CN110443297A - Clustering method, device and the computer storage medium of image - Google Patents
Clustering method, device and the computer storage medium of image Download PDFInfo
- Publication number
- CN110443297A CN110443297A CN201910696640.3A CN201910696640A CN110443297A CN 110443297 A CN110443297 A CN 110443297A CN 201910696640 A CN201910696640 A CN 201910696640A CN 110443297 A CN110443297 A CN 110443297A
- Authority
- CN
- China
- Prior art keywords
- image
- clustering
- cluster
- image data
- presentation graphics
- 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 34
- 238000012790 confirmation Methods 0.000 claims abstract description 14
- 238000010586 diagram Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 15
- 241001269238 Data Species 0.000 claims description 12
- 238000005286 illumination Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 description 5
- 230000001815 facial effect Effects 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/95—Hardware or software architectures specially adapted for image or video understanding structured as a network, e.g. client-server architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- G06V10/993—Evaluation of the quality of the acquired pattern
-
- 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
- G06V40/50—Maintenance of biometric data or enrolment thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of clustering method of image, device and computer storage medium, this method includes obtaining a plurality of image data;Clustering processing is carried out to a plurality of image data, obtains at least one image clustering;The presentation graphics of image clustering are obtained, presentation graphics are the cluster centre of the highest image data of mass value or image clustering in image clustering;Confirm image collection library in the matched cluster center of presentation graphics;Image clustering is saved to image collection library according to confirmation result.By the above-mentioned means, can greatly accelerate to cluster efficiency.
Description
Technical field
The present invention relates to the cluster fields of image, deposit more particularly to the clustering method, device and computer of a kind of image
Storage media.
Background technique
In existing clustering method, multiple cameras corresponds to multiple images cluster, and each image clustering includes gathering
Single picture is then compared with the cluster centre of multiple images cluster by class center one by one after video camera gets picture,
And carry out stepping.But stepping inefficiency, the camera chain that video camera reaches certain amount grade can not be adapted to.
Summary of the invention
The present invention provides clustering method, device and the computer storage medium of a kind of image, to solve the prior art
The lower problem of middle picture stepping efficiency.
In order to solve the above technical problems, one technical scheme adopted by the invention is that provide a kind of clustering method of image,
The described method includes: obtaining a plurality of image data;Clustering processing is carried out to a plurality of image data, obtains at least one image
Cluster;The presentation graphics of described image cluster are obtained, the presentation graphics are the mass value highest in described image cluster
Image data;Confirm image collection library in the matched cluster center of the presentation graphics;It will be described according to confirmation result
Image clustering is saved to described image set library.
In order to solve the above technical problems, another technical solution used in the present invention is to provide a kind of cluster dress of image
It sets, the clustering apparatus of described image includes: acquisition module, for obtaining a plurality of image data;Processing module, for described more
Image data carries out clustering processing, obtains at least one image clustering, obtains the presentation graphics of described image cluster, described
Presentation graphics are the cluster centre of the highest image data of mass value or described image cluster in described image cluster;Confirmation
Module, for confirm in image collection library with the matched cluster center of the presentation graphics;Preserving module, for according to confirmation
As a result described image cluster is saved to described image set library.
In order to solve the above technical problems, the cluster that another technical solution used in the present invention is to provide a kind of image determines
Device, including processor and memory are stored with computer program in memory, and processor is for executing computer program with reality
The step of cluster of existing above-mentioned image determines method.
In order to solve the above technical problems, another technical solution used in the present invention is to provide a kind of computer storage Jie
Matter, wherein being stored with computer program, computer program, which is performed, realizes the step of cluster of above-mentioned image determines method.
It is different from the prior art, it is poly- then to get image by carrying out clustering processing to a plurality of image data by the present invention
Class and presentation graphics, and presentation graphics are matched with the cluster center in image collection library, and will according to matching result
Image clustering is saved to image collection library.To greatly reduce calculation amount, cluster efficiency has been speeded.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, to this
For the those of ordinary skill of field, without creative efforts, it can also be obtained according to these attached drawings others
Attached drawing.
Fig. 1 is the flow diagram of one embodiment of clustering method of image of the present invention;
Fig. 2 is the sub-step flow diagram of Fig. 1 step S12;
Fig. 3 is the structural schematic diagram of one embodiment of cluster determining device of image of the present invention;
Fig. 4 is the structural schematic diagram of another embodiment of cluster determining device of image of the present invention;
Fig. 5 is the structural schematic diagram of one embodiment of computer storage medium of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that the described embodiments are merely a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It in video monitoring, needs to classify to a large amount of facial image, so that same category of facial image
Classification same category.Specifically, so that allowing the facial image of the same person as an individual classification.
It is the flow diagram of the clustering method first embodiment of image of the present invention, this implementation referring specifically to Fig. 1, Fig. 1
The clustering method of example diagram picture includes the following steps.
S11 obtains a plurality of image data.
In a particular embodiment, it can specifically be grabbed by a large amount of video camera real-time perfoming with real-time image acquisition data
It claps to obtain image, and characteristic processing is carried out to get image data, to be stored in preset data to the image of acquisition
In library.The a plurality of image data of preset quantity is then periodically obtained from the database.Specifically, which can carry out
Default, such as 1h, 12h or for 24 hours etc. is not construed as limiting here.Preset quantity can also specifically be preset, can be with
The calculation amount of computing unit is that reference is set, when there is the image data more than or equal to preset quantity in database,
The image data for then obtaining preset quantity then obtains in database when database has the image data less than preset quantity
Whole image datas.And mark has been obtained to the image data obtained in database, so that next time will not weigh
It is multiple to obtain.
S12 carries out clustering processing to a plurality of image data, obtains at least one image clustering.
Clustering processing is carried out to a plurality of image data, to obtain at least one image clustering, specifically, an image is poly-
Class is the set for the multiple images data that similarity reaches certain threshold value.
Referring to Fig. 2, Fig. 2 is the sub-step of step S12 in first embodiment in the clustering method of image of the present invention.This reality
Apply the clustering method of example diagram picture the following steps are included:
S121 calculates the similarity of every image data and other image datas;
The calculating of similarity is carried out to a plurality of image data, specifically, calculates every image data and other image datas
Similarity then get the result of n* (n-1) similarity by taking n image data as an example.Such as ABCDE five if it exists
Image data, then successively obtain A-B, A-C, A-D, the similarity of B-A, B-C, and so on, until the similarity of E-D.To
Determine the result of 20 similarities.
Similarity is greater than other image datas of similarity threshold as the similar image of each image data by S122
Collection.
Similarity is greater than other image datas of similarity threshold as the similar diagram image set of each image data.Specifically
Ground carries out similarity judgement to every image data, if the similarity of other image datas and this image data is greater than phase
Like degree threshold value, then using other image datas as the similar diagram image set of this image data.
In five image datas of ABCDE, if the similarity of A-B, B-A, B-C, C-B, D-E, E-D are greater than similarity threshold
Value.Then the similar diagram image set of image data A is (B), and the similar diagram image set of image data B is (A, C), and so on, picture number
Similar diagram image set according to E is (D).
S123, successively determines whether every image data is included into image clustering.
Then successively every image data is determined, judges whether to be included into image clustering.It such as can be to a plurality of
Image data carries out a sequence identification, and the sequence identified in sequence is successively determined a plurality of image data.It is such as right
Five data of ABCDE successively identifies 1-5, is then successively determined to ABCDE according to the sequence of 1-5.Or use other times
The mode gone through is determined, and can be determined with to guarantee every image data in a plurality of image data and will not be repeated.
S124, if image data is not included into image clustering, using image data and its similar diagram image set as new image
Cluster.
If image data is not included into image clustering, the similar diagram image set of image data and the image data is made
For new image clustering.
By taking five image datas of ABCDE as an example, when being successively determined to ABCDE, A is determined first, due to
There are no image clusterings, then A is not belonging to any image clustering, then using the similar diagram image set of A and A as an image clustering.
It regard A, (B) as an image clustering.
When being determined to D, since D is not included into already existing image clustering, then by D, (E) is new as one
Image clustering.
Its similar diagram image set is included into image clustering if image data has been included into image clustering by S125.
If the image data has been included into image clustering, the similar diagram image set of the image data is also included into figure
As in cluster.
By taking five image datas of ABCDE as an example, in the case where having existed image clustering (A (B)), carried out really to B
Periodically, it is suffered since B has been included into image clustering (A, (B)), then the similarity graph image set of B is also further included into image
Cluster (A, (B)) in, i.e., (A, C) is included into image clustering (A, (B)), thus get image clustering (A, (B), (A,
C))。
In a particular embodiment, each image clustering is deleted and is handled again, so that each in each image clustering
Identical image data only saves one.Become (A, B, C) after such as being deleted processing again to image clustering (A, (B), (A, C)).
By the above-mentioned means, being successively determined to each image data, so that each image data is included into
In image clustering.
S13, obtains the presentation graphics of image clustering, and presentation graphics are the highest image of mass value in image clustering
Data.
In getting at least one image clustering, the presentation graphics of image clustering are further obtained, an image is poly-
The presentation graphics of class are the highest image data of mass value in the image clustering.
In another embodiment, which is also possible to the cluster centre of the image clustering, and particularly
One is included into the image data of image clustering.In the image clustering (A, (B)) in S124 step, A be the image clustering (A,
(B)) cluster centre or representative diagram image set.
Specifically, presentation graphics can be an image data, be also possible to the set of multiple images data composition, this
In without limitation.
Specifically, mass value is the weighting of shaded coefficient, fuzzy coefficient, illumination tensor and three-dimensional perspective in image data
With.It can specifically be calculated with following publicity:
F=occlusion*k1+blur*k2+illumination*k3+Pitch*k4+Roll*k5+ Yaw*k6;
Wherein, f is the mass value of image data, and occlusion is shaded coefficient, and blur is fuzzy coefficient,
Illumination is illumination tensor, Pitch be in three-dimensional perspective pitch angle [- 90 (and on), 90 (under)], Roll three dimensional angular
Plane internal rotation angle [- 180 (counterclockwise), 180 (clockwise)] in degree, Yaw are the left-right rotary corner [- 90 in three-dimensional perspective
(left side), 90 (right sides)].K1-k6 is different weighting coefficient.Here the weighting coefficient of k1-k6 specifically can according to concrete condition into
Row is default.
S14, confirm image collection library in the matched cluster center of presentation graphics.
Whether have cluster center with presentation graphics matched, specifically, can calculate generation if determining in image collection library
The similarity of cluster center in table image and image collection library, and collection is determined according to whether similarity is greater than similarity threshold
Whether conjunction center matches with presentation graphics.If similarity is greater than threshold value, it is determined that the cluster center and presentation graphics
Match.
S15 saves image clustering to image collection library according to confirmation result.
In one embodiment, if not gathering image with the matched cluster center of presentation graphics in image collection library
Class is saved with new image collection to image collection library;
Specifically, image collection library specifically can be the set of multiple images set.
In a particular embodiment, image collection library is an empty library originally, then by first batch above-mentioned steps S11-S13 institute
The presentation graphics of the image clustering of acquisition go to be compared with the cluster center in image collection library, due to image collection library at this time
For empty library, then cluster center identical with presentation graphics can not be found in image collection library, then using image clustering as figure
Image set closes image collection new in library, and using presentation graphics as cluster center.
When determining subsequent batch operation via image clustering acquired in S11-S13, due to the image collection library
Through there are the image collection of part and cluster centers.Then successively by the collection in the presentation graphics of image clustering and image collection library
Conjunction center closely compares, and calculates whether similarity is greater than similarity threshold, similarity threshold here and above-described embodiment
Similarity threshold can be the same, is also possible to different, is not construed as limiting here.
If similarity is respectively less than certain threshold value, prove in image collection library not with the matched collection of the presentation graphics
Conjunction center, then using the image clustering as image collection new in image collection library, and using presentation graphics as cluster center.
In another embodiment, if having in image collection library with the matched cluster center of presentation graphics, image is gathered
Class is saved into the corresponding image collection of matched cluster center.
If image collection inventory is the matched cluster center of the presentation graphics, i.e., there are one in image collection library
The similarity of image collection, the presentation graphics of the cluster center and image clustering of the image collection is greater than certain similarity threshold
Value, then save the image clustering into the corresponding image collection of cluster center.
Specifically, when saving image clustering into the corresponding image collection of matched cluster center, compare the image
The mass value for the cluster center that the presentation graphics of cluster polymerize with institute picture to be saved, the calculation of mass value are above-mentioned
Step has been described, and which is not described herein again.
If the mass value of presentation graphics is greater than the mass value of cluster center, using presentation graphics as addition image
The cluster center of image collection after cluster.So that the cluster center of image collection is best in quality in whole image set
Image data.
In above-described embodiment, acquired image data is handled by periodicity, and successively to image data into
The primary cluster of row and secondary cluster.Image data is clustered in advance in primary cluster, so that similarity is greater than certain phase
Like degree threshold value image data as an image clustering, and determine mass value is best in image clustering image data as generation
Table image.And due to once clustering in, be successively to each image data carry out similarity calculation, with ABCDE five scheme
As data instance, the similarity of image data A and image data B are greater than similarity threshold, image data B and image data C's
Similarity reaches greater than similarity threshold, but image data A and image data C is being compared suitable, and similarity is less than phase
Like degree threshold value, but by the above-mentioned means, image data C and image data A can be still stored in same cluster.Pass through
Aforesaid way can be stored in a cluster in similar image data as far as possible.Greatly reduce error.
And in secondary cluster, presentation graphics are compared with the cluster center in image collection library.Thus by image
Data further progress saves.So that the image data that similarity reaches certain threshold value can be stored in image collection library
In one image collection.Since the image collection quantity in image collection library is generally large, after once clustering, only to representativeness
Image is compared.Calculation amount to be significantly reduced, can quickly classify to batch image data, enable to
It is stored in image collection library.
And further, for compared with the prior art, method provided by the present application is adapted to a large amount of video camera systems
System, and interconnected by establishing image collection library and each video camera, optimize the classification and storage of image data.
As shown in figure 3, the application also provides a kind of cluster determining device 300 of image, the cluster determining device of the image
300 include obtaining module 31, processing module 32, confirmation module 33 and preserving module 34.Wherein, module 31 is obtained for obtaining
A plurality of image data;Processing module 32 is used to carry out clustering processing to a plurality of image data, obtains at least one image clustering, and
The presentation graphics of image clustering are obtained, which is the highest image data of mass value in image clustering;Confirmation
Module 34 is used to confirm in image collection library and the matched cluster center of presentation graphics;Preserving module 35 is used to be tied according to confirmation
Fruit saves image clustering to image collection library.Its specific step above-described embodiment has been described, and which is not described herein again.
The cluster of above-mentioned image determines that method is generally realized by the cluster determining device of image, thus the present invention also proposes one
The cluster determining device of kind image.Referring to Fig. 4, the structure that Fig. 4 is one embodiment of cluster determining device of image of the present invention is shown
It is intended to.The cluster determining device 100 of the present embodiment image includes processor 12 and memory 11;Calculating is stored in memory 11
Machine program, processor 12 is for executing computer program to realize the step of cluster such as above-mentioned image determines method.
The cluster of above-mentioned image determines that the logical process of method is presented with computer program, in terms of computer program, if
It when selling or using, is storable in computer storage medium as independent software product, thus the present invention proposes one
Kind computer storage medium.Referring to Fig. 5, Fig. 5 is the structural schematic diagram of one embodiment of computer storage medium of the present invention, this reality
It applies and is stored with computer program 21 in a computer storage medium 200, above-mentioned match is realized when computer program is executed by processor
Network method or control method.
The computer storage medium 200 is specifically as follows USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. can store calculating
The medium of machine program, or may be the server for being stored with the computer program, which can be by the computer of storage
Program is sent to other equipment operation, or can also be with the computer program of the self-operating storage.The computer storage medium 200
It can be the combination of multiple entities from physical entity, such as multiple servers, server add memory or memory
Add the multiple combinations mode such as mobile hard disk.
Mode the above is only the implementation of the present invention is not intended to limit the scope of the invention, all to utilize this
Equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content, it is relevant to be applied directly or indirectly in other
Technical field is included within the scope of the present invention.
Claims (10)
1. a kind of clustering method of image, which is characterized in that the described method includes:
Obtain a plurality of image data;
Clustering processing is carried out to a plurality of image data, obtains at least one image clustering;
The presentation graphics of described image cluster are obtained, the presentation graphics are that the mass value in described image cluster is highest
The cluster centre of image data or described image cluster;
Confirm image collection library in the matched cluster center of the presentation graphics;
Described image cluster is saved to described image set library according to confirmation result.
2. clustering method according to claim 1, which is characterized in that described clustered described image according to confirmation result is protected
It deposits to described image set library, comprising:
If in described image set library not with the matched cluster center of presentation graphics, by image clustering with new image set
It closes and saves to described image set library;
If have in described image set library with the matched cluster center of presentation graphics, image clustering is saved to matched institute
It states in the corresponding image collection of cluster center.
3. clustering method according to claim 2, which is characterized in that described to save image clustering to the matched collection
In the corresponding image collection in conjunction center, include: later
Compare the mass value of the presentation graphics and the mass value of the cluster center;
If the mass value of the presentation graphics is greater than the mass value of the cluster center, using the presentation graphics as adding
The cluster center of described image set after adding described image to cluster.
4. clustering method according to claim 3, which is characterized in that the mass value is to block to be in described image data
The weighted sum of number, fuzzy coefficient, illumination tensor and three-dimensional perspective.
5. clustering method according to claim 1, which is characterized in that in the confirmation image collection library with the representativeness
The cluster center of images match, comprising:
Calculate the similarity of cluster center in the presentation graphics and described image set library;
The cluster center that confirmation similarity is greater than similarity threshold is matched with the presentation graphics.
6. clustering method according to claim 1, which is characterized in that described to be carried out at cluster to a plurality of image data
Reason, obtains at least one image clustering, comprising:
Calculate the similarity of every image data Yu other image datas;
Similarity is greater than other image datas of similarity threshold as the similar diagram image set of each image data;
Successively determine whether every image data is included into image clustering;
If described image data have been included into image clustering, its similar diagram image set is included into image clustering;
If described image data are not included into image clustering, using image data and its similar diagram image set as new image clustering.
7. clustering method according to claim 1, which is characterized in that described to obtain a plurality of image data, comprising:
Real-time image acquisition data;
Periodically acquire a plurality of image data of preset quantity.
8. a kind of clustering apparatus of image, which is characterized in that the clustering apparatus of described image includes:
Module is obtained, for obtaining a plurality of image data;
Processing module obtains at least one image clustering for carrying out clustering processing to a plurality of image data, described in acquisition
The presentation graphics of image clustering, the cluster centre presentation graphics are the highest image of mass value in described image cluster
The cluster centre of data or described image cluster;
Confirmation module, for confirm in image collection library with the matched cluster center of the presentation graphics;
Preserving module, for being saved described image cluster to described image set library according to confirmation result.
9. a kind of clustering apparatus of image, which is characterized in that the clustering apparatus of described image includes processor and memory;It is described
Computer program is stored in memory, the processor is for executing the computer program to realize such as claim 1-7
Any one of the method the step of.
10. a kind of computer storage medium, which is characterized in that the computer storage medium is stored with computer program, described
Computer program is performed the step of realization any one of such as claim 1-7 the method.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910696640.3A CN110443297B (en) | 2019-07-30 | 2019-07-30 | Image clustering method and device and computer storage medium |
PCT/CN2019/117421 WO2021017282A1 (en) | 2019-07-30 | 2019-11-12 | Systems and methods for image clustering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910696640.3A CN110443297B (en) | 2019-07-30 | 2019-07-30 | Image clustering method and device and computer storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110443297A true CN110443297A (en) | 2019-11-12 |
CN110443297B CN110443297B (en) | 2022-06-07 |
Family
ID=68432386
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910696640.3A Active CN110443297B (en) | 2019-07-30 | 2019-07-30 | Image clustering method and device and computer storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110443297B (en) |
WO (1) | WO2021017282A1 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111061899A (en) * | 2019-12-18 | 2020-04-24 | 深圳云天励飞技术有限公司 | Archive representative picture generation method and device and electronic equipment |
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 |
CN111104970A (en) * | 2019-12-04 | 2020-05-05 | 浙江大华技术股份有限公司 | Face clustering method and device |
WO2021128883A1 (en) * | 2019-12-23 | 2021-07-01 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for managing portrait data |
CN115687249A (en) * | 2022-12-30 | 2023-02-03 | 浙江大华技术股份有限公司 | Image gathering method and device, terminal and computer readable storage medium |
WO2023124134A1 (en) * | 2021-12-28 | 2023-07-06 | 上海商汤智能科技有限公司 | File processing method and apparatus, electronic device, computer storage medium and program |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113240092A (en) * | 2021-05-31 | 2021-08-10 | 深圳市商汤科技有限公司 | Neural network training and face recognition method, device, equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100303342A1 (en) * | 2009-06-02 | 2010-12-02 | Yahoo! Inc. | Finding iconic images |
US20180210898A1 (en) * | 2013-06-10 | 2018-07-26 | Oath Inc. | Image-based faceted system and method |
CN108875834A (en) * | 2018-06-22 | 2018-11-23 | 北京达佳互联信息技术有限公司 | Image clustering method, device, computer equipment and storage medium |
CN109213732A (en) * | 2018-06-28 | 2019-01-15 | 努比亚技术有限公司 | A kind of method, mobile terminal and computer readable storage medium improving photograph album classification |
CN109242045A (en) * | 2018-09-30 | 2019-01-18 | 北京达佳互联信息技术有限公司 | Image clustering processing method, device, electronic equipment and storage medium |
CN110069989A (en) * | 2019-03-15 | 2019-07-30 | 上海拍拍贷金融信息服务有限公司 | Face image processing process and device, computer readable storage medium |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7831599B2 (en) * | 2005-03-04 | 2010-11-09 | Eastman Kodak Company | Addition of new images to an image database by clustering according to date/time and image content and representative image comparison |
US8351661B2 (en) * | 2009-12-02 | 2013-01-08 | At&T Intellectual Property I, L.P. | System and method to assign a digital image to a face cluster |
US8983210B2 (en) * | 2010-03-01 | 2015-03-17 | Microsoft Corporation | Social network system and method for identifying cluster image matches |
US9613455B1 (en) * | 2014-11-02 | 2017-04-04 | Google Inc. | Local georeferenced data |
-
2019
- 2019-07-30 CN CN201910696640.3A patent/CN110443297B/en active Active
- 2019-11-12 WO PCT/CN2019/117421 patent/WO2021017282A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100303342A1 (en) * | 2009-06-02 | 2010-12-02 | Yahoo! Inc. | Finding iconic images |
US20180210898A1 (en) * | 2013-06-10 | 2018-07-26 | Oath Inc. | Image-based faceted system and method |
CN108875834A (en) * | 2018-06-22 | 2018-11-23 | 北京达佳互联信息技术有限公司 | Image clustering method, device, computer equipment and storage medium |
CN109213732A (en) * | 2018-06-28 | 2019-01-15 | 努比亚技术有限公司 | A kind of method, mobile terminal and computer readable storage medium improving photograph album classification |
CN109242045A (en) * | 2018-09-30 | 2019-01-18 | 北京达佳互联信息技术有限公司 | Image clustering processing method, device, electronic equipment and storage medium |
CN110069989A (en) * | 2019-03-15 | 2019-07-30 | 上海拍拍贷金融信息服务有限公司 | Face image processing process and device, computer readable storage medium |
Non-Patent Citations (2)
Title |
---|
JING WANG.ET AL: ""Texture-based Segmentation for Extracting Image Shape Features"", 《IEEE》 * |
齐美彬等: ""大规模图像集中的代表性图像选取"", 《自动化学报》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111104970A (en) * | 2019-12-04 | 2020-05-05 | 浙江大华技术股份有限公司 | Face clustering method and device |
CN111104970B (en) * | 2019-12-04 | 2023-04-18 | 浙江大华技术股份有限公司 | Face clustering method and device |
CN111061899A (en) * | 2019-12-18 | 2020-04-24 | 深圳云天励飞技术有限公司 | Archive representative picture generation method and device and electronic equipment |
CN111061899B (en) * | 2019-12-18 | 2022-04-26 | 深圳云天励飞技术股份有限公司 | Archive representative picture generation method and device and electronic equipment |
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 |
WO2021128883A1 (en) * | 2019-12-23 | 2021-07-01 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for managing portrait data |
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 |
WO2023124134A1 (en) * | 2021-12-28 | 2023-07-06 | 上海商汤智能科技有限公司 | File processing method and apparatus, electronic device, computer storage medium and program |
CN115687249A (en) * | 2022-12-30 | 2023-02-03 | 浙江大华技术股份有限公司 | Image gathering method and device, terminal and computer readable storage medium |
CN115687249B (en) * | 2022-12-30 | 2023-03-28 | 浙江大华技术股份有限公司 | Image gathering method and device, terminal and computer readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
WO2021017282A1 (en) | 2021-02-04 |
CN110443297B (en) | 2022-06-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110443297A (en) | Clustering method, device and the computer storage medium of image | |
US9619734B2 (en) | Classification of land based on analysis of remotely-sensed earth images | |
CN104350471B (en) | Method and system for detecting anomalies in real-time in processing environment | |
EP3646254A1 (en) | Gpu enhanced graph model build and scoring engine | |
KR20130122662A (en) | Method and system for comparing images | |
CN110764898B (en) | Task allocation method and device, readable storage medium and terminal equipment | |
CN111783713B (en) | Weak supervision time sequence behavior positioning method and device based on relation prototype network | |
CN110363091A (en) | Face identification method, device, equipment and storage medium in the case of side face | |
EP3012781A1 (en) | Method and apparatus for extracting feature correspondences from multiple images | |
CN108205570A (en) | A kind of data detection method and device | |
CN111914908A (en) | Image recognition model training method, image recognition method and related equipment | |
CN113052245B (en) | Image clustering method and device, electronic equipment and storage medium | |
WO2023221790A1 (en) | Image encoder training method and apparatus, device, and medium | |
CN115063589A (en) | Knowledge distillation-based vehicle component segmentation method and related equipment | |
CN110569380B (en) | Image tag obtaining method and device, storage medium and server | |
CN115049878A (en) | Target detection optimization method, device, equipment and medium based on artificial intelligence | |
CN115631399A (en) | Training method of image quality evaluation model, image quality evaluation method and device | |
CN114973057A (en) | Video image detection method based on artificial intelligence and related equipment | |
CN113837202A (en) | Feature point extraction method, image reconstruction method and device | |
CN116580249B (en) | Method, system and storage medium for classifying beats based on ensemble learning model | |
CN110163103A (en) | A kind of live pig Activity recognition method and apparatus based on video image | |
CN110309790B (en) | Scene modeling method and device for road target detection | |
CN112149528A (en) | Panorama target detection method, system, medium and equipment | |
CN114723883A (en) | Three-dimensional scene reconstruction method, device, equipment and storage medium | |
Geng et al. | Shelf Product Detection Based on Deep Neural Network |
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 |