sent by the main processor;
and the analysis module 106 is configured to analyze the network data in a feature library or a feature table that establishes a binding relationship with the designated network data source, so as to obtain a feature value corresponding to the network data.
In one possible implementation, the apparatus 100 further includes:
the third receiving module is used for receiving the characteristic value or the original data sent by the main processor; wherein the original data comprises multimedia data;
and the extraction module is used for extracting the characteristic value in the original data according to a preset characteristic extraction algorithm when the original data is received.
In one possible implementation, the apparatus 100 further includes:
the second establishing module is used for establishing a co-data processing library in the storage medium mounted by the coprocessor;
the third establishing module is used for establishing a feature library in the co-processing database;
the generating module is used for generating a feature table in the feature library;
and the storage module is used for storing the characteristic value into the characteristic table.
In one possible implementation, the apparatus 100 includes:
the checking module is used for checking the validity of the characteristic value when the characteristic value sent by the main processor is received;
the detection module is used for detecting whether the characteristic value is encrypted or not when the validity check result of the characteristic value is legal;
and the decryption module is used for decrypting the characteristic value when the characteristic value is encrypted.
In one possible implementation, the apparatus 100 further includes:
the inserting module is used for inserting the characteristic value carried by the characteristic value inserting instruction into the characteristic table appointed by the main processor when receiving the characteristic value inserting instruction sent by the main processor, generating a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction and sending the characteristic value ID to the main processor;
the updating module is used for updating the characteristic value corresponding to the characteristic value ID to be updated according to the characteristic value ID to be updated and the new characteristic value or original data carried by the characteristic value updating instruction when the characteristic value updating instruction sent by the main processor is received, and returning an updating result to the main processor;
the deleting module is used for deleting the characteristic value corresponding to the characteristic value ID to be deleted carried by the characteristic value deleting instruction when the characteristic value deleting instruction sent by the main processor is received, and returning a deleting result to the main processor;
and the first query module is used for querying the characteristic value corresponding to the characteristic value ID to be queried carried by the characteristic value query instruction when receiving the characteristic value query instruction sent by the main processor, and returning a first query result to the main processor.
In one possible implementation, the apparatus 100 includes:
the encryption module is used for encrypting the return result and sending the encrypted return result to the main processor; the returned result comprises a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction, an updating result, a deleting result, a first query result and a second query result.
In one possible implementation, the apparatus 100 further includes:
the fourth receiving module is used for receiving the feature library query instruction sent by the main processor;
the second query module is used for querying the feature library information appointed by the main processor in the collaborative data processing library according to the feature library query instruction and returning a second query result; the feature library information comprises at least one of the number of feature libraries, the number of feature tables in any feature library, the number of feature values in any feature table and the number of feature values in any feature library.
According to the embodiment, the feature library is stored in the storage medium of the coprocessor, so that the time required for reading the feature library is shortened, the upgrading cost of equipment is reduced, and meanwhile, the coprocessor can identify the feature values of various different scenes at the same time by dynamically binding a designated network data source with a designated plurality of feature libraries or a plurality of feature tables, the identification efficiency is improved, and the storage requirement is reduced.
It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present application, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein again.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
The embodiment of the application also provides a co-data processor system, which comprises a main processor and at least one coprocessor, wherein each coprocessor is correspondingly provided with a storage medium, a computer program capable of running on the corresponding coprocessor is stored in the storage medium, and the steps in any of the method embodiments are realized when the coprocessor executes the computer program.
Embodiments of the present application also provide a computer readable storage medium storing a computer program which, when executed by a processor, implements steps that may implement the various method embodiments described above.
Embodiments of the present application provide a computer program product which, when run on a mobile terminal, causes the mobile terminal to perform steps that may be performed in the various method embodiments described above.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. With this understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by hardware related to computer program instructions, where the computer program may be stored on a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing device/terminal apparatus, recording medium, computer Memory, read-Only Memory (ROM), random access Memory (RAM, random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/network device and method may be implemented in other manners. For example, the apparatus/network device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.
Claims (10)
1. A feature library management method, for use with a coprocessor, the method comprising:
receiving a binding instruction sent by a main processor;
identifying at least one feature library to be bound or at least one feature table to be bound and a designated network data source contained in the binding instruction; the feature library is stored in a co-data processing library, at least one feature table is stored in each feature library, and the binding instruction comprises an address of a specified network data source and a name of at least one specified feature library to be bound or a name of at least one specified feature table to be bound;
according to the binding instruction, the original binding relation of the network data source appointed by the main processor is released;
establishing a binding relation between the designated network data source and at least one feature library to be bound or at least one feature table to be bound;
receiving network data from the designated network data source sent by the main processor;
and analyzing the network data in a feature library or a feature table which establishes a binding relation with the appointed network data source to obtain a feature value corresponding to the network data.
2. The feature library management method of claim 1, further comprising, prior to receiving the binding instruction sent by the host processor:
establishing a co-data processing library in a storage medium mounted by the coprocessor;
establishing a feature library in a cooperative data processing library;
generating a feature table in the feature library;
and storing the characteristic value into the characteristic table.
3. The feature library management method of claim 1, further comprising:
receiving the characteristic value or the original data sent by the main processor; wherein the original data comprises multimedia data;
and extracting the characteristic value in the original data according to a preset characteristic extraction algorithm when the original data is received.
4. The feature library management method of claim 1, wherein the identifying and analyzing the network data in the feature library or feature table that establishes a binding relationship with the designated network data source, to obtain the feature value corresponding to the network data, comprises:
when the characteristic value sent by the main processor is received, performing validity check on the characteristic value;
detecting whether the characteristic value is encrypted or not when the validity check result of the characteristic value is legal;
decrypting the feature value when the feature value is encrypted.
5. The feature library management method of claim 1, further comprising:
when a characteristic value inserting instruction sent by the main processor is received, inserting a characteristic value carried by the characteristic value inserting instruction into a characteristic table appointed by the main processor, generating a characteristic value ID corresponding to the characteristic value carried by the characteristic value inserting instruction, and sending the characteristic value ID to the main processor;
when a characteristic value updating instruction sent by the main processor is received, updating a characteristic value corresponding to the characteristic value ID to be updated according to the characteristic value ID to be updated and new characteristic value or original data carried by the characteristic value updating instruction, and returning an updating result to the main processor;
deleting the characteristic value corresponding to the characteristic value ID to be deleted carried by the characteristic value deleting instruction when the characteristic value deleting instruction sent by the main processor is received, and returning a deleting result to the main processor;
and inquiring a characteristic value corresponding to a characteristic value ID to be inquired carried by the characteristic value inquiry instruction when the characteristic value inquiry instruction sent by the main processor is received, and returning a first inquiry result to the main processor.
6. The feature library management method of claim 1, further comprising:
receiving a feature library query instruction sent by the main processor;
inquiring the feature library information appointed by the main processor in the collaborative data processing library according to the feature library inquiry instruction, and returning a second inquiry result; the feature library information comprises at least one of the number of feature libraries, the number of feature tables in any feature library, the number of feature values in any feature table and the number of feature values in any feature library.
7. A coprocessor, comprising:
the first receiving module is used for receiving the binding instruction sent by the main processor;
the identification module is used for identifying at least one feature library to be bound or at least one feature table to be bound and a designated network data source contained in the binding instruction; the feature library is stored in a co-data processing library, at least one feature table is stored in each feature library, and the binding instruction comprises an address of a specified network data source and a name of at least one specified feature library to be bound or a name of at least one specified feature table to be bound;
the releasing module is used for releasing the original binding relation of the network data source appointed by the main processor according to the binding instruction;
the first establishing module is used for establishing a binding relation between the appointed network data source and at least one feature library to be bound or at least one feature table to be bound;
a second receiving module, configured to receive network data from the specified network data source sent by the main processor;
and the analysis module is used for analyzing the network data in a feature library or a feature table which establishes a binding relation with the appointed network data source to obtain a feature value corresponding to the network data.
8. The coprocessor of claim 7, wherein prior to receiving the bind instruction sent by the host processor, further comprising:
the second establishing module is used for establishing a co-data processing library in the storage medium mounted by the coprocessor;
the third establishing module is used for establishing a feature library in the collaborative data processing library;
the generating module is used for generating a feature table in the feature library;
and the storage module is used for storing the characteristic value into the characteristic table.
9. A co-data processor system comprising a main processor and at least one co-processor, each co-processor being loaded with a storage medium, the storage medium having stored therein a computer program executable on the corresponding co-processor, the co-processor implementing the method of any one of claims 1 to 6 when executing the computer program.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911381948.5A CN111221803B (en) | 2019-12-27 | 2019-12-27 | Feature library management method and coprocessor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911381948.5A CN111221803B (en) | 2019-12-27 | 2019-12-27 | Feature library management method and coprocessor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111221803A CN111221803A (en) | 2020-06-02 |
| CN111221803B true CN111221803B (en) | 2024-01-09 |
Family
ID=70811165
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911381948.5A Active CN111221803B (en) | 2019-12-27 | 2019-12-27 | Feature library management method and coprocessor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111221803B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409470B (en) * | 2023-12-15 | 2024-03-15 | 千巡科技(深圳)有限公司 | Face recognition feature data dynamic matching method, system, device and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101208662A (en) * | 2005-03-21 | 2008-06-25 | 德克斯泰拉股份有限公司 | Modular applications for mobile data system |
| CN101442489A (en) * | 2008-12-30 | 2009-05-27 | 北京畅讯信通科技有限公司 | Method for recognizing flux based on characteristic library |
| CN102571813A (en) * | 2012-02-12 | 2012-07-11 | 周良文 | Response system with bound disposition of communication terminals and networking identity |
| CN108189804A (en) * | 2017-12-29 | 2018-06-22 | 威马智慧出行科技（上海）有限公司 | A kind of face identification system and face identification method for vehicle |
| CN109119084A (en) * | 2018-07-13 | 2019-01-01 | 广东电网有限责任公司 | A kind of dispatch call method and system based on speech recognition |
| CN109917918A (en) * | 2019-03-07 | 2019-06-21 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment based on augmented reality |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101919771B1 (en) * | 2017-06-12 | 2019-02-11 | 주식회사 티맥스데이터 | Optimization technique for database application |
| US11120001B2 (en) * | 2017-11-30 | 2021-09-14 | International Business Machines Corporation | Table discovery in distributed and dynamic computing systems |
-
2019
- 2019-12-27 CN CN201911381948.5A patent/CN111221803B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101208662A (en) * | 2005-03-21 | 2008-06-25 | 德克斯泰拉股份有限公司 | Modular applications for mobile data system |
| CN101442489A (en) * | 2008-12-30 | 2009-05-27 | 北京畅讯信通科技有限公司 | Method for recognizing flux based on characteristic library |
| CN102571813A (en) * | 2012-02-12 | 2012-07-11 | 周良文 | Response system with bound disposition of communication terminals and networking identity |
| CN108189804A (en) * | 2017-12-29 | 2018-06-22 | 威马智慧出行科技（上海）有限公司 | A kind of face identification system and face identification method for vehicle |
| CN109119084A (en) * | 2018-07-13 | 2019-01-01 | 广东电网有限责任公司 | A kind of dispatch call method and system based on speech recognition |
| CN109917918A (en) * | 2019-03-07 | 2019-06-21 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment based on augmented reality |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111221803A (en) | 2020-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109189367B (en) | Data processing method, device, server and storage medium | |
| CN110008428B (en) | News data processing method and device, blockchain node equipment and storage medium | |
| CN112949545B (en) | Method, apparatus, computing device and medium for recognizing face image | |
| CN111797430A (en) | Data verification method, device, server and storage medium | |
| CN112990484B (en) | Model joint training method, device and equipment based on asymmetric federated learning | |
| CN104346550A (en) | Information processing method and electronic equipment | |
| CN112101200A (en) | Human face anti-recognition method, system, computer equipment and readable storage medium | |
| CN111221803B (en) | Feature library management method and coprocessor | |
| CN114745373A (en) | File transmission method, device, equipment and storage medium | |
| CN116311389B (en) | Fingerprint identification method and device | |
| WO2023115968A1 (en) | Method and device for identifying violation data at user end, medium, and program product | |
| GB2546567A (en) | Method of associating a person with a digital object | |
| EP3398108A1 (en) | A method and system for visual privacy protection for mobile and wearable devices | |
| CN113052045B (en) | Method, apparatus, computing device and medium for identifying finger vein image | |
| CN111191612B (en) | Video image matching method, device, terminal equipment and readable storage medium | |
| CN113297617A (en) | Authority data acquisition method and device, computer equipment and storage medium | |
| CN113890754A (en) | Data transmission method, terminal, system and readable storage medium | |
| CN111177765A (en) | Financial big data processing method, storage medium and system | |
| CN112364683A (en) | Case evidence fixing method and device | |
| CN112364682A (en) | Case searching method and device | |
| CN112118366A (en) | Method and device for transmitting face picture data | |
| CN111783914A (en) | Information processing method, information processing device and mobile terminal | |
| CN107944231A (en) | Copyright registration method, apparatus, system, electronic equipment and computer-readable medium | |
| CN112565311A (en) | Rapid evidence obtaining method for telecommunication fraud case victim | |
| Rytsar et al. | Interactive segmentation with hidden object-based annotations: toward Smart Media |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |