CN110147455A - A kind of face matching retrieval device and method - Google Patents
A kind of face matching retrieval device and method Download PDFInfo
- Publication number
- CN110147455A CN110147455A CN201710760859.6A CN201710760859A CN110147455A CN 110147455 A CN110147455 A CN 110147455A CN 201710760859 A CN201710760859 A CN 201710760859A CN 110147455 A CN110147455 A CN 110147455A
- Authority
- CN
- China
- Prior art keywords
- index
- facial image
- face
- face characteristic
- database
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000001815 facial effect Effects 0.000 claims abstract description 103
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 230000015654 memory Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000004048 modification Effects 0.000 claims description 6
- 238000012986 modification Methods 0.000 claims description 6
- 230000032258 transport Effects 0.000 claims 1
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 3
- 210000000887 face Anatomy 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—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
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Library & Information Science (AREA)
- Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of faces to match retrieval device and method, is related to the fields such as image procossing, matching, retrieval.Face matching retrieval device disclosed by the invention, compare server, secondary relatively server including at least index, wherein: index compares server, for the face characteristic of facial image to be retrieved to be compared with face characteristic in the index database pre-established, obtain corresponding index, wherein, respectively index respectively corresponds a kind of face characteristic in the index database；Secondary relatively server, for comparing in all face characteristics that the index that server obtains stores in corresponding region in the facial feature database pre-established in the index, the face characteristic of facial image to be retrieved is retrieved, comparison result is got.
Description
Technical field
The present invention relates to the fields such as image procossing, matching, retrieval, more particularly to one kind to carry out fastly in magnanimity facial image
The scheme of speed matching retrieval.
Background technique
With image procossing, the development of mode identification technology, recognition of face, compare using more and more extensive.Especially
With the development of video big data, it is frequently necessary to one facial image of input in police field, then in the face database of magnanimity
Find out matching face.The face database of police field usually contains millions of or even up to ten million facial images.How from
Face matching is fast and accurately carried out in the face database of magnanimity, retrieves the problem of becoming current urgent need to resolve.
The method for carrying out Rapid matching in magnanimity face at present, generallys use the parallel processing manners such as distribution, passes through increasing
Add machine, the face alignment task of magnanimity is broken the whole up into parts, is evenly distributed to every machine and handles up, then remerge each ratio
To result.
Summary of the invention
The problem of provided herein is a kind of faces to match retrieval device and method, can solve existing face retrieval low efficiency.
Disclosed herein is a kind of faces to match retrieval device, compares server, secondary relatively server including at least index,
Wherein:
Index compares server, for by the face characteristic of facial image to be retrieved and people in the index database that pre-establishes
Face feature is compared, and obtains corresponding index, wherein respectively index respectively corresponds a kind of face characteristic in the index database；
Secondary relatively server, for comparing index that server obtains in the face characteristic pre-established in the index
In all face characteristics stored in corresponding region in database, the face characteristic of facial image to be retrieved is examined
Rope gets comparison result.
Optionally, above-mentioned apparatus, further includes:
Features Management server, for being managed operation to the facial feature database；
The management operation includes appointing one or more operations as follows: newly-increased facial image deletes facial image, modification face
Image, inquiry facial image.
Optionally, in above-mentioned apparatus, the index compares server, for when needing the newly-increased facial image being put in storage,
Newly-increased facial image to be put in storage is compared with the face characteristic in the index database, is indexed accordingly；
The Features Management server, for according to obtained index, by newly-increased facial image to be put in storage store to
In the facial feature database in the corresponding region of the index.
There is disclosed herein a kind of faces to match search method, comprising:
The face characteristic of facial image to be retrieved is compared with face characteristic in the index database pre-established, is obtained
Corresponding index；
In the facial feature database pre-established, the face stored in region corresponding with obtained index is special
In sign, the face characteristic of facial image to be retrieved is retrieved, comparison result is got；
Wherein, respectively index respectively corresponds a kind of face characteristic in the index database.
Optionally, the above method, by the face characteristic of facial image to be retrieved and people in the index database that pre-establishes
Before face feature is compared, further includes:
All kinds of face characteristics are found out from the facial image obtained in advance according to clustering method as index, and establish index
Library.
Optionally, in the above method, the face characteristic by facial image to be retrieved and the index database pre-established
Middle face characteristic is compared, and obtains corresponding index and includes:
It, will be in the face characteristic of facial image to be retrieved and the index database pre-established using the algorithm for calculating distance
Face characteristic is compared, using similarity in comparison result setting range index as corresponding index；
Wherein, the used algorithm for calculating distance generates cluster calculation used by index with when pre-establishing index database
The algorithm of distance is identical.
Optionally, described in the facial feature database pre-established in the above method, it is opposite with obtained index
In the face characteristic stored in the region answered, the face characteristic of facial image to be retrieved is retrieved, comprising:
It is right in the face characteristic stored in region corresponding with obtained index using the algorithm for calculating distance
The face characteristic of facial image to be retrieved is retrieved；
Wherein, the used algorithm for calculating distance generates cluster calculation used by index with when pre-establishing index database
The algorithm of distance is identical.
Optionally, the above method, further includes:
Operation is managed to the facial feature database, the management operation includes appointing one or more operations as follows:
Newly-increased facial image deletes facial image, modification facial image, inquiry facial image.
Optionally, in the above method, when operating the management that the facial feature database carries out to increase facial image newly,
Include:
Newly-increased facial image to be put in storage is compared with the face characteristic in the index database, obtains corresponding rope
Draw；
According to obtained index, newly-increased facial image to be put in storage is stored into the facial feature database and is somebody's turn to do
It indexes in corresponding region.
There is disclosed herein a kind of retrieval device, including memory, processor and it is stored on the memory and can be
The computer program run on the processor, which is characterized in that the processor is realized such as when executing the computer program
The processing of the upper method.
Compared with prior art, technical scheme has the advantages that
It is indexed by establishing, magnanimity face characteristic is first carried out to rough classification (the corresponding a kind of face feature of i.e. one index),
In this way, first determining the type of face characteristic in face retrieval, so that it may matched range will be needed to be substantially reduced, promote identification
Speed；
In addition, by establishing index, it is only necessary to compare, subtract with a limited number of representative face characteristics (indexing)
Lack the matched number of needs, to save computing resource, reduces hardware cost；
The promotion of recognition speed reduces waiting time when user carries out face retrieval in magnanimity face database, improves
The working efficiency of user experience and user.
Detailed description of the invention
Fig. 1 is that face Rapid matching retrieves apparatus structure schematic diagram in the embodiment of the present invention；
Fig. 2 is face Rapid matching search method flow chart in the embodiment of the present invention；
Fig. 3 is that magnanimity face Rapid matching retrieves apparatus structure schematic diagram in alternative embodiment of the present invention；
Fig. 4 is the process flow diagram that index compares server in alternative embodiment of the present invention；
Fig. 5 is that index flow chart is generated in alternative embodiment of the present invention；
Fig. 6 is face characteristic data loading flow chart in alternative embodiment of the present invention；
Fig. 7 is that face characteristic retrieves entire flow figure in alternative embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with specific embodiment pair
Technical solution of the present invention is described in further detail.It should be noted that in the absence of conflict, embodiments herein and
Feature in embodiment can be arbitrarily combined with each other.
Present inventor proposes, constructs face index in the face database of magnanimity by way of cluster, is carrying out people
When face is retrieved, face to be retrieved and a limited number of indexes are compared first, then needed using index acquisition secondary
Then the face range of comparison carries out Secondary Match in a small range again, to improve face retrieval efficiency.
Based on above-mentioned thought, the present embodiment provides a kind of retrieval dresses that Rapid matching is completely carried out in magnanimity face
Set, as shown in Figure 1, specifically include that index compare server, it is secondary relatively server,
Index compares server, by matching face characteristic to be processed with the face characteristic in index database, obtains
Take index corresponding to face characteristic to be processed.Wherein, respectively index respectively corresponds a kind of face characteristic in index database.Herein
A kind of face characteristic is the face characteristic of a kind of representative feature.The characteristic features being referred to herein can be people
The locality characteristic of face, such as eyes, mouth, shape of face feature, are also possible to the assemblage characteristic of face, such as the various parts of face
Any combination etc. between property feature.As long as magnanimity face can be divided into multiclass face i.e. according to different characteristic features
It can.In practical application, it can be clustered from magnanimity face using clustering method and come out the face with different characteristic features.Two
Secondary relatively server, it is corresponding in facial feature database according to the calculated index of index server in face recognition process
Face within the scope of (i.e. with the face characteristic stored in the corresponding region of the index) carry out face characteristic it is secondary compared with, inspection
Rope.
In practical application, index compares server may following two application scenarios in the present apparatus:
1) it when face characteristic data loading, is compared, is obtained similar to face characteristic to be put in storage by face characteristic first
Highest index is spent, the corresponding subregion of facial feature database is then stored according to index；
2) when face retrieval, the face characteristic in face characteristic and index database to be retrieved is compared, obtains and meets
One or several indexes of certain similarity.In processing, index compares server and loads index from index data base first
It caches or to calculating for indexing the specialized hardware compared, such as FPGA, GPU；Then by extracted facial image to be processed
Feature be compared to obtain corresponding index with the feature in index database.
It is worth noting that, index compare server face to be retrieved is carried out in index database when feature compares calculating away from
From algorithm need and establish index database to generate index when used cluster calculation with a distance from mode and secondary relatively service
Device calculates distance algorithm when carrying out secondary compare to face to be retrieved is consistent.
Wherein, index database is stored with all indexes employed in this system, and each index being referred to herein is right respectively
(i.e. same class face is special for the one group of representative face characteristic that should cluster out in the face of magnanimity by clustering method
Sign).
Facial feature database can store all face characteristics employed in this system according to index partition,
Wherein, facial feature database is divided into the corresponding region with each index, respectively indexes and is stored with the rope in corresponding region
Draw all face characteristics of corresponding a kind of face characteristic.For example, with each face characteristic for indexing and being stored in corresponding region, including
All face characteristics of the similarity of representative face characteristic corresponding with the index in setting similar threshold value.
In addition, above-mentioned apparatus can also include Features Management server, for managing face characteristic.It may include pair
The increasing of facial feature database deletes, changes, looking into, obtaining the operation such as corresponding face characteristic data by index.And secondary relatively service
Device carries out quadratic search to facial image to be retrieved, can be through Features Management server and realizes.For example, secondary ratio
It can be when carrying out quadratic search compared with server, it will be corresponding with index in facial feature database by Features Management server
Whole face characteristic data are real-time or are pre-loaded to secondary relatively server local or spatial cache, to carry out quadratic search.
Above-mentioned index compares server, secondary relatively server, Features Management server, index database, feature database.This five part can be by
Function is disposed on different machines respectively, and same machine or a few machines can also be deployed in by required computing resource
On.
Other than the search operaqtion of above-mentioned facial image, there are also the in-stockroom operation of new facial image, i.e. index compares clothes
Device be engaged in when needing the facial image being put in storage, facial image to be put in storage is compared with the face characteristic in index database, is obtained
To corresponding index.Facial image to be put in storage is stored to face according to obtained index by Features Management server again
In property data base in subregion corresponding with the index.
The present embodiment also provides a kind of magnanimity face Rapid matching search method, as shown in Fig. 2, including following operation:
Step 200: face characteristic in the face characteristic of facial image to be retrieved and the index database pre-established is carried out
Retrieval, obtains corresponding index；
Wherein, respectively index respectively corresponds one group of representative face characteristic (i.e. a kind of face spy in the index database
Sign)；
In the step, it includes one or several indexes for meeting certain similarity that retrieval, which obtains corresponding index,.
For example, can using the algorithm for calculating distance, by the face characteristic of facial image to be retrieved with pre-establish
Face characteristic in index database is compared, and one or more indexes by similarity in comparison result in setting range are used as phase
Corresponding index.Wherein, the used algorithm for calculating distance and cluster used by generation index when pre-establishing index database
The algorithm for calculating distance is identical.
It is noted that the locality characteristic that can be face for the characteristic features being referred to herein, such as eyes, mouth
Bar, the features such as shape of face, any combination being also possible between the assemblage characteristic of face, such as the various locality characteristics of face.
As long as magnanimity face can be divided into multiclass face according to different characteristic features.It, can be using poly- in practical application
Class method is clustered from magnanimity face comes out the face with different characteristic features.
Step 201: in the institute that obtained index stores in corresponding region in the facial feature database pre-established
There is face characteristic to enclose interior, quadratic search is carried out to the face characteristic of facial image to be retrieved, gets comparison result.
Wherein, face characteristic corresponding with each index includes in the facial feature database, tool corresponding with each index
Institute of the similarity of representational face characteristic (a kind of face characteristic i.e. corresponding with each index) in setting similar threshold value
There is face characteristic.
In the face characteristic stored in region corresponding with obtained index, to the face of facial image to be retrieved
Feature can be realized when being retrieved using the algorithm for calculating distance.Wherein, it is used calculate distance algorithm with build in advance
The algorithm that cluster calculation distance used by indexing is generated when vertical index database is identical.
On the basis of aforesaid operations, facial feature database can also be managed, the management of progress operation can be with
Including appointing one or more operations as follows:
Newly-increased facial image deletes facial image, modification facial image, inquiry facial image.
For increasing the operation of facial image newly, this process is as follows:
Newly-increased facial image to be put in storage is compared with the face characteristic in the index database, obtains corresponding rope
Draw；
According to obtained index, newly-increased facial image to be put in storage is stored into the facial feature database and is somebody's turn to do
It indexes in corresponding region.
In addition, it is also provided herein a kind of retrieval device, including memory, processor and is stored on the memory and can
The computer program run on the processor, wherein the processor realizes institute as above when executing the computer program
The processing for the method stated.
Illustrate the specific implementation process of above scheme with reference to the accompanying drawing.
Magnanimity face Rapid matching retrieves structure drawing of device, as shown in figure 3, the device includes: that index compares server, two
Secondary relatively server, Features Management server, index database, facial feature database.
Index database stores index information employed in this system, wherein the index stored is by clustering method in sea
(one group of representative face characteristic is a kind of face to the representative face characteristic for clustering out in the face of amount
Feature).Index compares server and is mainly used for calculating index corresponding to given face characteristic；In face storage, ratio is indexed
Compared with server calculate with the most similar index of face characteristic to be put in storage, and by the relevant information of face characteristic to be put in storage be stored in
Most similar index region corresponding in facial feature database；In face retrieval, the index compares service
Device calculates all indexes for meeting certain similar threshold value with face to be retrieved, and using these indexes, is relatively taken by secondary
Business device carries out the matching retrieval of face.Secondary relatively server by utilizing index compares the calculated index of server and passes through feature pipe
Reason server obtains the face characteristic for needing to compare, and face characteristic to be retrieved is done ratio with these face characteristics got
Compared with obtaining comparison result.Features Management server for managing face characteristic, specifically: the increasing of facial feature database, delete,
Change, look into, obtaining corresponding human face data by index.
It is worth noting that in order to make it easy to understand, index compares server, secondary relatively server, spy in this structure chart
Levy management server, index database, feature database number all only one, but practical solution and not only limit one, can be according to industry
Business amount configures suitable number, these three certain servers can also may not be individual clothes on same machine
Business, but three different functional modules, process or thread.
Index compares the process flow of server when retrieval, as shown in figure 4, the process includes the following steps:
Step S401: the index in load index database, index compare server and index database are loaded into memory or is used to compare
Compared with hardware (such as: FPGA, GPU).
Step S402: the face characteristic in the face characteristic and index database of extracted facial image to be processed is compared
Relatively obtain corresponding index.
When being compared in this step with the feature in index database, the algorithm for calculating distance is needed and is adopted when generating index
The algorithm for calculate distance when calculating the mode and secondary comparison of distance in cluster when feature compares is consistent.
It is noted that when the face characteristic of facial image to be processed is compared with the face characteristic in index database, with
The similarity of corresponding representative face characteristic is indexed in setting similar threshold value, is considered as, facial image to be processed
Face characteristic it is corresponding with this index.Thus it can also be seen that the corresponding index of the facial image to be processed compared may
It is one, it is also possible to multiple.
Index process is generated, as shown in figure 5, the process includes the following steps:
Step S501: the face characteristic of magnanimity is extracted；
In the step, optimal retrieval precision and performance in order to obtain, the corresponding face of extracted face characteristic is best
It is the face in subsequent face database to be retrieved.
Step S502: representative face characteristic is found out as index by clustering method.
It should be noted that calculating being consistent apart from calculation of comparing with feature of mode of distance in cluster.
Face characteristic data loading process, as shown in fig. 6, the process includes the following steps:
Step S601: the face characteristic of face to be put in storage is extracted.
Step S602: the feature in face characteristic to be put in storage and index database is done into similarity-rough set, obtains most like rope
Draw.
Step S603-1: face characteristic to be put in storage is inserted by the calculated most like index of step S402 institute in face
Corresponding region in property data base.
Step S603-2: the face characteristic data that Features Management server is loaded are updated.
After above-mentioned each operation, the entire flow of face characteristic retrieval, as shown in fig. 7, the process includes the following steps:
Step S701: data initialization stage, index compare server and index database are loaded into memory or is used to compare
Hardware, the hardware that the characteristic in facial feature database is loaded into memory or is used to compare by Features Management server；
Hardware for comparing, such as: FPGA, GPU etc..
Step S702: index comparison phase, the face characteristic of face picture to be retrieved on index server with all ropes
Draw and be compared, obtains one or several qualified indexes.
In the step, comparing can be realized in CPU by the comparison algorithm that Software Coding is realized, higher in order to obtain
Performance can also be realized by hardware such as FPGA, GPU.
Wherein, when being compared with all indexes, the similarity with the face characteristic of face picture to be retrieved is in default phase
It may be considered qualified index like the index in the range of degree threshold value.
Step S703: secondary comparison phase, secondary relatively server is by the feature of facial image to be retrieved in step S702
In do quadratic search within the scope of face characteristic corresponding to obtained index, get comparison result.
In the step, comparing can be realized in CPU by the comparison algorithm that Software Coding is realized, higher in order to obtain
Performance can also be realized by hardware such as FPGA, GPU.
Step S704: result output stage, the page obtain search result and show.
Wherein it is possible to by comparison result obtained in step S703, there are memories or the interim table or local file of database
In, with the displaying for the page.
Those of ordinary skill in the art will appreciate that all or part of the steps in the above method can be instructed by program
Related hardware is completed, and described program can store in computer readable storage medium, such as read-only memory, disk or CD
Deng.Optionally, one or more integrated circuits can be used also to realize in all or part of the steps of above-described embodiment.Accordingly
Ground, each module/unit in above-described embodiment can take the form of hardware realization, can also use the shape of software function module
Formula is realized.The application is not limited to the combination of the hardware and software of any particular form.
The above, preferred embodiments only of the invention, is not intended to limit the scope of the present invention.It is all this
Within the spirit and principle of invention, any modification, equivalent substitution, improvement and etc. done should be included in protection model of the invention
Within enclosing.
Claims (10)
1. a kind of face matching retrieval device compares server, secondary relatively server including at least index, in which:
Index compares server, for face in the face characteristic of facial image to be retrieved and the index database pre-established is special
Sign is compared, and obtains corresponding index, wherein respectively index respectively corresponds a kind of face characteristic in the index database；
Secondary relatively server, for comparing index that server obtains in the face characteristic data pre-established in the index
In all face characteristics stored in corresponding region in library, the face characteristic of facial image to be retrieved is retrieved, is obtained
Get comparison result.
2. device as described in claim 1, which is characterized in that further include:
Features Management server, for being managed operation to the facial feature database；
The management operation includes appointing one or more operations as follows: newly-increased facial image deletes facial image, modification face figure
Picture, inquiry facial image.
3. device as claimed in claim 2, which is characterized in that
The index compares server, for when needing the newly-increased facial image being put in storage, by newly-increased facial image to be put in storage
It is compared with the face characteristic in the index database, is indexed accordingly；
The Features Management server, for according to obtained index, newly-increased facial image to be put in storage to be stored to described
In facial feature database in the corresponding region of the index.
4. a kind of face matches search method, comprising:
The face characteristic of facial image to be retrieved is compared with face characteristic in the index database pre-established, is obtained opposite
The index answered；
In the facial feature database pre-established, the face characteristic that is stored in region corresponding with obtained index
It is interior, the face characteristic of facial image to be retrieved is retrieved, comparison result is got；
Wherein, respectively index respectively corresponds a kind of face characteristic in the index database.
5. method as claimed in claim 4, which is characterized in that by the face characteristic of facial image to be retrieved with build in advance
Before face characteristic is compared in vertical index database, further includes:
All kinds of face characteristics are found out from the facial image obtained in advance according to clustering method as index, and establish index database.
6. method as claimed in claim 5, which is characterized in that the face characteristic by facial image to be retrieved and in advance
Face characteristic is compared in the index database of foundation, obtains corresponding index and includes:
Using the algorithm for calculating distance, by the face in the face characteristic of facial image to be retrieved and the index database pre-established
Feature is compared, using similarity in comparison result setting range index as corresponding index；
Wherein, cluster calculation distance used by the used algorithm for calculating distance is indexed with generation when pre-establishing index database
Algorithm it is identical.
7. the method as described in claim 4,5 or 6, which is characterized in that it is described in the facial feature database pre-established,
In the face characteristic stored in region corresponding with obtained index, the face characteristic of facial image to be retrieved is carried out
Retrieval, comprising:
Using the algorithm for calculating distance, in the face characteristic stored in region corresponding with obtained index, to be checked
The face characteristic of the facial image of rope is retrieved；
Wherein, cluster calculation distance used by the used algorithm for calculating distance is indexed with generation when pre-establishing index database
Algorithm it is identical.
8. the method for claim 7, which is characterized in that further include:
Operation is managed to the facial feature database, the management operation includes appointing one or more operations as follows:
Newly-increased facial image deletes facial image, modification facial image, inquiry facial image.
9. method according to claim 8, which is characterized in that the management operation carried out to the facial feature database is new
When increasing facial image, comprising:
Newly-increased facial image to be put in storage is compared with the face characteristic in the index database, is indexed accordingly；
According to obtained index, newly-increased facial image to be put in storage is stored into the facial feature database and the index
In corresponding region.
10. a kind of retrieval device, including memory, processor and it is stored on the memory and can transports on the processor
Capable computer program, which is characterized in that the processor is realized when executing the computer program as appointed in claim 4-9
The processing of method described in one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710760859.6A CN110147455B (en) | 2017-08-30 | 2017-08-30 | Face matching retrieval device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710760859.6A CN110147455B (en) | 2017-08-30 | 2017-08-30 | Face matching retrieval device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110147455A true CN110147455A (en) | 2019-08-20 |
| CN110147455B CN110147455B (en) | 2024-03-22 |
Family
ID=67588060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710760859.6A Active CN110147455B (en) | 2017-08-30 | 2017-08-30 | Face matching retrieval device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110147455B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765292A (en) * | 2019-10-24 | 2020-02-07 | 重庆紫光华山智安科技有限公司 | Image retrieval method, training method and related device |
| CN110874419A (en) * | 2019-11-19 | 2020-03-10 | 山东浪潮人工智能研究院有限公司 | Quick retrieval technology for face database |
| CN111009065A (en) * | 2019-12-09 | 2020-04-14 | 金现代信息产业股份有限公司 | Face recognition access control system optimization method and system based on clustering algorithm |
| CN111091106A (en) * | 2019-12-23 | 2020-05-01 | 浙江大华技术股份有限公司 | Image clustering method and device, storage medium and electronic device |
| CN111782855A (en) * | 2020-07-15 | 2020-10-16 | 上海依图网络科技有限公司 | Face image processing method, device, equipment and medium |
| CN112232290A (en) * | 2020-11-06 | 2021-01-15 | 四川云从天府人工智能科技有限公司 | Data clustering method, server, system, and computer-readable storage medium |
| CN112487222A (en) * | 2020-11-30 | 2021-03-12 | 江苏正赫通信息科技有限公司 | Method for quickly searching and effectively storing similar human faces |
| WO2021047305A1 (en) * | 2019-09-10 | 2021-03-18 | 中兴通讯股份有限公司 | Face search method and device, apparatus, and computer-readable storage medium |
| CN112559793A (en) * | 2021-02-23 | 2021-03-26 | 成都旺小宝科技有限公司 | Retrieval method of face image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207898A (en) * | 2013-03-19 | 2013-07-17 | 天格科技（杭州）有限公司 | Method for rapidly retrieving similar faces based on locality sensitive hashing |
| CN104765768A (en) * | 2015-03-09 | 2015-07-08 | 深圳云天励飞技术有限公司 | Mass face database rapid and accurate retrieval method |
| CN106776849A (en) * | 2016-11-28 | 2017-05-31 | 西安交通大学 | A kind of method and guide system to scheme quick-searching sight spot |
-
2017
- 2017-08-30 CN CN201710760859.6A patent/CN110147455B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207898A (en) * | 2013-03-19 | 2013-07-17 | 天格科技（杭州）有限公司 | Method for rapidly retrieving similar faces based on locality sensitive hashing |
| CN104765768A (en) * | 2015-03-09 | 2015-07-08 | 深圳云天励飞技术有限公司 | Mass face database rapid and accurate retrieval method |
| CN106776849A (en) * | 2016-11-28 | 2017-05-31 | 西安交通大学 | A kind of method and guide system to scheme quick-searching sight spot |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021047305A1 (en) * | 2019-09-10 | 2021-03-18 | 中兴通讯股份有限公司 | Face search method and device, apparatus, and computer-readable storage medium |
| CN110765292A (en) * | 2019-10-24 | 2020-02-07 | 重庆紫光华山智安科技有限公司 | Image retrieval method, training method and related device |
| CN110874419A (en) * | 2019-11-19 | 2020-03-10 | 山东浪潮人工智能研究院有限公司 | Quick retrieval technology for face database |
| CN111009065A (en) * | 2019-12-09 | 2020-04-14 | 金现代信息产业股份有限公司 | Face recognition access control system optimization method and system based on clustering algorithm |
| CN111091106B (en) * | 2019-12-23 | 2023-10-10 | 浙江大华技术股份有限公司 | Image clustering method and device, storage medium and electronic device |
| CN111091106A (en) * | 2019-12-23 | 2020-05-01 | 浙江大华技术股份有限公司 | Image clustering method and device, storage medium and electronic device |
| CN111782855A (en) * | 2020-07-15 | 2020-10-16 | 上海依图网络科技有限公司 | Face image processing method, device, equipment and medium |
| CN112232290A (en) * | 2020-11-06 | 2021-01-15 | 四川云从天府人工智能科技有限公司 | Data clustering method, server, system, and computer-readable storage medium |
| CN112232290B (en) * | 2020-11-06 | 2023-12-08 | 四川云从天府人工智能科技有限公司 | Data clustering method, server, system and computer readable storage medium |
| CN112487222A (en) * | 2020-11-30 | 2021-03-12 | 江苏正赫通信息科技有限公司 | Method for quickly searching and effectively storing similar human faces |
| CN112487222B (en) * | 2020-11-30 | 2021-11-30 | 江苏正赫通信息科技有限公司 | Method for quickly searching and effectively storing similar human faces |
| CN112559793B (en) * | 2021-02-23 | 2021-07-13 | 成都旺小宝科技有限公司 | Retrieval method of face image |
| CN112559793A (en) * | 2021-02-23 | 2021-03-26 | 成都旺小宝科技有限公司 | Retrieval method of face image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110147455B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147455A (en) | A kind of face matching retrieval device and method | |
| US9454580B2 (en) | Recommendation system with metric transformation | |
| US10650274B2 (en) | Image clustering method, image clustering system, and image clustering server | |
| US11204707B2 (en) | Scalable binning for big data deduplication | |
| CN109766436A (en) | A kind of matched method and apparatus of data element of the field and knowledge base of tables of data | |
| CN111522989A (en) | Method, computing device, and computer storage medium for image retrieval | |
| EP3470992B1 (en) | Efficient storage and utilization of a hierarchical data set | |
| CN111507285A (en) | Face attribute recognition method and device, computer equipment and storage medium | |
| WO2015001416A1 (en) | Multi-dimensional data clustering | |
| CN108052535B (en) | Visual feature parallel rapid matching method and system based on multiprocessor platform | |
| CN109416688B (en) | Method and system for flexible high performance structured data processing | |
| WO2017095439A1 (en) | Incremental clustering of a data stream via an orthogonal transform based indexing | |
| CN111177372A (en) | Scientific and technological achievement classification method, device, equipment and medium | |
| CN112784102B (en) | Video retrieval method and device and electronic equipment | |
| CN113761017A (en) | Similarity searching method and device | |
| CN110209895B (en) | Vector retrieval method, device and equipment | |
| CN107452003A (en) | A kind of method and device of the image segmentation containing depth information | |
| Chen et al. | Learning to generalize in heterogeneous federated networks | |
| CN104850600B (en) | A kind of method and apparatus for searching for the picture comprising face | |
| Yin et al. | Content‐Based Image Retrial Based on Hadoop | |
| US10409992B2 (en) | Investigation apparatus, computer-readable recording medium, and investigation method | |
| Kang et al. | Interactive hierarchical tag clouds for summarizing spatiotemporal social contents | |
| Horak et al. | FCA as a tool for inaccuracy detection in content-based image analysis | |
| US20150142712A1 (en) | Rule discovery system, method, apparatus, and program | |
| CN105279172A (en) | Video matching method and device |
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 |