CN111061706B - Face recognition algorithm model cleaning method and device and storage medium - Google Patents
Face recognition algorithm model cleaning method and device and storage medium Download PDFInfo
- Publication number
- CN111061706B CN111061706B CN201911084161.2A CN201911084161A CN111061706B CN 111061706 B CN111061706 B CN 111061706B CN 201911084161 A CN201911084161 A CN 201911084161A CN 111061706 B CN111061706 B CN 111061706B
- Authority
- CN
- China
- Prior art keywords
- feature
- data
- algorithm model
- face image
- features
- 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
Images
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
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a face recognition algorithm model cleaning method, a face recognition algorithm model cleaning device and a storage medium. The method comprises the following steps: acquiring a stored face image from a database; acquiring the characteristics of the face image, and packaging the characteristics into characteristic packaging data with a preset data structure, wherein the characteristic packaging data comprises the characteristics and description information; and storing the feature encapsulation data, and associating the face image with the feature encapsulation data. According to the face recognition algorithm model cleaning method, the face recognition algorithm model cleaning device and the storage medium, the features of the face image are obtained, the features are packaged into feature packaging data with a preset data structure, then the feature packaging data are stored, the face image and the feature packaging data are associated, and the preset data structure comprises the features and at least one piece of description information, so that the face recognition algorithm model cleaning method, the face recognition algorithm model cleaning device and the storage medium are convenient to recognize during subsequent upgrading, the feature data extracted by models of all versions are prevented from being stored at the same time, the face recognition efficiency is improved, and data redundancy is avoided.
Description
Technical Field
The present application relates to the field of face recognition, and in particular, to a method and an apparatus for cleaning a face recognition algorithm model, and a storage medium.
Background
In the prior art, after a feature extraction algorithm model is upgraded every time, feature extraction needs to be performed on a face image in a database again, feature data extracted by an old version of the model and feature data extracted by a new version of the model are both stored in the database, when face recognition is performed, features of each face need to be compared with feature data extracted by each version of the model, so that face recognition efficiency is low, and data redundancy is caused because the feature data extracted by each version of the model occupies a large storage space.
Disclosure of Invention
The application provides a face recognition algorithm model cleaning method, a face recognition algorithm model cleaning device and a storage medium, and can solve the problems of low face recognition efficiency and data redundancy in the prior art.
In order to solve the technical problem, the application adopts a technical scheme that: a face recognition algorithm model cleaning method is provided, and the method comprises the following steps:
acquiring a stored face image from a database;
acquiring the characteristics of the face image, and packaging the characteristics of the face image into characteristic packaging data with a preset data structure, wherein the characteristic packaging data comprises the characteristics and description information of the face image;
and storing the feature encapsulation data, and associating the face image with the feature encapsulation data.
In order to solve the above technical problem, another technical solution adopted by the present application is: there is provided a face recognition algorithm model cleaning apparatus, the apparatus comprising:
the acquisition module is used for acquiring the stored face image from the database;
the feature packaging module is used for acquiring the features of the face image and packaging the features of the face image into feature packaging data with a preset data structure, wherein the feature packaging data comprises the features and description information of the face image;
and the storage association module is used for storing the feature packaging data and associating the face image with the feature packaging data.
In order to solve the above technical problem, another technical solution adopted by the present application is: there is provided a face recognition algorithm model washing apparatus, the apparatus comprising a processor, and a memory coupled to the processor,
the memory stores program instructions for implementing the face recognition algorithm model cleaning method;
the processor is configured to execute the program instructions stored in the memory for performing a face recognition algorithm model cleaning.
In order to solve the above technical problem, another technical solution adopted by the present application is: a storage medium is provided, in which program instructions capable of implementing the above-described face recognition algorithm model cleaning method are stored.
The beneficial effect of this application is: according to the face recognition algorithm model cleaning method, the face recognition algorithm model cleaning device and the storage medium, the features of the face image are obtained, the features are packaged into feature packaging data with a preset data structure, then the feature packaging data are stored, and the face image is associated with the feature packaging data.
Drawings
FIG. 1 is a schematic flow chart of a face recognition algorithm model cleaning method according to a first embodiment of the present invention;
FIG. 2 is a diagram illustrating a default data structure in the method according to the first embodiment of the present invention;
fig. 3 is a schematic diagram illustrating header information in a preset data structure in the method according to the first embodiment of the present invention;
FIG. 4 is a schematic flow chart of a cleaning method for a face recognition algorithm model according to a second embodiment of the present invention;
FIG. 5 is a schematic flow chart of a cleaning method for a face recognition algorithm model according to a third embodiment of the present invention;
FIG. 6 is a schematic flow chart of a cleaning method for a face recognition algorithm model according to a fourth embodiment of the present invention;
FIG. 7 is a schematic structural diagram of a face recognition algorithm model device according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a face recognition algorithm model device according to another embodiment of the present invention;
fig. 9 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first", "second" and "third" in this application are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any indication of the number of technical features indicated. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of the feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless explicitly specifically limited otherwise. In the embodiment of the present application, all the directional indicators (such as upper, lower, left, right, front, and rear … …) are used only to explain the relative positional relationship between the components, the movement, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indicator is changed accordingly. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
Fig. 1 is a schematic flow chart of a face recognition algorithm model cleaning method according to a first embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 1 if the results are substantially the same. As shown in fig. 1, the face recognition algorithm model cleaning method includes the following steps:
s101, the stored face image is obtained from a database.
In this embodiment, the database is a local database or a cloud database, for example, the database is a smart public security face database, which includes a black and white list database, a static database and a snapshot database, and data interaction can be performed between the databases.
S102, obtaining the characteristics of the face image, and packaging the characteristics into characteristic packaging data with a preset data structure, wherein the characteristic packaging data comprises the characteristics and at least one piece of description information.
In this embodiment, the manner of obtaining the features of the face image may be to obtain original feature data of the face image in a database, and encapsulate and store the original feature data of the face image that has been stored in the database.
In this embodiment, the manner of acquiring the features of the face image may be to re-extract the features of the face image.
In this embodiment, the preset data structure may include data content, the feature encapsulation data stored according to the preset data structure may include different data content, and other attributes except for the feature are added to the feature encapsulation data, where the feature encapsulation data includes the feature and at least one piece of description information; further, the at least one description information may include at least one of feature extraction algorithm model identification information, custom attribute information, and verification information; further, the customized attribute information may be basic feature information of the face image, for example, wearing glasses.
In an optional embodiment, the preset data structure may further include a data type and/or a data storage mode. Depending on the programming language, the data types may vary. The data storage mode can comprise a linear table, a stack, a queue, a binary tree and the like. The preset data structure can be adapted to subsequent comparison requirements and upgrading requirements.
In an alternative embodiment, the preset data structure is shown in fig. 2, and the data structure of the packet header information in the preset data structure is shown in fig. 3, where. And the dhfv is identification information of the feature extraction algorithm model, the check packet is check information, the user-defined data is user-defined attribute information, and the characteristic value is the acquired characteristic of the face image. The step of encapsulating the feature into feature encapsulation data having a preset data structure specifically includes: firstly, acquiring feature extraction algorithm model identification information, custom attribute information and verification information of the feature; then, data packaging is carried out on the features, the feature extraction algorithm model identification information, the custom attribute information and the verification information to form a feature package of the features; finally, acquiring packet header information according to the feature packet, and adding a packet header in the feature packet according to the packet header information to obtain the feature encapsulation data.
S103, storing the feature encapsulation data, and associating the face image with the feature encapsulation data.
In this embodiment, the feature encapsulation data is stored in a database, and is associated with the face image in the database to establish a first image file.
In an optional embodiment, since the feature encapsulation data has a data structure adapted to the comparison requirement and the upgrade requirement, the feature encapsulation data can be directly read after being acquired from the database during subsequent comparison and upgrade, and data conversion or decoding is not required.
In an alternative embodiment, after the feature encapsulation data is stored, the original feature data of the face image is deleted from its storage location. The method avoids the data redundancy caused by the fact that the feature data extracted by the models of the versions occupy large storage space.
FIG. 4 is a flowchart illustrating a method for cleaning a face recognition algorithm model according to a second embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 4 if the results are substantially the same. As shown in fig. 4, the method for cleaning the face recognition algorithm model comprises the following steps:
s201, the stored face image is obtained from the database.
S202, obtaining the characteristics of the face image, and comparing the characteristics of the face image with the stored characteristic packaging data to obtain a comparison result.
And S203, judging whether the features of the face image are matched with the stored feature encapsulation data according to the preset matching conditions and the comparison result.
And S204, when the matching is successful, associating the face image serving as a side file with the matched feature packaging data.
S205, when the matching fails, the features of the face image are packaged into feature packaging data with a preset data structure, the feature packaging data are stored, and the stored face image is used as a main file to be associated with the feature packaging data.
In this embodiment, the preset matching condition may be that the similarity is greater than or equal to a preset similarity threshold.
In this embodiment, after the features of the face image are obtained, before data encapsulation and storage are performed, it is determined whether the features of the face image can be matched with stored feature encapsulation data, and if matching is successful, it indicates that the features of the current face image and the feature encapsulation data on matching both point to the same object (the same person), the features of the object have already been data encapsulated and stored, and it is not necessary to perform data encapsulation and storage on the features of the current face image, and directly associate the current face image with the feature encapsulation data on matching, so as to implement one person (object) one file. And if the matching fails, the characteristic of the object corresponding to the current face image is not subjected to data packaging and storage.
In an optional implementation manner, in the process of extracting the features of the face image, a load balancing strategy is adopted, specifically, a first calculation region of a computer is divided into a plurality of calculation sub-regions, the plurality of calculation sub-regions are mapped to a plurality of calculation nodes in a one-to-one correspondence manner, and a feature extraction operator is configured at each calculation node; and distributing a plurality of face images to the computing nodes in a rotating or random mode so as to execute the task of extracting the features of the face images in parallel. The first calculation region may be a memory region or a part of a memory region. Further, each computing node may be further configured with a retrieval operator, the retrieval operator is used for feature comparison, after the feature extraction operator extracts features of the face image at each computing node, the retrieval operator compares the features of the face image with stored feature encapsulation data to obtain a comparison result, according to a preset matching condition, whether the features of the face image are matched with the stored feature encapsulation data is judged according to the comparison result, and step S202 and step S203 are executed in parallel. Furthermore, each computing node is configured with a scheduling algorithm, performs task progress management and task scheduling, performs feedback when the current task is interrupted, and initiates execution of the next task when the current task is completed.
FIG. 5 is a flowchart illustrating a method for cleaning a face recognition algorithm model according to a third embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 5 if the results are substantially the same. As shown in fig. 5, the method for cleaning the face recognition algorithm model comprises the following steps:
s301, the stored face image is obtained from the database.
S302, obtaining the characteristics of the face image, and packaging the characteristics into characteristic packaging data with a preset data structure, wherein the characteristic packaging data comprises the characteristics and at least one piece of description information.
S303, storing the feature encapsulation data, and associating the face image with the feature encapsulation data.
S304, obtaining the face image to be recognized, and extracting the characteristics of the face image to be recognized.
S305, comparing the characteristics of the facial image to be recognized with the stored characteristic packaging data to obtain a comparison result.
And S306, judging whether the features of the facial image to be recognized are matched with the stored feature encapsulation data according to a preset matching condition and the comparison result.
And S307, when the matching is successful, acquiring a main file and/or a sub-file associated with the matched feature packaging data, and outputting the main file and/or the sub-file as a recognition result of the facial image to be recognized.
For specific reference to steps S301 to S303, reference is made to the first embodiment and the second embodiment, which are not described in detail herein.
In this embodiment, the preset matching condition may be that the similarity is greater than or equal to a preset similarity threshold.
In this embodiment, feature recognition is performed on a face image to be recognized, features of the face image to be recognized are compared with feature encapsulation data stored in a database, so that the features are matched with the feature encapsulation data, and all main files and sub-files related to the matched feature encapsulation data are output as recognition results, so as to realize a function of searching a picture with a picture.
In an optional embodiment, in step S304, the features of the image of the face to be recognized and the basic feature information may be extracted, where the basic feature information is the appearance description information of the face, such as: wearing glasses, skin tone or clothing, etc. In step S305, first, the basic feature information is compared with description information in stored feature encapsulation data to obtain a first comparison result; then, comparing the characteristics with the characteristics of the characteristic packaging data in the first comparison result to obtain a second comparison result; and finally, judging whether the features of the face image to be recognized are matched with the stored feature packaging data or not according to the second comparison result. In the embodiment, the basic feature information is used for preliminary comparison, and then the feature recognition is performed to accelerate the recognition speed.
FIG. 6 is a flowchart illustrating a cleaning method of a face recognition algorithm model according to a fourth embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 6 if the results are substantially the same. As shown in fig. 6, the method for cleaning the face recognition algorithm model includes the following steps:
s401, the stored face image is obtained from the database.
S402, obtaining the characteristics of the face image, and packaging the characteristics into characteristic packaging data with a preset data structure, wherein the characteristic packaging data comprises the characteristics and at least one piece of description information.
And S403, storing the feature encapsulation data, and associating the face image with the feature encapsulation data.
S404, receiving an algorithm model upgrading instruction.
S405, the stored face image and feature encapsulation data related to the face image are obtained.
S406, judging whether the face image meets the upgrading condition according to the feature extraction algorithm model identification information in the feature encapsulation data.
And S407, when the judgment result is yes, re-extracting the features of the face image by adopting the upgraded feature extraction algorithm model, updating the features in the feature encapsulation data into the re-extracted features, and storing the feature encapsulation data.
Steps S401 to S403 refer to the first embodiment and the second embodiment specifically, and are not described in detail herein.
In this embodiment, the upgrade condition is: the feature extraction algorithm model identification information is matched with the current feature extraction algorithm model.
In this embodiment, feature data of a face image in a database is upgraded, and in step S406, it is determined whether the current face image and feature encapsulation data satisfy an upgrade condition, for example, a wisdom public security face database is usually maintained by several companies, when a company a upgrades, it is determined according to feature extraction algorithm model identification information in feature encapsulation data that features of the face image are extracted by the company, and a version number is an old version, and the face image is upgraded; and judging that the features of the face image are extracted by other companies according to the feature extraction algorithm model identification information in the feature encapsulation data, or judging that the features of the face image are extracted by the company but have the latest version number, and not upgrading the face image.
FIG. 7 is a schematic structural diagram of a face recognition algorithm model cleaning device according to an embodiment of the present invention. As shown in fig. 7, the apparatus 50 includes: an acquisition module 51, a feature encapsulation module 52 and a storage association module 53.
The obtaining module 51 is configured to obtain a stored face image from a database.
And a feature encapsulation module 52, configured to obtain features of the face image, and encapsulate the features into feature encapsulation data with a preset data structure, where the preset data structure includes the features and at least one piece of description information.
And a storage association module 53, configured to store the feature encapsulation data, and associate the face image with the feature encapsulation data.
Optionally, the manner of obtaining the features of the face image may be to obtain original feature data of the face image in a database, and encapsulate and store the original feature data of the face image already stored in the database.
Optionally, the manner of obtaining the features of the face image may be to re-extract the features of the face image.
Optionally, the preset data structure may include data content, the feature encapsulation data stored according to the preset data structure may include different data content, and other attributes except the feature are added to the feature encapsulation data, where the feature encapsulation data includes the feature and at least one piece of description information; further, the at least one description information may include at least one of feature extraction algorithm model identification information, custom attribute information, and verification information; further, the customized attribute information may be basic feature information of the face image, for example, wearing glasses.
In an optional embodiment, the preset data structure may further include a data type and/or a data storage mode. Depending on the programming language, the data types may vary. The data storage mode can comprise a linear table, a stack, a queue, a binary tree and the like. The preset data structure can adapt to subsequent comparison requirements and upgrading requirements.
Referring to fig. 8, fig. 8 is a schematic structural diagram of a face recognition algorithm model cleaning device according to an embodiment of the present invention. As shown in fig. 8, the face recognition algorithm model washing device 60 includes a processor 61 and a memory 62 coupled to the processor 61.
The memory 62 stores program instructions for implementing the face recognition algorithm model cleaning method according to any of the embodiments described above.
The processor 61 is configured to execute program instructions stored in the memory 62 for performing model cleaning of the face recognition algorithm.
The processor 61 may also be referred to as a CPU (Central Processing Unit). The processor 61 may be an integrated circuit chip having signal processing capabilities. The processor 61 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Referring to fig. 9, fig. 9 is a schematic structural diagram of a storage medium according to an embodiment of the invention. The storage medium of the embodiment of the present invention stores program instructions 71 capable of implementing all the methods described above, where the program instructions 71 may be stored in the storage medium in the form of a software product, and include several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage device includes: various media capable of storing program codes, such as a usb disk, a portable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is only a logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit. The above embodiments are merely examples and are not intended to limit the scope of the present disclosure, and all modifications, equivalents, and flow charts using the contents of the specification and drawings are included in the scope of the present disclosure.
Claims (11)
1. A face recognition algorithm model cleaning method is characterized by comprising the following steps:
acquiring a stored face image from a database;
acquiring the characteristics of the face image;
acquiring feature extraction algorithm model identification information, user-defined attribute information and verification information of the features;
performing data encapsulation on the features, the feature extraction algorithm model identification information, the custom attribute information and the verification information to form a feature package of the features;
acquiring packet header information according to the feature packet, and adding a packet header to the feature packet according to the packet header information to acquire feature encapsulation data;
and storing the feature packaging data, and associating the face image with the feature packaging data.
2. The method for cleaning a human face recognition algorithm model according to claim 1, wherein the obtaining the features of the human face image comprises:
acquiring original characteristic data of the face image in a database;
or, re-extracting the features of the face image.
3. The method for cleaning a face recognition algorithm model according to claim 1 or 2, wherein after the storing the feature encapsulation data, the method further comprises:
and deleting the original characteristic data of the face image from the storage position of the original characteristic data.
4. The method of claim 1, wherein the obtaining of the feature extraction algorithm model identification information, the custom attribute information, and the verification information of the features comprises:
comparing the characteristics of the face image with stored characteristic packaging data to obtain a comparison result;
judging whether the features of the face image are matched with the stored feature encapsulation data or not according to a preset matching condition and the comparison result;
when the matching is successful, the face image is used as a side file to be associated with the matched feature packaging data;
when the matching fails, sequentially executing the feature extraction algorithm model identification information, the user-defined attribute information and the verification information for obtaining the features; performing data encapsulation on the features, the feature extraction algorithm model identification information, the custom attribute information and the verification information to form a feature package of the features; the step of obtaining packet header information according to the feature packet, and adding a packet header to the feature packet according to the packet header information to obtain the feature encapsulation data, and the step of storing the feature encapsulation data and associating the stored face image as a main file with the feature encapsulation data.
5. The method of claim 4, wherein associating the stored face image as a master with the feature encapsulation data further comprises:
acquiring a face image to be recognized, and extracting the characteristics of the face image to be recognized;
comparing the characteristics of the facial image to be recognized with the stored characteristic packaging data to obtain a comparison result;
judging whether the features of the facial image to be recognized are matched with the stored feature encapsulation data or not according to preset matching conditions and the comparison result;
and when the matching is successful, acquiring a main file and/or a sub-file associated with the matched feature packaging data, and outputting the main file and/or the sub-file as the recognition result of the facial image to be recognized.
6. The method for cleaning a face recognition algorithm model according to claim 1, after storing the feature pack data and associating the face image with the feature pack data, further comprising:
receiving an algorithm model upgrading instruction;
acquiring a stored face image and feature encapsulation data associated with the face image;
judging whether the face image meets an upgrading condition according to feature extraction algorithm model identification information in the feature encapsulation data;
and if so, re-extracting the features of the face image by adopting an upgraded feature extraction algorithm model, updating the features in the feature encapsulation data into the re-extracted features, and storing the feature encapsulation data.
7. The method for cleaning a face recognition algorithm model according to claim 6, wherein the upgrade condition is: the feature extraction algorithm model identification information is matched with the current feature extraction algorithm model.
8. The method for cleaning a face recognition algorithm model according to claim 1, wherein in the step of extracting the features of the face image,
dividing a first calculation region of a computer into a plurality of calculation sub-regions, mapping the calculation sub-regions to a plurality of calculation nodes in a one-to-one correspondence manner, and configuring a feature extraction operator at each calculation node;
distributing a plurality of face images to the computing nodes in a training or random mode so as to execute the task of extracting the features of the face images in parallel.
9. A face recognition algorithm model cleaning apparatus, the apparatus comprising:
the acquisition module is used for acquiring the stored face image from the database;
the feature encapsulation module is used for acquiring the features of the face image and acquiring feature extraction algorithm model identification information, custom attribute information and verification information of the features; performing data encapsulation on the features, the feature extraction algorithm model identification information, the custom attribute information and the verification information to form a feature package of the features; acquiring packet header information according to the feature packet, and adding a packet header to the feature packet according to the packet header information to acquire feature encapsulation data;
and the storage association module is used for storing the feature packaging data and associating the face image with the feature packaging data.
10. A face recognition algorithm model cleaning device, characterized in that the device comprises a processor and a memory coupled with the processor,
the memory stores program instructions for implementing the face recognition algorithm model cleaning method according to any one of claims 1-8;
the processor is configured to execute the program instructions stored in the memory for performing a face recognition algorithm model cleaning.
11. A storage medium having stored therein program instructions which, when executed by a processor, implement the face recognition algorithm model cleaning method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911084161.2A CN111061706B (en) | 2019-11-07 | 2019-11-07 | Face recognition algorithm model cleaning method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911084161.2A CN111061706B (en) | 2019-11-07 | 2019-11-07 | Face recognition algorithm model cleaning method and device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111061706A CN111061706A (en) | 2020-04-24 |
| CN111061706B true CN111061706B (en) | 2023-04-18 |
Family
ID=70298514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911084161.2A Active CN111061706B (en) | 2019-11-07 | 2019-11-07 | Face recognition algorithm model cleaning method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111061706B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308031A (en) * | 2020-11-25 | 2021-02-02 | 浙江大华系统工程有限公司 | Universal face recognition and face feature information base generation method, device and equipment |
| CN112989096B (en) * | 2021-03-05 | 2023-03-14 | 浙江大华技术股份有限公司 | Face feature migration method, electronic device and storage medium |
| CN113139508B (en) * | 2021-05-12 | 2023-11-14 | 深圳他米科技有限公司 | Hotel safety early warning method, device and equipment based on artificial intelligence |
| CN116910695B (en) * | 2023-09-11 | 2024-01-05 | 哈尔滨工程大学三亚南海创新发展基地 | Marking method of equipment monitoring result and checking method of equipment monitoring data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108229378A (en) * | 2017-12-29 | 2018-06-29 | 浪潮金融信息技术有限公司 | Face image data generation method and device, computer storage media, terminal |
| CN108924430A (en) * | 2018-08-27 | 2018-11-30 | Oppo广东移动通信有限公司 | photo storage method, device, terminal and computer readable storage medium |
| CN109241852A (en) * | 2018-08-10 | 2019-01-18 | 广州杰赛科技股份有限公司 | Face identification method, device with supplementary features, computer equipment |
| CN109284722A (en) * | 2018-09-29 | 2019-01-29 | 佳都新太科技股份有限公司 | Image processing method, device, face recognition device and storage medium |
| CN110019891A (en) * | 2017-12-29 | 2019-07-16 | 浙江宇视科技有限公司 | Image storage method, image search method and device |
| CN110209866A (en) * | 2019-05-30 | 2019-09-06 | 苏州浪潮智能科技有限公司 | A kind of image search method, device, equipment and computer readable storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4283315B2 (en) * | 2007-02-01 | 2009-06-24 | シャープ株式会社 | Image history storage method, image history storage system, image history storage program, and recording medium |
-
2019
- 2019-11-07 CN CN201911084161.2A patent/CN111061706B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108229378A (en) * | 2017-12-29 | 2018-06-29 | 浪潮金融信息技术有限公司 | Face image data generation method and device, computer storage media, terminal |
| CN110019891A (en) * | 2017-12-29 | 2019-07-16 | 浙江宇视科技有限公司 | Image storage method, image search method and device |
| CN109241852A (en) * | 2018-08-10 | 2019-01-18 | 广州杰赛科技股份有限公司 | Face identification method, device with supplementary features, computer equipment |
| CN108924430A (en) * | 2018-08-27 | 2018-11-30 | Oppo广东移动通信有限公司 | photo storage method, device, terminal and computer readable storage medium |
| CN109284722A (en) * | 2018-09-29 | 2019-01-29 | 佳都新太科技股份有限公司 | Image processing method, device, face recognition device and storage medium |
| CN110209866A (en) * | 2019-05-30 | 2019-09-06 | 苏州浪潮智能科技有限公司 | A kind of image search method, device, equipment and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 杨曼.基于Hadoop平台的海量视频检索系统设计与实现.《中国优秀硕士学位论文电子期刊》.2016,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111061706A (en) | 2020-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111061706B (en) | Face recognition algorithm model cleaning method and device and storage medium | |
| CN108830138B (en) | Livestock identification method, device and storage medium | |
| CN110019080B (en) | Data access method and device | |
| CN108536745B (en) | Shell-based data table extraction method, terminal, equipment and storage medium | |
| CN109271363B (en) | File storage method and device | |
| CN107590291A (en) | A kind of searching method of picture, terminal device and storage medium | |
| CN112328823A (en) | Training method and device for multi-label classification model, electronic equipment and storage medium | |
| CN107798082B (en) | File label processing method and device | |
| CN107832148A (en) | Performance optimization method and equipment | |
| CN112667415A (en) | Data calling method and device, readable storage medium and electronic equipment | |
| CN110602212A (en) | Application service management method, device and storage medium | |
| CN116384109A (en) | Novel power distribution network-oriented digital twin model automatic reconstruction method and device | |
| CN111338716A (en) | Data processing method and device based on rule engine and terminal equipment | |
| CN111597553A (en) | Process processing method, device, equipment and storage medium in virus searching and killing | |
| CN114090083B (en) | Visual component configuration method, device, equipment and storage medium | |
| CN109120509A (en) | A kind of method and device that information is collected | |
| CN115617748A (en) | Material list information analysis method, device, equipment and storage medium | |
| CN114116699A (en) | Data processing method and device, electronic equipment and computer readable storage medium | |
| CN114138813A (en) | Attribute configuration method and related device | |
| CN112783500B (en) | Method and device for generating compiling optimization information and electronic equipment | |
| CN114218261A (en) | Data query method and device, storage medium and electronic equipment | |
| CN112181796A (en) | Information acquisition method, device, server and storage medium | |
| CN110471708B (en) | Method and device for acquiring configuration items based on reusable components | |
| CN114157662A (en) | Cloud platform parameter adaptation method and device, terminal equipment and storage medium | |
| CN111858590A (en) | Storage system metadata organization method, system, terminal and storage medium |
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 |