CN113688712A - A portrait recognition method, device, electronic device and storage medium - Google Patents
A portrait recognition method, device, electronic device and storage medium Download PDFInfo
- Publication number
- CN113688712A CN113688712A CN202110947338.8A CN202110947338A CN113688712A CN 113688712 A CN113688712 A CN 113688712A CN 202110947338 A CN202110947338 A CN 202110947338A CN 113688712 A CN113688712 A CN 113688712A
- Authority
- CN
- China
- Prior art keywords
- portrait
- database
- target
- determining
- human
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012545 processing Methods 0.000 claims description 21
- 238000012216 screening Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000001914 filtration Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000000344 soap Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007175 bidirectional communication Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Image Analysis (AREA)
Abstract
本发明实施例公开了一种人像识别方法、装置、电子设备及存储介质。该方法包括：获取目标人像数据的数据信息；其中，所述数据信息包括人像信息和位置信息；根据所述人像信息和/或所述位置信息从至少一个候选人像数据库中确定第一目标人像数据库，以及确定所述第一目标人像数据库的数据库对比顺序；根据所述人像信息、所述第一目标人像数据库和所述数据库对比顺序确定人像识别结果。通过运行本发明实施例所提供的技术方案，可以解决当应用场景为多种时，需要分别针对各自场景的特征库进行人像识别；当特征库数据量较大时，需要从海量数据中进行人像识别等情况均可能降低人像识别的效率的问题，实现提高人像识别的效率和通用性的效果。
Embodiments of the present invention disclose a method, device, electronic device and storage medium for identifying a person. The method includes: acquiring data information of target portrait data; wherein the data information includes portrait information and position information; determining a first target portrait database from at least one candidate portrait database according to the portrait information and/or the position information , and determine the database comparison sequence of the first target portrait database; determine the portrait recognition result according to the portrait information, the first target portrait database, and the database comparison sequence. By running the technical solutions provided by the embodiments of the present invention, it can be solved that when there are multiple application scenarios, it is necessary to perform portrait recognition for the feature databases of respective scenarios; Recognition and other situations may reduce the efficiency of portrait recognition, and achieve the effect of improving the efficiency and versatility of portrait recognition.
Description
Technical Field
The present invention relates to computer technologies, and in particular, to a method and an apparatus for identifying a portrait, an electronic device, and a storage medium.
Background
The traditional portrait recognition method is that large-scale portrait features in practical application scenes are extracted according to industry experience or expert experience, and then the extracted features are summarized, summarized and statistically processed to form a specific feature library for subsequent feature comparison, namely, analysis and optimization are performed only aiming at specific problems of a single application scene. When the application scenes are various, the portrait identification is needed to be carried out respectively aiming at the feature libraries of the respective scenes; when the data size of the feature library is large, the efficiency of face recognition may be reduced in cases where face recognition needs to be performed from mass data.
Disclosure of Invention
The embodiment of the invention provides a portrait recognition method and device, electronic equipment and a storage medium, and aims to improve the efficiency and universality of portrait recognition.
In a first aspect, an embodiment of the present invention provides a portrait identification method, where the method includes:
acquiring data information of target portrait data; wherein the data information comprises portrait information and location information;
determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database;
and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
In a second aspect, an embodiment of the present invention further provides a portrait recognition apparatus, where the apparatus includes:
the data information acquisition module is used for acquiring data information of the target portrait data; wherein the data information comprises portrait information and location information;
the database sequence determining module is used for determining a first target human database from at least one candidate human database according to the portrait information and/or the position information and determining a database comparison sequence of the first target human database;
and the portrait recognition result determining module is used for determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the portrait recognition method as described above.
In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the portrait recognition method as described above.
The embodiment of the invention obtains the data information of the target portrait data; wherein the data information comprises portrait information and location information; determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database; and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence. When multiple application scenes exist, the face recognition needs to be carried out on the feature libraries of the respective scenes; when the data volume of the feature library is large, the face recognition efficiency can be reduced under the conditions that the face recognition needs to be carried out from mass data, and the like, and the effect of improving the face recognition efficiency and the universality is achieved.
Drawings
Fig. 1 is a flowchart of a human image recognition method according to an embodiment of the present invention;
fig. 2 is a flowchart of a portrait recognition method according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a portrait recognition system according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a portrait recognition apparatus according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a portrait recognition method according to an embodiment of the present invention, where the present embodiment is applicable to a situation of portrait recognition in a multi-application scenario, and the method may be executed by a portrait recognition apparatus according to an embodiment of the present invention, and the apparatus may be implemented in a software and/or hardware manner. Referring to fig. 1, the method for identifying a person provided in this embodiment includes:
The target portrait data is obtained by processing all acquired portrait data according to a preset algorithm.
The portrait information of the target portrait data is information related to the portrait in the target portrait data, for example, a portrait image.
The position information of the target portrait data is position information obtained when the target portrait data is obtained, and may be determined according to a scene where a data acquisition device of the target portrait data is located. The physical area may be a specific website, a branch, a building, etc., and the organization information may be a preset number corresponding to an organization, etc., which is not limited in this embodiment.
In this embodiment, optionally, before acquiring the data information of the target portrait data, the method further includes:
acquiring portrait data to be processed, and performing preset image processing on the portrait data to be processed to obtain target portrait data; wherein the preset image processing includes at least one of image deduplication processing and image screening processing.
The acquisition of the portrait data to be processed can be shooting videos through data acquisition equipment, and the portrait data in the video stream is extracted to serve as the portrait data to be processed. The image deduplication processing can be to judge the quality score of the portrait data of the same object, optimize deduplication, and use the portrait data with the highest quality as the target portrait data of the object, so that the subsequent repeated processing of multiple portrait data of the same object is avoided, the computing resources are wasted, and the accuracy of subsequent portrait identification is improved by obtaining the portrait data with the best quality.
The image screening processing can be a preset white list, the portrait data in the white list is filtered, the filtered target portrait data is obtained, the white list can contain hall service personnel, internal staff, unscheduled visitor personnel and the like, and the embodiment does not limit the situation, so that the consumption of computing resources is avoided.
The first target person database is determined from at least one candidate person database according to the person image information and/or the location information, where the person database may store the person image characteristics, which is not limited in this embodiment. The first target human database may be a human database conforming to a current human image data acquisition scene, and for example, if the position information is a position of a bank at the time of human image data acquisition, it may be determined that the first target human database is a human database corresponding to a position a. And if the portrait information is the portrait image quality score, determining the portrait information as a first target portrait database according to a portrait database corresponding to the image quality score.
The first target human database is determined according to the position information and the human image information, the human databases corresponding to the position information and the human image information can be directly used as the first target human database, the human databases corresponding to the position information and the human image information can be ranked according to preset weights of the position information and the human image information, the human database which is ranked in front is used as the first target human database, and the first target human database is not limited by the embodiment.
The database comparison order of the first target human database is determined according to the portrait information and/or the position information, whether a specific database comparison order is preset at the position or not can be determined according to the position information, if yes, the specific comparison order is used as the database comparison order, and if not, the preset comparison order is used as the database comparison order.
For example, the preset comparison order may be the order of comparing according to the data size of the first target human database, according to the administrative level size of the database, and the like, which is not limited in this embodiment; the specific comparison sequence may be a comparison sequence different from a preset comparison sequence, for example, the preset sequence is a human database of a city area to be compared first, and then a human database of a website is compared, and the specific comparison sequence may be a human database of a website to be compared first, and then a human database of a city area to be compared, so that different comparison sequences are set according to different application scenarios.
And step 130, determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
According to the database comparison sequence, the first target human database is sequentially compared with the human image information, and a corresponding human image recognition result can be obtained, wherein the human image recognition result can be one of no matching object, one matching object and a plurality of matching objects, and the embodiment does not limit the result.
The comparison mode can be that the portrait characteristic value in the portrait information is extracted and searched in the first target portrait database, whether the first target portrait database has the portrait characteristics meeting the condition that the similarity exceeds a preset threshold value or not is determined, and if the first target portrait database has the portrait characteristics, the information corresponding to the portrait characteristics is used as the portrait recognition result.
Optionally, after the portrait recognition result is determined to be recognized, subsequent portrait recognition result pushing and business process processing are performed according to a preset business rule, for example, when the portrait recognition result is successful, the recognized portrait related information is pushed to a specified platform, and the portrait related information may be a portrait image, corresponding identity information, and the like, which is not limited in this embodiment.
According to the technical scheme provided by the embodiment, data information of target portrait data is acquired; wherein the data information comprises portrait information and location information; determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database; and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence. When multiple scenes and/or a large number of users to be identified appear, a database to be compared and a database comparison sequence can be determined according to portrait information and/or position information, and therefore the range to be compared and identified is divided. The method has the advantages that the situation that each scene only corresponds to the respective fixed human database is avoided, when the human databases used in different application scenes are overlapped, human database resources are wasted, meanwhile, the respective associated human databases are determined for the different application scenes, and the efficiency, the universality and the accuracy of follow-up human image recognition are improved.
Example two
Fig. 2 is a flowchart of a portrait session identification method according to a second embodiment of the present invention, and the technical solution is supplementary explained for a process of determining a portrait session identification result according to the portrait session information, the first target portrait database, and the database comparison order. Compared with the scheme, the scheme is specifically optimized in that the step of determining the portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence comprises the following steps:
determining a current human database from the first target human database according to the database comparison sequence, and comparing the human image information with the current human database to obtain at least one current human image comparison result;
judging whether the current portrait comparison result meeting a preset first threshold exists or not;
if not, determining a new current human image database from the first target human image database according to the database comparison sequence, and comparing the human image information with the new current human image database to obtain a new current human image comparison result;
and determining the portrait recognition result according to the new current portrait comparison result. Specifically, the flowchart of the portrait recognition method is shown in fig. 2:
And determining a current human database from the first target human database according to the database comparison sequence, wherein the current human database is a human database currently subjected to comparison operation. Illustratively, the database comparison order is database C, database D, and database A. Then database C is the current human database at the beginning of the compare operation.
And comparing the portrait information with the current portrait database to obtain at least one current portrait comparison result, wherein the current portrait comparison result can be a result of comparing each portrait characteristic value in the portrait information with all characteristic values in the current portrait database.
And 240, judging whether the current portrait comparison result meeting a preset first threshold exists.
And judging whether each current portrait comparison result meets a preset first threshold value, if so, indicating that the current portrait identification result is successful, and obtaining a corresponding matching object. The preset first threshold is a contrast threshold corresponding to the current human database.
All successful current portrait recognition results can be used as final portrait recognition results, which is not limited in this embodiment.
And 250, if not, determining a new current human database from the first target human database according to the database comparison sequence, and comparing the human image information with the new current human database to obtain a new current human image comparison result.
If not, the fact that no matched portrait feature data exists in the current portrait database is indicated, and the next portrait database in the database comparison sequence is determined to serve as the new current portrait database.
And comparing the portrait information with the new current portrait database according to the mode to obtain a new current portrait comparison result.
And step 260, determining the portrait recognition result according to the new current portrait comparison result.
And if the new current portrait comparison results are all failed, continuously determining a next human database of the new current human database according to the database comparison sequence, performing corresponding comparison operation, and circulating the steps until the comparison of all the databases in the first target human database is finished. According to the actual comparison condition, the new current portrait comparison result which is successfully compared can be used as the portrait recognition result, or the new current portrait comparison results which are all unsuccessfully compared can be used as the portrait recognition result.
In this embodiment, optionally, the determining the portrait recognition result according to the new current portrait comparison result includes:
judging whether all the new current portrait comparison results meet a preset second threshold corresponding to each new current portrait comparison result;
if the first target human database is not satisfied, determining a second target human database except the first target human database from the candidate human database according to a preset sequence among human databases;
and determining the portrait recognition result according to the portrait information and the second target portrait database.
And judging whether all the new current portrait comparison results meet preset second threshold values corresponding to the new current portrait comparison results, wherein the preset second threshold values of the new current portrait comparison results are determined according to the human database corresponding to the new current portrait comparison results. And determining a proper contrast threshold according to the attributes of different human database, and improving the accuracy of subsequent human image recognition.
And if all the comparison results do not meet the respective threshold values, the comparison results of the first target human database are all failed. At this time, a second target human database except the first target human database is determined from the candidate human database according to the preset sequence between the human databases, illustratively, the first target human database is a website-level human image feature library and a branch-level human image feature library, and then the second target human database may be a full-line-level human image feature library, that is, the preset sequence may be a rank sequence between human databases, and this embodiment does not limit this. Therefore, when the face recognition result corresponding to the first target face database is failure, comparison operation can be further performed, the comparison range is expanded, and accuracy of face recognition is improved.
In this embodiment, optionally, after determining whether the current portrait comparison result meeting the preset first threshold exists, the method further includes:
if yes, determining result sorting of the current portrait comparison result;
determining a target portrait comparison result from the current portrait comparison results according to the result sorting;
judging whether the target portrait comparison result meeting a preset identification threshold exists or not;
and if so, determining the target portrait comparison result as the portrait identification result.
If there are a plurality of current portrait comparison results meeting the preset first threshold, determining the result ranking of the current portrait comparison results, where the determination mode may be ranking from high to low according to the similarity degree, and this embodiment does not limit this. The current portrait comparison result with the highest ranking rank may be used as the target portrait comparison result, which is not limited in this embodiment.
And comparing the target portrait comparison result with a preset identification threshold, and determining the target portrait comparison result as a portrait identification result if the target portrait comparison result is greater than or equal to the preset identification threshold. And judging whether the most similar portrait comparison result meets the portrait identification requirement or not, and avoiding a plurality of identification results, thereby improving the accuracy of portrait identification.
The embodiment of the invention judges whether the current portrait comparison result meeting a preset first threshold exists or not; if not, determining a new current human image database from the first target human image database according to the database comparison sequence, and comparing the human image information with the new current human image database to obtain a new current human image comparison result; and determining a portrait recognition result according to the new current portrait comparison result, namely, continuing to perform comparison operation according to the database comparison sequence when the portrait comparison is unsuccessful, thereby improving the portrait recognition efficiency.
EXAMPLE III
In order to make it more clear to those skilled in the art to understand the present invention, the present application further provides a portrait recognition system, fig. 3 is a schematic structural diagram of a portrait recognition system provided in a third embodiment of the present invention, as shown in fig. 3:
wherein, leading software and hardware integration module of image acquisition: the module is respectively arranged in a network point, a branch business hall and each office building according to different levels and is used for receiving RTSP video streams transmitted by the portrait acquisition equipment, and filtering and outputting portrait encrypted data by the terminal.
Wherein, the leading software and hardware integration module of image acquisition includes:
the video stream processing module: the method has the functions of processing the video stream and extracting the portrait, can position the portrait in the video stream, judges the portrait quality score, optimizes the duplication removal and outputs the portrait data with the highest quality.
The device management and encryption module: the event platform management module is used for pre-filtering the white list portrait photos pre-embedded by the event platform management module, the white list comprises hall service personnel, internal working personnel and unscheduled visitor personnel, and outputting portrait encrypted data
The event platform front module: the system is used for realizing bidirectional communication and data interaction among the modules.
Wherein, the event platform front module includes:
the preposed data receiving module: the image acquisition device is used for receiving encrypted portrait data on the image acquisition front software and hardware integrated module.
The preposed data pushing module: the system is used for forming a SOAP message according to the position information such as the physical area where the image acquisition equipment is located in the image acquisition front-end software and hardware integrated module, the mechanism information and the like, and sending the SOAP message to the event platform core module.
The event platform management module: the system is used for monitoring and maintaining the running state and the application version of the image acquisition preposed software and hardware integrated module through an equipment management interface and a data synchronization function deployed on the module.
Wherein, the event platform management module includes:
the device management and data synchronization module: the system is used for issuing the white list data in batches through TCP communication with the image acquisition prepositive software and hardware integrated module.
An equipment monitoring module: the method is used for monitoring and maintaining the running state and the application version of the image acquisition front-mounted software and hardware integrated module.
N portrait comparison service module: the method is used for storing mass portrait characteristic values and decrypting portrait encrypted data.
Wherein, the portrait comparison service module includes:
portrait feature library management module: the portrait characteristic library is composed of a large number of portrait characteristics which are stored in the module in a multi-level mode according to actual physical areas and scene logic rules.
1: n people image comparison module: the module decrypts the sent portrait data, extracts characteristic values and searches in a characteristic library, and screens and judges whether the portrait data is accurately identified according to a set threshold value.
An event platform core module: the system is used for comprehensively analyzing and forming a first target human database and database comparison sequence according to the physical region classification of the human database, business logic rules and the like, interacting with the human image comparison service module to obtain a recognition result, and finishing subsequent data pushing.
Wherein, the event platform core module includes:
multi-scenario application and accurate identification module (not shown in the figure): the module adopts a micro-service architecture, a front-end module of a docking event platform and a face recognition comparison service module, and has the functions of physical area positioning, scene information division, multi-scene sharing, potential face library generation, white list filtering and timing duplicate removal, recognition result pushing and the like. The method has the main functions of generating a potential portrait feature library and a comparison priority sequence according to an actual physical region and a scene rule, reducing the functions to be compared, improving the recognition quality of the multi-scene portrait and distinguishing the comparison priority and the comparison range of different scenes.
By establishing the event platform core module, when a large number of multi-scene clients appear, a specific priority and a comparison rule can be formed according to the actual physical area and scene characteristics, the range to be compared and identified is divided, and an identification range analysis model is constructed. The accuracy of multi-scene recognition is improved by inputting the rule model, and 1: the limitation of the N technical application scheme promotes the intelligent level in the field of banking and internal management scenes, and reduces the labor cost and the management cost.
Example four
Fig. 4 is a schematic structural diagram of a portrait recognition apparatus according to a fourth embodiment of the present invention. The device can be realized by hardware and/or software, can execute the portrait recognition method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. As shown in fig. 4, the apparatus includes:
a data information obtaining module 410, configured to obtain data information of the target portrait data; wherein the data information comprises portrait information and location information;
a database order determining module 420, configured to determine a first target person database from at least one candidate person database according to the portrait information and/or the location information, and determine a database comparison order of the first target person database;
and the portrait recognition result determining module 430 is configured to determine a portrait recognition result according to the portrait information, the first target portrait database, and the database comparison sequence.
According to the technical scheme provided by the embodiment, data information of target portrait data is acquired; wherein the data information comprises portrait information and location information; determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database; and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence. When multiple scenes and/or a large number of users to be identified appear, a database to be compared and a database comparison sequence can be determined according to portrait information and/or position information, and therefore the range to be compared and identified is divided. The method has the advantages that each scene only corresponds to the respective fixed human database, when the human databases used in different application scenes are overlapped, human database resources are wasted, meanwhile, the respective associated human databases are determined for the different application scenes, and the efficiency and the accuracy of subsequent human image recognition are improved.
On the basis of the above technical solutions, optionally, the portrait recognition result determining module includes:
a current portrait comparison result obtaining unit, configured to determine a current portrait database from the first target portrait database according to the database comparison sequence, compare the portrait information with the current portrait database, and obtain at least one current portrait comparison result;
the current portrait comparison result judging unit is used for judging whether the current portrait comparison result meeting a preset first threshold exists or not;
a new current portrait comparison result obtaining unit, configured to determine a new current portrait database from the first target portrait database according to the database comparison sequence if the current portrait comparison result determination unit determines that the current portrait comparison result is negative, compare the portrait information with the new current portrait database, and obtain a new current portrait comparison result;
and the first portrait recognition result determining unit is used for determining the portrait recognition result according to the new current portrait comparison result.
On the basis of the foregoing technical solutions, optionally, the first person recognition result determining unit includes:
a new current portrait comparison result judgment subunit, configured to judge whether all the new current portrait comparison results satisfy preset second thresholds corresponding to the new current portrait comparison results;
a second target human database determining subunit, configured to determine, if the new current human image comparison result judgment subunit judges that none of the candidate human databases is satisfied, a second target human database other than the first target human database from the candidate human database according to a preset sequence between human databases;
and the portrait recognition result determining subunit is used for determining the portrait recognition result according to the portrait information and the second target portrait database.
On the basis of the above technical solutions, optionally, the apparatus further includes:
the result ordering determining unit is used for determining the result ordering of the current portrait comparison result after the new current portrait comparison result judging subunit;
the target portrait comparison result determining unit is used for determining a target portrait comparison result from the current portrait comparison results according to the result sorting;
the target portrait comparison result judging unit is used for judging whether the target portrait comparison result meeting a preset identification threshold exists or not;
and the second portrait identification result determining unit is used for determining the target portrait comparison result as the portrait identification result if the target portrait comparison result judging unit judges that the target portrait comparison result is positive.
On the basis of the above technical solutions, optionally, the apparatus further includes:
the data image processing module is used for acquiring portrait data to be processed before the data information acquisition module and carrying out preset image processing on the portrait data to be processed to obtain the target portrait data; wherein the preset image processing includes at least one of image deduplication processing and image screening processing.
EXAMPLE five
Fig. 5 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention, as shown in fig. 5, the electronic device includes a processor 50, a memory 51, an input device 52, and an output device 53; the number of the processors 50 in the electronic device may be one or more, and one processor 50 is taken as an example in fig. 5; the processor 50, the memory 51, the input device 52 and the output device 53 in the electronic apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 5.
The memory 51 is used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the portrait recognition method in the embodiment of the present invention. The processor 50 executes various functional applications and data processing of the electronic device by executing software programs, instructions and modules stored in the memory 51, that is, implements the above-described portrait recognition method.
The memory 51 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 51 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 51 may further include memory located remotely from the processor 50, which may be connected to the electronic device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
EXAMPLE six
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for face recognition, the method including:
acquiring data information of target portrait data; wherein the data information comprises portrait information and location information;
determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database;
and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the portrait recognition method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the portrait recognition apparatus, the included units and modules are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A face recognition method, comprising:
acquiring data information of target portrait data; wherein the data information comprises portrait information and location information;
determining a first target human database from at least one candidate human database according to the portrait information and/or the location information, and determining a database comparison order of the first target human database;
and determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
2. The method of claim 1, wherein determining a portrait session based on the portrait information, the first target people database, and the database comparison order comprises:
determining a current human database from the first target human database according to the database comparison sequence, and comparing the human image information with the current human database to obtain at least one current human image comparison result;
judging whether the current portrait comparison result meeting a preset first threshold exists or not;
if not, determining a new current human image database from the first target human image database according to the database comparison sequence, and comparing the human image information with the new current human image database to obtain a new current human image comparison result;
and determining the portrait recognition result according to the new current portrait comparison result.
3. The method of claim 2, wherein determining the portrait recognition result from the new current portrait comparison result comprises:
judging whether all the new current portrait comparison results meet a preset second threshold corresponding to each new current portrait comparison result;
if the first target human database is not satisfied, determining a second target human database except the first target human database from the candidate human database according to a preset sequence among human databases;
and determining the portrait recognition result according to the portrait information and the second target portrait database.
4. The method according to claim 2, after determining whether there is the current portrait comparison result satisfying a preset first threshold, further comprising:
if yes, determining result sorting of the current portrait comparison result;
determining a target portrait comparison result from the current portrait comparison results according to the result sorting;
judging whether the target portrait comparison result meeting a preset identification threshold exists or not;
and if so, determining the target portrait comparison result as the portrait identification result.
5. The method of claim 1, further comprising, prior to obtaining data information for the target portrait data:
acquiring portrait data to be processed, and performing preset image processing on the portrait data to be processed to obtain target portrait data; wherein the preset image processing includes at least one of image deduplication processing and image screening processing.
6. A face recognition apparatus, comprising:
the data information acquisition module is used for acquiring data information of the target portrait data; wherein the data information comprises portrait information and location information;
the database sequence determining module is used for determining a first target human database from at least one candidate human database according to the portrait information and/or the position information and determining a database comparison sequence of the first target human database;
and the portrait recognition result determining module is used for determining a portrait recognition result according to the portrait information, the first target portrait database and the database comparison sequence.
7. The apparatus of claim 6, wherein the face recognition result determining module comprises:
a current portrait comparison result obtaining unit, configured to determine a current portrait database from the first target portrait database according to the database comparison sequence, compare the portrait information with the current portrait database, and obtain at least one current portrait comparison result;
the current portrait comparison result judging unit is used for judging whether the current portrait comparison result meeting a preset first threshold exists or not;
a new current portrait comparison result obtaining unit, configured to determine a new current portrait database from the first target portrait database according to the database comparison sequence if the current portrait comparison result determination unit determines that the current portrait comparison result is negative, compare the portrait information with the new current portrait database, and obtain a new current portrait comparison result;
and the first portrait recognition result determining unit is used for determining the portrait recognition result according to the new current portrait comparison result.
8. The apparatus of claim 7, wherein the first human recognition result determining unit comprises:
a new current portrait comparison result judgment subunit, configured to judge whether all the new current portrait comparison results satisfy preset second thresholds corresponding to the new current portrait comparison results;
a second target human database determining subunit, configured to determine, if the new current human image comparison result judgment subunit judges that none of the candidate human databases is satisfied, a second target human database other than the first target human database from the candidate human database according to a preset sequence between human databases;
and the portrait recognition result determining subunit is used for determining the portrait recognition result according to the portrait information and the second target portrait database.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the portrait recognition method of any of claims 1-5.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the portrait recognition method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110947338.8A CN113688712A (en) | 2021-08-18 | 2021-08-18 | A portrait recognition method, device, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110947338.8A CN113688712A (en) | 2021-08-18 | 2021-08-18 | A portrait recognition method, device, electronic device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113688712A true CN113688712A (en) | 2021-11-23 |
Family
ID=78580702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110947338.8A Pending CN113688712A (en) | 2021-08-18 | 2021-08-18 | A portrait recognition method, device, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113688712A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105760461A (en) * | 2016-02-04 | 2016-07-13 | 上海卓易科技股份有限公司 | Automatic album establishing method and device |
| CN108885698A (en) * | 2018-07-05 | 2018-11-23 | 深圳前海达闼云端智能科技有限公司 | Face identification method, device and server |
| CN109086669A (en) * | 2018-06-29 | 2018-12-25 | 汉王科技股份有限公司 | Recognition of face auth method, device, electronic equipment |
| CN109523591A (en) * | 2018-11-21 | 2019-03-26 | 北京锐安科技有限公司 | Processing method, device, equipment and the storage medium of picture |
| CN109919091A (en) * | 2019-03-06 | 2019-06-21 | 广州佳都数据服务有限公司 | Face safety inspection method, device and electronic equipment based on dynamic white list |
| CN110799972A (en) * | 2017-07-21 | 2020-02-14 | 北京市商汤科技开发有限公司 | Dynamic human face image storage method and device, electronic equipment, medium and program |
| CN112241696A (en) * | 2020-09-28 | 2021-01-19 | 深圳市商汤科技有限公司 | Image processing method and device, electronic device and storage medium |
| CN112488078A (en) * | 2020-12-23 | 2021-03-12 | 浙江大华技术股份有限公司 | Face comparison method and device and readable storage medium |
| CN113128437A (en) * | 2021-04-27 | 2021-07-16 | 北京市商汤科技开发有限公司 | Identity recognition method and device, electronic equipment and storage medium |
-
2021
- 2021-08-18 CN CN202110947338.8A patent/CN113688712A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105760461A (en) * | 2016-02-04 | 2016-07-13 | 上海卓易科技股份有限公司 | Automatic album establishing method and device |
| CN110799972A (en) * | 2017-07-21 | 2020-02-14 | 北京市商汤科技开发有限公司 | Dynamic human face image storage method and device, electronic equipment, medium and program |
| CN109086669A (en) * | 2018-06-29 | 2018-12-25 | 汉王科技股份有限公司 | Recognition of face auth method, device, electronic equipment |
| CN108885698A (en) * | 2018-07-05 | 2018-11-23 | 深圳前海达闼云端智能科技有限公司 | Face identification method, device and server |
| CN109523591A (en) * | 2018-11-21 | 2019-03-26 | 北京锐安科技有限公司 | Processing method, device, equipment and the storage medium of picture |
| CN109919091A (en) * | 2019-03-06 | 2019-06-21 | 广州佳都数据服务有限公司 | Face safety inspection method, device and electronic equipment based on dynamic white list |
| CN112241696A (en) * | 2020-09-28 | 2021-01-19 | 深圳市商汤科技有限公司 | Image processing method and device, electronic device and storage medium |
| CN112488078A (en) * | 2020-12-23 | 2021-03-12 | 浙江大华技术股份有限公司 | Face comparison method and device and readable storage medium |
| CN113128437A (en) * | 2021-04-27 | 2021-07-16 | 北京市商汤科技开发有限公司 | Identity recognition method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 林增敏等: "融合人脸跟踪和聚类的人脸图像去重方法", 《计算机科学》, no. 2020, 15 November 2020 (2020-11-15), pages 1 - 5 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110232373B (en) | Face clustering method, device, equipment and storage medium | |
| WO2019223313A1 (en) | Personnel file establishment method and apparatus | |
| US11727053B2 (en) | Entity recognition from an image | |
| JP6973876B2 (en) | Face recognition methods, face recognition devices and computer programs that execute face recognition methods | |
| CN110147722A (en) | A kind of method for processing video frequency, video process apparatus and terminal device | |
| US8995729B2 (en) | Accelerated comparison using scores from coarse and fine matching processes | |
| CN106203333A (en) | Face identification method and system | |
| WO2019080908A1 (en) | Image processing method and apparatus for implementing image recognition, and electronic device | |
| CN110298240B (en) | Automobile user identification method, device, system and storage medium | |
| US11816117B2 (en) | Data retrieval method and apparatus | |
| CN110929125A (en) | Search recall method, apparatus, device and storage medium thereof | |
| WO2020125100A1 (en) | Image search method, apparatus, and device | |
| CN109711298B (en) | Method and system for efficient face characteristic value retrieval based on faiss | |
| CN111061837A (en) | Topic identification method, device, equipment and medium | |
| CN106844571A (en) | Recognize method, device and the computing device of synonym | |
| KR20210033940A (en) | How to Train Neural Networks for Human Facial Recognition | |
| CN111368867A (en) | Archive classification method and system and computer readable storage medium | |
| CN110929081A (en) | Picture query method, computer equipment and storage medium | |
| CN111651625A (en) | Image retrieval method, device, electronic device and storage medium | |
| JP5430636B2 (en) | Data acquisition apparatus, method and program | |
| CN111444362B (en) | Malicious picture interception method, device, equipment and storage medium | |
| CN112380169A (en) | Storage device, data processing method, device, apparatus, medium, and system | |
| CN109783678B (en) | Image searching method and device | |
| CN111950352A (en) | Hierarchical face clustering method, system, device and storage medium | |
| CN113688712A (en) | A portrait recognition method, device, electronic device 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 |