CN112070144A - Image clustering method and device, electronic equipment and storage medium - Google Patents
Image clustering method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112070144A CN112070144A CN202010917917.3A CN202010917917A CN112070144A CN 112070144 A CN112070144 A CN 112070144A CN 202010917917 A CN202010917917 A CN 202010917917A CN 112070144 A CN112070144 A CN 112070144A
- Authority
- CN
- China
- Prior art keywords
- images
- image
- target image
- clustering
- neighborhood
- 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 68
- 238000012545 processing Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000013507 mapping Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000005192 partition Methods 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011982 device technology Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000638 solvent extraction Methods 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The application discloses an image clustering method, an image clustering device, electronic equipment and a storage medium, and relates to the technical field of electronic equipment. The method comprises the steps of obtaining a plurality of images and obtaining a plurality of clustering factors aiming at the images, wherein each image in the images is used as a data point, obtaining a target image from the images, obtaining a plurality of neighborhood radiuses corresponding to the clustering factors, obtaining a plurality of neighborhoods of the target image by taking the target image as a data point center and the neighborhood radiuses as radiuses, respectively obtaining the images in each neighborhood of the neighborhoods, obtaining a plurality of image sets, obtaining a target image set based on the image sets, and clustering the images contained in the target image set. According to the method and the device, when a plurality of clustering factors exist, a plurality of neighborhoods are obtained by setting a plurality of neighborhood radius divisions, and images contained in a target image set obtained based on a plurality of image sets corresponding to the plurality of neighborhoods are clustered, so that the accuracy of image clustering is improved.
Description
Technical Field
The present application relates to the field of electronic device technologies, and in particular, to an image clustering method and apparatus, an electronic device, and a storage medium.
Background
With the development of science and technology, electronic equipment is more and more widely used and has more and more functions, and the electronic equipment becomes one of the necessary things in daily life of people. The photo album of the electronic device generally has an image classification function, that is, images are selected from the photo album and classified according to preset conditions, and at present, the electronic device does not accurately classify the images.
Disclosure of Invention
In view of the above problems, the present application provides an image clustering method, an apparatus, an electronic device, and a storage medium to solve the above problems.
In a first aspect, an embodiment of the present application provides an image clustering method, where the method includes: acquiring a plurality of images, and acquiring a plurality of clustering factors for the plurality of images, wherein each image in the plurality of images serves as a data point; acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radii corresponding to the plurality of clustering factors; taking the target image as a data point center and the neighborhood radiuses as radiuses, and acquiring a plurality of neighborhoods of the target image, wherein the neighborhoods correspond to the neighborhood radiuses one by one; respectively acquiring images in each neighborhood of the plurality of neighborhoods to obtain a plurality of image sets; and obtaining a target image set based on the plurality of image sets, and clustering the images contained in the target image set.
In a second aspect, an embodiment of the present application provides an image gathering device, including: an image acquisition module, configured to acquire a plurality of images, and acquire a plurality of clustering factors for the plurality of images, where each image in the plurality of images serves as a data point; a neighborhood radius obtaining module, configured to obtain a target image from the multiple images, and obtain multiple neighborhood radii corresponding to the multiple clustering factors; the neighborhood acquisition module is used for acquiring a plurality of neighborhoods of the target image by taking the target image as a data point center and taking the plurality of neighborhood radiuses as radiuses, wherein the plurality of neighborhoods correspond to the plurality of neighborhood radiuses in a one-to-one mode; an image set acquisition module, configured to acquire an image located in each of the multiple neighborhoods, respectively, and obtain multiple image sets; and the image clustering module is used for obtaining a target image set based on the plurality of image sets and clustering the images contained in the target image set.
In a third aspect, an embodiment of the present application provides an electronic device, including a memory and a processor, the memory being coupled to the processor, the memory storing instructions, and the processor performing the above method when the instructions are executed by the processor.
In a fourth aspect, the present application provides a computer-readable storage medium, in which a program code is stored, and the program code can be called by a processor to execute the above method.
The image clustering method, the device, the electronic device and the storage medium provided by the embodiment of the application acquire a plurality of images and acquire a plurality of clustering factors aiming at the plurality of images, wherein each image in the plurality of images is used as a data point, a target image is acquired from the plurality of images, a plurality of neighborhood radiuses corresponding to the plurality of clustering factors are acquired, a plurality of neighborhoods of the target image are acquired by taking the target image as a data point center and the plurality of neighborhood radiuses as radiuses, the plurality of neighborhoods are in one-to-one correspondence with the plurality of neighborhood radiuses, the images in each neighborhood in the plurality of neighborhoods are respectively acquired to acquire a plurality of image sets, a target image set is acquired based on the plurality of image sets, the images contained in the target image set are clustered, and therefore, when the plurality of clustering factors exist, the plurality of neighborhoods are acquired by setting the plurality of neighborhood radius divisions, and clustering images contained in a target image set obtained based on a plurality of image sets corresponding to a plurality of neighborhoods, so that the accuracy of image clustering is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart illustrating an image clustering method according to an embodiment of the present application;
FIG. 2 is a flow chart of an image clustering method according to another embodiment of the present application;
FIG. 3 is a flow chart illustrating an image clustering method according to still another embodiment of the present application;
FIG. 4 is a flow chart illustrating an image clustering method according to another embodiment of the present application;
FIG. 5 is a flow chart illustrating step S470 of the image clustering method illustrated in FIG. 4 of the present application;
FIG. 6 is a flow chart illustrating step S472 of the image clustering method illustrated in FIG. 5 of the present application;
FIG. 7 is a block diagram illustrating an image clustering apparatus provided in an embodiment of the present application;
FIG. 8 is a block diagram of an electronic device for executing an image clustering method according to an embodiment of the present application;
fig. 9 illustrates a storage unit for storing or carrying program codes for implementing an image clustering method according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
At present, when an electronic device clusters images in an album, the images are mainly clustered according to a Clustering condition by using a Density-Based Clustering of Applications with Noise (DBSCAN) Clustering algorithm, for example, the images are clustered according to shooting position information of the images by using the DBSCAN Clustering algorithm. However, the inventor finds that, when image clustering is performed at present, images are clustered from only one fixed dimension, and other dimensions are not used as references, so that the problem of inaccurate image classification is caused.
In view of the above problems, the inventors have found through long-term research and provide an image clustering method, an image clustering device, an electronic device, and a storage medium provided in the embodiments of the present application, where multiple neighborhoods are obtained by setting multiple neighborhood radius partitions when multiple clustering factors exist, and clustering images included in a target image set obtained based on multiple image sets corresponding to multiple neighborhoods, so as to improve the accuracy of image clustering. The specific image clustering method is described in detail in the following embodiments.
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating an image clustering method according to an embodiment of the present application. The image clustering method is used for obtaining a plurality of neighborhoods by setting a plurality of neighborhood radius divisions when a plurality of clustering factors exist, and clustering images contained in a target image set obtained based on a plurality of image sets corresponding to the plurality of neighborhoods, so that the accuracy of image clustering is improved. In a specific embodiment, the image clustering method is applied to the image clustering device 200 shown in fig. 7 and the electronic device 100 (fig. 8) equipped with the image clustering device 200. The specific process of the present embodiment will be described below by taking an electronic device as an example, and it is understood that the electronic device applied in the present embodiment may be a smart phone, a tablet computer, a wearable electronic device, and the like, which is not limited herein. As will be described in detail with respect to the flow shown in fig. 1, the image clustering method may specifically include the following steps:
step S110: a plurality of images are acquired, and a plurality of clustering factors for the plurality of images are acquired, wherein each image in the plurality of images serves as a data point.
In this embodiment, a plurality of images may be acquired, and a plurality of clustering factors for the plurality of images may be acquired. For example, a plurality of images may be obtained from an album of the electronic device and a plurality of clustering factors for the plurality of images may be obtained, a plurality of images may be obtained from a cache of a chat-class application of the electronic device and a plurality of clustering factors for the plurality of images may be obtained, a server connected to the electronic device may obtain a plurality of images and a plurality of clustering factors for the plurality of images may be obtained, and so on. The image may include a still picture, a moving picture, a video, and the like, which is not limited herein. The corresponding units, spans and representation modes of the plurality of clustering factors may differ, and if the plurality of clustering factors may include the shooting time information and the shooting position information, the units, spans and representation modes of the shooting time information and the shooting position information differ.
In some embodiments, the plurality of clustering factors for the plurality of images may be configured automatically when the electronic device is shipped from a factory, or may be configured manually by a user during use of the electronic device, and the like, which is not limited herein. The clustering factor for the multiple images may be used to represent a clustering condition that has an influence on the clustering of the images, for example, the shooting position information corresponding to an image may influence the clustering of the multiple images, so that the shooting position information may be used as one clustering factor, the shooting time information corresponding to an image may influence the clustering of the multiple images, so that the shooting time information may be used as one clustering factor, the shooting object information corresponding to an image may influence the clustering of the multiple images, so that the shooting object information may be used as one clustering factor, and the like, which is not limited herein. As a manner, the plurality of clustering factors may include a first clustering factor, a second clustering factor, and an nth clustering factor … …, and the number of specific clustering factors may be adjusted and updated according to requirements, which is not limited herein.
In some embodiments, after acquiring the plurality of images, each image in the plurality of images may be treated as one data point. For example, assuming that the plurality of images includes a first image, a second image, a third image, a fourth image, and a fifth image, the first image, the second image, the third image, the fourth image, and the fifth image may be respectively taken as one data point, such as the first image as a first data point, the second image as a second data point, the third image as a third data point, the fourth image as a fourth data point, and the fifth image as a fifth data point.
Step S120: and acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radiuses corresponding to the plurality of clustering factors.
In this embodiment, after obtaining the plurality of images and the plurality of clustering factors for the plurality of images, the plurality of images may be screened, the target image may be obtained from the plurality of images, and the plurality of neighborhood radii corresponding to the plurality of clustering factors may be obtained.
In some embodiments, after acquiring the plurality of images, an image may be randomly selected from the plurality of images as a target image, images may be sequentially selected as the target image in the order of the plurality of images, and so on. In some embodiments, after acquiring the plurality of images, acquiring images that are not subjected to clustering processing from the plurality of images, and randomly selecting images from the images that are not subjected to clustering processing as target images, the images may be sequentially selected as the target images in the order of the images that are not subjected to clustering processing, the images may be sequentially selected as the target images in the reverse order, and the like.
In some embodiments, the electronic device may store a plurality of clustering factors and a plurality of neighborhood radii eps, and store a correspondence between the plurality of clustering factors and the plurality of neighborhood radii, where the correspondence may be that one clustering factor corresponds to one or more neighborhood radii. The corresponding relationship between the clustering factor and the neighborhood radius may be stored in the electronic device after being preset by a user, may be stored after being preset automatically by the electronic device, or may be sent to the electronic device after being preset by the server, which is not limited herein.
As one way, the electronic device may create a mapping table, which may include a plurality of clustering factors and a plurality of neighborhood radii, for example, the mapping table may be as shown in table 1, where a clustering factor is represented by a and a neighborhood radius is represented by B, and then, through the mapping table, the electronic device may correspondingly set and store the corresponding relationship between a clustering factor and a neighborhood radius locally in the electronic device.
TABLE 1
| Clustering factor | Neighborhood radius |
| A1 | B1 |
| A2 | B2 |
| A3 | B3 |
| A4 | B4 |
Further, in this embodiment, after obtaining the plurality of clustering factors, the mapping relationship table may be used to search for clustering factors consistent with the plurality of clustering factors for the plurality of images, and then based on the corresponding relationship between the clustering factors and the neighborhood radii in the mapping relationship table, the neighborhood radii corresponding to the clustering factors consistent with the plurality of clustering factors for the plurality of images may be searched for, and the neighborhood radii corresponding to the plurality of clustering factors for the plurality of images are determined. For example, when the plurality of clustering factors for the plurality of images are consistent with the clustering factor a1 and the clustering factor a2, it may be further determined that the neighborhood radius corresponding to the clustering factor a1 is B1 and the neighborhood radius B1 is B1, and the neighborhood radius corresponding to the clustering factor a2 is B2 and the neighborhood radius B2 is B2, based on the mapping relation table.
Step S130: and taking the target image as a data point center and the neighborhood radiuses as radiuses, and acquiring a plurality of neighborhoods of the target image, wherein the neighborhoods correspond to the neighborhood radiuses in a one-to-one mode.
In this embodiment, after the target image and the plurality of neighborhood radii corresponding to the plurality of clustering factors are obtained, the target image may be used as a data point center, the plurality of neighborhood radii are used as radii, and a plurality of neighborhoods of the data point center are drawn out and used as a plurality of neighborhoods of the target image.
In some embodiments, it is assumed that the plurality of neighborhood radii includes a first neighborhood radius, a second neighborhood radius, a third neighborhood radius … …, a nth neighborhood radius, and that the first neighborhood radius, the second neighborhood radius, and the third neighborhood radius … … are all different in units, spans, and representations. Then, a first neighborhood of the target image may be obtained with the target image as a center of the data point and the radius of the first neighborhood as a radius, a second neighborhood of the target image may be obtained with the target image as the center of the data point and the radius of the second neighborhood as a radius, a third neighborhood of the target image … … may be obtained with the target image as the center of the data point and the radius of the nth neighborhood as a radius, and an nth neighborhood of the target image may be obtained.
Step S140: and respectively acquiring images in each of the plurality of neighborhoods to obtain a plurality of image sets.
In this embodiment, after obtaining the plurality of neighborhoods based on the plurality of clustering factors, the images located in each of the plurality of neighborhoods may be obtained respectively, so as to obtain a plurality of image sets. After obtaining the plurality of neighborhoods, the image in each of the plurality of neighborhoods may be obtained, the image included in each of the neighborhoods is used as one image set, and the plurality of image sets are obtained based on the one image set corresponding to each of the plurality of neighborhoods.
In some embodiments, assuming that the plurality of neighborhoods includes a first neighborhood, a second neighborhood, and an nth neighborhood of a third neighborhood … …, an image located within the first neighborhood may be obtained, a first set of images is obtained based on the image located within the first neighborhood, an image located within the second neighborhood is obtained, a second set of images is obtained based on the image located within the second neighborhood, an image located within the third neighborhood is obtained, a third set of images … … is obtained based on the image located within the third neighborhood, an nth set of images is obtained based on the image located within the nth neighborhood, and a plurality of sets of images is obtained based on the first set of images, the second set of images, and the nth set of images … ….
Step S150: and obtaining a target image set based on the plurality of image sets, and clustering the images contained in the target image set.
In this embodiment, after acquiring a plurality of image sets, a target image set may be obtained based on the plurality of image sets, and images included in the target image set may be clustered.
In some embodiments, after acquiring the plurality of image sets, images included in each of the plurality of image sets may be acquired, respectively, where the images included in each of the image sets may include the same image or different images, and therefore, the images included in each of the image sets may be compared to acquire images existing in the plurality of image sets at the same time, and the target image set may be obtained based on the images existing in the plurality of image sets at the same time. In some embodiments, after acquiring the plurality of image sets, an intersection of the plurality of image sets may be acquired, with the intersection being the target image set.
In some embodiments, after obtaining the target image set, the images included in the target images may be aggregated, for example, an image classification set may be created based on the images included in the target image set, the images included in the target image set may be added to a pre-created image classification set, and so on. After the target image set is obtained, the clustering factor corresponding to the target image set may be obtained, after the clustering factor corresponding to the target image set is obtained, whether the pre-created image classification set includes the image classification set created based on the clustering factor corresponding to the target image set may be detected, when it is detected that the pre-created image classification set includes the image classification set created based on the clustering factor corresponding to the target image set, the image included in the target image set may be added to the pre-created image classification set to which the target image set belongs, and when it is detected that the pre-created image classification set does not include the image classification set created based on the clustering factor corresponding to the target image set, the image classification set may be created based on the clustering factor corresponding to the target image set, and the image included in the target image set may be added to the created image classification set.
An embodiment of the present application provides an image clustering method, which includes obtaining a plurality of images, and obtaining a plurality of clustering factors for the plurality of images, where each image in the plurality of images serves as a data point, obtaining a target image from the plurality of images, and obtaining a plurality of neighborhood radii corresponding to the plurality of clustering factors, and obtaining a plurality of neighborhoods of the target image with the target image as a data point center and the plurality of neighborhood radii as radii, where the plurality of neighborhoods correspond to the plurality of neighborhood radii one-to-one, obtaining an image located in each of the plurality of neighborhoods, respectively, obtaining a plurality of image sets, obtaining a target image set based on the plurality of image sets, and clustering images included in the target image set, so that the plurality of neighborhoods are obtained by setting the plurality of neighborhood radius divisions when the plurality of clustering factors exist, and clustering images included in the target image set obtained based on the plurality of image sets corresponding to the plurality of neighborhoods, and the accuracy of image clustering is improved.
Referring to fig. 2, fig. 2 is a schematic flow chart illustrating an image clustering method according to another embodiment of the present application. As will be explained in detail with respect to the flow shown in fig. 2, in this embodiment, the plurality of clustering factors include a first clustering factor and a second clustering factor, where the first clustering factor corresponds to a first neighborhood radius and the second clustering factor corresponds to a second neighborhood radius, and the image clustering method may specifically include the following steps:
step S210: a plurality of images are acquired, and a plurality of clustering factors for the plurality of images are acquired, wherein each image in the plurality of images serves as a data point.
Step S220: and acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radiuses corresponding to the plurality of clustering factors.
For the detailed description of steps S210 to S220, refer to steps S110 to S120, which are not described herein again.
Step S230: and acquiring a first neighborhood of the target image by taking the target image as a data point center and the radius of the first neighborhood as a radius, and acquiring a second neighborhood of the target image by taking the target image as the data point center and the radius of the second neighborhood as a radius.
In some embodiments, the plurality of clustering factors includes a first clustering factor and a second clustering factor, the first clustering factor corresponds to a first neighborhood radius, and the second clustering factor corresponds to a second neighborhood radius, and as one mode, the first clustering factor may include shooting position information, and the second clustering factor may include shooting time information. The shooting position information may be two-dimensional data, and the shooting time information may be one-dimensional data.
As a manner, after acquiring a first neighborhood radius corresponding to the target image and the shooting position information, the first neighborhood of the target image may be acquired by taking the target image as a data point center and taking the first neighborhood radius as a radius, where the first neighborhood may be a circular range with the target image as a center and N1 meters as a radius, which is not limited herein. As another mode, after acquiring the target image and the second neighborhood radius corresponding to the shooting time information, the second neighborhood of the target image may be obtained by taking the target image as the center of the data point and taking the second neighborhood radius as the radius, where the second neighborhood may be a circular range with the target image as the center and the N2 hours as the radius, which is not limited herein.
Step S240: and acquiring images in the first neighborhood to acquire a first image set, acquiring images in the second neighborhood to acquire a second image set.
In this embodiment, after obtaining the first neighborhood based on the shooting position information and the first neighborhood radius and the second neighborhood based on the shooting time information and the second neighborhood radius, the image located in the first neighborhood may be obtained, and the first image set based on the image located in the first neighborhood may be obtained, and the image located in the second neighborhood may be obtained, and the second image set based on the image located in the second neighborhood may be obtained.
Step S250: and obtaining a target image set based on the first image set and the second image set, and clustering the images contained in the target image set.
In this embodiment, after the first image set and the second image set are acquired, a target image set may be obtained based on the first image set and the second image set, and images included in the target image set may be clustered.
In some embodiments, after acquiring the first image set and the second image set, images included in the first image set and the second image set may be acquired, respectively, images included in the first image set and the second image set may be compared to acquire images concurrently existing in the first image set and the second image set, and the target image set may be obtained based on the images concurrently existing in the first image set and the second image set. In some embodiments, after acquiring the first image set and the second image set, an intersection of the first image set and the second image set may be acquired, and the intersection may be used as the target image set.
In another embodiment of the present application, an image clustering method obtains a plurality of images and obtains a plurality of clustering factors for the plurality of images, where each image in the plurality of images is used as a data point, obtains a target image from the plurality of images, obtains a plurality of neighborhood radii corresponding to the plurality of clustering factors, obtains a first neighborhood of the target image with the target image as a data point center and the first neighborhood radius as a radius, obtains a second neighborhood of the target image with the target image as a data point center and the second neighborhood radius as a radius, obtains an image located in the first neighborhood, obtains a first image set, obtains an image located in the second neighborhood, obtains a second image set, obtains a target image set based on the first image set and the second image set, and clusters images included in the target image set. Compared with the image clustering method shown in fig. 1, in this embodiment, when the first clustering factor and the second clustering factor exist, the first neighborhood is obtained by setting the first neighborhood radius partition corresponding to the first clustering factor, the second neighborhood is obtained by setting the second neighborhood radius partition corresponding to the second clustering factor, and the images included in the target image set obtained based on the first image set corresponding to the first neighborhood and the second image set corresponding to the second neighborhood are clustered, so that the accuracy of image clustering is improved.
Referring to fig. 3, fig. 3 is a schematic flow chart illustrating an image clustering method according to still another embodiment of the present application. As will be described in detail with respect to the flow shown in fig. 3, the image clustering method may specifically include the following steps:
step S310: a plurality of images are acquired, and a plurality of clustering factors for the plurality of images are acquired, wherein each image in the plurality of images serves as a data point.
Step S320: and acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radiuses corresponding to the plurality of clustering factors.
Step S330: and taking the target image as a data point center and the neighborhood radiuses as radiuses, and acquiring a plurality of neighborhoods of the target image, wherein the neighborhoods correspond to the neighborhood radiuses in a one-to-one mode.
Step S340: and respectively acquiring images in each of the plurality of neighborhoods to obtain a plurality of image sets.
For detailed description of steps S310 to S340, please refer to steps S110 to S140, which are not described herein again.
Step S350: and acquiring the intersection of the plurality of image sets, and taking the intersection as the target image set.
In some embodiments, after acquiring the plurality of image sets, an intersection of the plurality of image sets may be acquired, with the intersection being the target image set. For example, the plurality of image sets include a first image set and a second image set, the first image set includes a first image, a second image, a third image, and a fourth image, and the second image set includes a first image, a third image, a fourth image, and a fifth image, and intersecting the first image set and the second image set may obtain a target image set including the first image, the third image, and the fourth image.
Step S360: and clustering the images contained in the target image set.
For the detailed description of step S360, please refer to step S150, which is not described herein.
In an image aggregation method provided in another embodiment of the present application, a plurality of images are obtained, and a plurality of clustering factors for the plurality of images are obtained, where each image in the plurality of images is used as a data point, a target image is obtained from the plurality of images, a plurality of neighborhood radii corresponding to the plurality of clustering factors are obtained, a plurality of neighborhoods of the target image are obtained with the target image as a data point center and the plurality of neighborhood radii as radii, the plurality of neighborhoods correspond to the plurality of neighborhood radii one to one, the images located in each of the plurality of neighborhoods are obtained respectively, a plurality of image sets are obtained, an intersection of the plurality of image sets is obtained, the intersection is used as a target image set, and the images included in the target image set are clustered. Compared with the image aggregation method shown in fig. 1, in this embodiment, an intersection of the plurality of image sets is further obtained as a set of target images, so that the effect of improving the accuracy of image clustering is achieved by improving the accuracy of the determined target image set.
Referring to fig. 4, fig. 4 is a schematic flowchart illustrating an image clustering method according to another embodiment of the present application. As will be described in detail with respect to the flow shown in fig. 4, the image clustering method may specifically include the following steps:
step S410: a plurality of images are acquired, and a plurality of clustering factors for the plurality of images are acquired, wherein each image in the plurality of images serves as a data point.
Step S420: and acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radiuses corresponding to the plurality of clustering factors.
Step S430: and taking the target image as a data point center and the neighborhood radiuses as radiuses, and acquiring a plurality of neighborhoods of the target image, wherein the neighborhoods correspond to the neighborhood radiuses in a one-to-one mode.
Step S440: and respectively acquiring images in each of the plurality of neighborhoods to obtain a plurality of image sets.
Step S450: a target image set is obtained based on the plurality of image sets.
For the detailed description of steps S410 to S450, refer to steps S110 to S150, which are not described herein again.
Step S460: acquiring the number of images contained in the target image set.
In some embodiments, after the target image set is acquired, the number of images included in the target image set may be acquired. Since each of the plurality of images can be used as one data point, after the target image set is acquired, the number of data points included in the target image set can be acquired as the number of images included in the target image set.
Step S470: when the number of the images contained in the target image set is larger than a number threshold value, determining the target images as core objects, and clustering the images contained in the target image set based on the target images.
In some embodiments, a number threshold value MinPts for determining the number of images included in the target image set may be set and stored in advance. Therefore, in this embodiment, after acquiring the number of images included in the target image set, the number of images included in the target image set may be compared with a number threshold to determine whether the number of images included in the target image set is greater than the number threshold, where when the determination result indicates that the number of images included in the target image set is greater than the number threshold, it may be determined that the target image is in a certain cluster, a data point corresponding to the target image may be marked as a core point, the target image is determined as a core object, and the images included in the target image set are clustered based on the target image. Likewise, the above-described manner may also be employed for other images in the plurality of images until all images in the plurality of images have been clustered.
Referring to fig. 5, fig. 5 is a flowchart illustrating step S470 of the image clustering method illustrated in fig. 4 according to the present application. As will be explained in detail with respect to the flow shown in fig. 5, the method may specifically include the following steps:
step S471: when the number of images contained in the target image set is larger than a number threshold, determining the target images as core objects, and establishing a target cluster based on the target images.
In some embodiments, a number threshold value MinPts for determining the number of images included in the target image set may be set and stored in advance. Therefore, in this embodiment, after acquiring the number of images included in the target image set, the number of images included in the target image set may be compared with a number threshold to determine whether the number of images included in the target image set is greater than the number threshold, where when the determination result indicates that the number of images included in the target image set is greater than the number threshold, it may be determined that the target image is in a certain cluster, a data point corresponding to the target image may be marked as a core point, the target image may be determined as a core object, and a target cluster is established based on the target image. Based on the method, the target clusters established based on the target images are integrated with a plurality of clustering factors, so that the image classification precision is higher.
Step S472: and acquiring images with a specified relation with the target images from the target image set, and adding the images with the specified relation with the target images to the target cluster.
In some embodiments, after the target cluster is established based on the target image, an image having a specified relationship with the target image may be acquired from the target image set, and the image having the specified relationship with the target image may be added to the target cluster to classify images belonging to the target cluster among the plurality of images.
Referring to fig. 6, fig. 6 is a flowchart illustrating step S472 of the image clustering method illustrated in fig. 5 according to the present application. As will be explained in detail with respect to the flow shown in fig. 6, the method may specifically include the following steps:
step S4721: and acquiring a core object and an edge object communicated with the target image from the target image set.
In some embodiments, after the target cluster is established based on the target image, a core object and an edge object communicating with the target image may be acquired from the target image set. The core object communicated with the target image represents that the image is communicated with the target image, the number of the images contained in the neighborhood range of the image is larger than a number threshold, the edge object communicated with the target image represents that the image is communicated with the target image, and the number of the images contained in the neighborhood range of the image is not larger than the number threshold. As one way, communicating with the target image may include having a direct or indirect relationship with the target image, for example, the shooting position information of the image is close to the shooting position information of the target image, the shooting time information of the image is close to the shooting time information of the target image, and the like, which is not limited herein.
Step S4722: adding a core object and an edge object communicated with the target image to the target cluster.
In some embodiments, after obtaining the core object and the edge object communicated with the target image, the core object and the edge object communicated with the target image may be added to the target cluster to improve the accuracy of image classification.
Another embodiment of the present application provides an image clustering method, which obtains a plurality of images, and obtains a plurality of clustering factors for the plurality of images, wherein each of the plurality of images is taken as a data point, a target image is obtained from the plurality of images, and a plurality of neighborhood radii corresponding to the plurality of clustering factors are obtained, taking the target image as a data point center and a plurality of neighborhood radii as radii to obtain a plurality of neighborhoods of the target image, wherein the plurality of neighborhoods correspond to the plurality of neighborhood radii one to one, respectively acquire an image located in each of the plurality of neighborhoods, acquire a plurality of image sets, acquire a target image set based on the plurality of image sets, acquire the number of images included in the target image set, and when the number of the images contained in the target image set is larger than the number threshold value, determining the target image as a core object, and clustering the images contained in the target image set based on the target image. Compared with the image clustering method shown in fig. 1, in this embodiment, when the number of images included in the target image set is greater than the number threshold, the target image is determined as a core object, and the images included in the target image set are clustered based on the target image, so as to improve the accuracy of image clustering.
Referring to the drawings, fig. 7 is a block diagram illustrating an image clustering apparatus according to an embodiment of the present disclosure. As will be explained below with respect to the block diagram shown in fig. 7, the image clustering device 200 includes: an image acquisition module 210, a neighborhood radius acquisition module 220, a neighborhood acquisition module 230, an image set acquisition module 240, and an image clustering module 250, wherein:
an image obtaining module 210 configured to obtain a plurality of images, and obtain a plurality of clustering factors for the plurality of images, wherein each image in the plurality of images serves as a data point.
A neighborhood radius obtaining module 220, configured to obtain a target image from the multiple images, and obtain multiple neighborhood radii corresponding to the multiple clustering factors.
A neighborhood acquiring module 230, configured to acquire a plurality of neighborhoods of the target image by using the target image as a data point center and the plurality of neighborhood radii as radii, where the plurality of neighborhoods correspond to the plurality of neighborhood radii one to one.
Further, the plurality of clustering factors includes a first clustering factor and a second clustering factor, the first clustering factor corresponds to a first neighborhood radius, and the second clustering factor corresponds to a second neighborhood radius, the neighborhood obtaining module 230 includes: a neighborhood acquisition submodule, wherein:
and the neighborhood acquisition submodule is used for acquiring a first neighborhood of the target image by taking the target image as a data point center and the first neighborhood radius as a radius, and acquiring a second neighborhood of the target image by taking the target image as the data point center and the second neighborhood radius as the radius.
An image set obtaining module 240, configured to obtain images located in each of the multiple neighborhoods, respectively, to obtain multiple image sets.
Further, the plurality of clustering factors include a first clustering factor and a second clustering factor, the first clustering factor corresponds to a first neighborhood radius, and the second clustering factor corresponds to a second neighborhood radius, and the image set obtaining module 240 includes: an image set acquisition sub-module, wherein:
and the image set acquisition sub-module is used for acquiring the images in the first neighborhood to obtain a first image set, acquiring the images in the second neighborhood to obtain a second image set.
An image clustering module 250, configured to obtain a target image set based on the plurality of image sets, and cluster images included in the target image set.
Further, the image clustering module 250 includes: a first target image set acquisition sub-module and a first image clustering sub-module, wherein:
and the first target image set acquisition sub-module is used for acquiring the intersection of the plurality of image sets and taking the intersection as the target image set.
And the first image clustering submodule is used for clustering the images contained in the target image set.
Further, the image clustering module 250 includes: a second target image set acquisition submodule, an image quantity acquisition submodule and a second image clustering submodule, wherein:
and the second target image set acquisition sub-module is used for acquiring a target image set based on the plurality of image sets.
And the image quantity acquisition sub-module is used for acquiring the quantity of the images contained in the target image set.
And the second image clustering sub-module is used for determining the target image as a core object when the number of the images contained in the target image set is greater than a number threshold value, and clustering the images contained in the target image set based on the target image.
Further, the second image clustering sub-module includes: a target cluster creating unit and an image adding unit, wherein:
and the target cluster establishing unit is used for determining the target image as a core object when the number of the images contained in the target image set is greater than a number threshold value, and establishing a target cluster based on the target image.
And the image adding unit is used for acquiring an image with a specified relation with the target image from the target image set and adding the image with the specified relation with the target image to the target cluster.
Further, the image adding unit includes: an object acquisition subunit and an image addition subunit, wherein:
and the object acquisition subunit is used for acquiring a core object and an edge object communicated with the target image from the target image set.
And the image adding subunit is used for adding a core object and an edge object which are communicated with the target image to the target cluster.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, the coupling between the modules may be electrical, mechanical or other type of coupling.
In addition, functional modules in the embodiments of the present application may be integrated into one processing module, or each of the modules may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.
Referring to fig. 8, a block diagram of an electronic device 100 according to an embodiment of the present disclosure is shown. The electronic device 100 may be a smart phone, a tablet computer, an electronic book, or other electronic devices capable of running an application. The electronic device 100 in the present application may include one or more of the following components: a processor 110, a memory 120, and one or more applications, wherein the one or more applications may be stored in the memory 120 and configured to be executed by the one or more processors 110, the one or more programs configured to perform a method as described in the aforementioned method embodiments.
The Memory 120 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 120 may be used to store instructions, programs, code sets, or instruction sets. The memory 120 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing various method embodiments described below, and the like. The data storage area may also store data created by the electronic device 100 during use (e.g., phone book, audio-video data, chat log data), and the like.
Referring to fig. 9, a block diagram of a computer-readable storage medium according to an embodiment of the present application is shown. The computer-readable medium 300 has stored therein a program code that can be called by a processor to execute the method described in the above-described method embodiments.
The computer-readable storage medium 300 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 300 includes a non-volatile computer-readable storage medium. The computer readable storage medium 300 has storage space for program code 310 for performing any of the method steps of the method described above. The program code can be read from or written to one or more computer program products. The program code 310 may be compressed, for example, in a suitable form.
In summary, the image clustering method, the apparatus, the electronic device, and the storage medium according to the embodiments of the present application acquire a plurality of images and acquire a plurality of clustering factors for the plurality of images, wherein each image in the plurality of images serves as a data point, an object image is acquired from the plurality of images, a plurality of neighborhood radii corresponding to the plurality of clustering factors are acquired, a plurality of neighborhoods of the object image are acquired with the object image as a data point center and the plurality of neighborhood radii as radii, the plurality of neighborhoods are corresponding to the plurality of neighborhood radii one to one, the images in each of the plurality of neighborhoods are acquired, a plurality of image sets are acquired, the object image set is acquired based on the plurality of image sets, and the images included in the object image set are clustered, so that the plurality of neighborhoods are acquired by setting the plurality of neighborhood radii for partitioning when the plurality of clustering factors exist, and clustering images contained in a target image set obtained based on a plurality of image sets corresponding to a plurality of neighborhoods, so that the accuracy of image clustering is improved.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. An image clustering method, characterized in that the method comprises:
acquiring a plurality of images, and acquiring a plurality of clustering factors for the plurality of images, wherein each image in the plurality of images serves as a data point;
acquiring a target image from the plurality of images, and acquiring a plurality of neighborhood radii corresponding to the plurality of clustering factors;
taking the target image as a data point center and the neighborhood radiuses as radiuses, and acquiring a plurality of neighborhoods of the target image, wherein the neighborhoods correspond to the neighborhood radiuses one by one;
respectively acquiring images in each neighborhood of the plurality of neighborhoods to obtain a plurality of image sets;
and obtaining a target image set based on the plurality of image sets, and clustering the images contained in the target image set.
2. The method of claim 1, wherein the plurality of clustering factors includes a first clustering factor and a second clustering factor, the first clustering factor corresponds to a first neighborhood radius, the obtaining a plurality of neighborhoods of the target image centered on the target image and the plurality of neighborhood radii is performed by:
taking the target image as a data point center, taking the radius of the first neighborhood as a radius, acquiring a first neighborhood of the target image, and taking the target image as the data point center, taking the radius of the second neighborhood as a radius, acquiring a second neighborhood of the target image;
the respectively acquiring images located in each of the plurality of neighborhoods, obtaining a plurality of image sets, including:
and acquiring images in the first neighborhood to acquire a first image set, acquiring images in the second neighborhood to acquire a second image set.
3. The method of claim 2, wherein the first clustering factor comprises shot location information and the second clustering factor comprises shot time information.
4. The method of claim 1, wherein obtaining a target image set based on the plurality of image sets and clustering images included in the target image set comprises:
acquiring an intersection of the plurality of image sets, and taking the intersection as the target image set;
and clustering the images contained in the target image set.
5. The method of claim 1, wherein obtaining a target image set based on the plurality of image sets and clustering images included in the target image set comprises:
obtaining a target image set based on the plurality of image sets;
acquiring the number of images contained in the target image set;
when the number of the images contained in the target image set is larger than a number threshold value, determining the target images as core objects, and clustering the images contained in the target image set based on the target images.
6. The method of claim 5, wherein determining the target image as a core object and clustering the images included in the target image set based on the target image when the number of images included in the target image set is greater than a number threshold comprises:
when the number of images contained in the target image set is larger than a number threshold, determining the target images as core objects, and establishing target clusters based on the target images;
and acquiring images with a specified relation with the target images from the target image set, and adding the images with the specified relation with the target images to the target cluster.
7. The method according to claim 6, wherein the obtaining an image having a specified relationship with the target image from the target image set and adding the image having the specified relationship with the target image to the target cluster comprises:
acquiring a core object and an edge object communicated with the target image from the target image set;
adding a core object and an edge object communicated with the target image to the target cluster.
8. An image clustering apparatus, characterized in that the apparatus comprises:
an image acquisition module, configured to acquire a plurality of images, and acquire a plurality of clustering factors for the plurality of images, where each image in the plurality of images serves as a data point;
a neighborhood radius obtaining module, configured to obtain a target image from the multiple images, and obtain multiple neighborhood radii corresponding to the multiple clustering factors;
the neighborhood acquisition module is used for acquiring a plurality of neighborhoods of the target image by taking the target image as a data point center and taking the plurality of neighborhood radiuses as radiuses, wherein the plurality of neighborhoods correspond to the plurality of neighborhood radiuses in a one-to-one mode;
an image set acquisition module, configured to acquire an image located in each of the multiple neighborhoods, respectively, and obtain multiple image sets;
and the image clustering module is used for obtaining a target image set based on the plurality of image sets and clustering the images contained in the target image set.
9. An electronic device comprising a memory and a processor, the memory coupled to the processor, the memory storing instructions that, when executed by the processor, the processor performs the method of any of claims 1-7.
10. A computer-readable storage medium, having stored thereon program code that can be invoked by a processor to perform the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010917917.3A CN112070144B (en) | 2020-09-03 | 2020-09-03 | Image clustering method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010917917.3A CN112070144B (en) | 2020-09-03 | 2020-09-03 | Image clustering method, device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112070144A true CN112070144A (en) | 2020-12-11 |
| CN112070144B CN112070144B (en) | 2024-08-27 |
Family
ID=73665868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010917917.3A Active CN112070144B (en) | 2020-09-03 | 2020-09-03 | Image clustering method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112070144B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116127337A (en) * | 2022-09-08 | 2023-05-16 | 北京中关村科金技术有限公司 | Risk mining method, device, storage medium and equipment based on position and image |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390165A (en) * | 2012-05-10 | 2013-11-13 | 北京百度网讯科技有限公司 | Picture clustering method and device |
| CN103778146A (en) * | 2012-10-23 | 2014-05-07 | 富士通株式会社 | Image clustering device and method |
| CN106503656A (en) * | 2016-10-24 | 2017-03-15 | 厦门美图之家科技有限公司 | A kind of image classification method, device and computing device |
| CN107943984A (en) * | 2017-11-30 | 2018-04-20 | 广东欧珀移动通信有限公司 | Image processing method, device, computer equipment and computer-readable recording medium |
| CN107977431A (en) * | 2017-11-30 | 2018-05-01 | 广东欧珀移动通信有限公司 | Image processing method, device, computer equipment and computer-readable recording medium |
| CN108710916A (en) * | 2018-05-22 | 2018-10-26 | 重庆完美空间科技有限公司 | The method and device of picture classification |
| CN110110795A (en) * | 2019-05-10 | 2019-08-09 | 厦门美图之家科技有限公司 | Image classification method and device |
| CN111091152A (en) * | 2019-12-17 | 2020-05-01 | 上海云从汇临人工智能科技有限公司 | Image clustering method, system, device and machine readable medium |
| CN111291678A (en) * | 2020-02-06 | 2020-06-16 | 北京爱笔科技有限公司 | Face image clustering method and device based on multi-feature fusion |
| CN111488490A (en) * | 2020-03-31 | 2020-08-04 | 北京奇艺世纪科技有限公司 | Video clustering method, device, server and storage medium |
-
2020
- 2020-09-03 CN CN202010917917.3A patent/CN112070144B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390165A (en) * | 2012-05-10 | 2013-11-13 | 北京百度网讯科技有限公司 | Picture clustering method and device |
| CN103778146A (en) * | 2012-10-23 | 2014-05-07 | 富士通株式会社 | Image clustering device and method |
| CN106503656A (en) * | 2016-10-24 | 2017-03-15 | 厦门美图之家科技有限公司 | A kind of image classification method, device and computing device |
| CN107943984A (en) * | 2017-11-30 | 2018-04-20 | 广东欧珀移动通信有限公司 | Image processing method, device, computer equipment and computer-readable recording medium |
| CN107977431A (en) * | 2017-11-30 | 2018-05-01 | 广东欧珀移动通信有限公司 | Image processing method, device, computer equipment and computer-readable recording medium |
| US20190163767A1 (en) * | 2017-11-30 | 2019-05-30 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Image processing method, image processing device, computer device, and computer readable storage medium |
| CN108710916A (en) * | 2018-05-22 | 2018-10-26 | 重庆完美空间科技有限公司 | The method and device of picture classification |
| CN110110795A (en) * | 2019-05-10 | 2019-08-09 | 厦门美图之家科技有限公司 | Image classification method and device |
| CN111091152A (en) * | 2019-12-17 | 2020-05-01 | 上海云从汇临人工智能科技有限公司 | Image clustering method, system, device and machine readable medium |
| CN111291678A (en) * | 2020-02-06 | 2020-06-16 | 北京爱笔科技有限公司 | Face image clustering method and device based on multi-feature fusion |
| CN111488490A (en) * | 2020-03-31 | 2020-08-04 | 北京奇艺世纪科技有限公司 | Video clustering method, device, server and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 秦佳睿等: "自适应局部半径的DBSCAN聚类算法", 小型微型计算机系统, no. 10, 15 October 2018 (2018-10-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116127337A (en) * | 2022-09-08 | 2023-05-16 | 北京中关村科金技术有限公司 | Risk mining method, device, storage medium and equipment based on position and image |
| CN116127337B (en) * | 2022-09-08 | 2023-12-08 | 北京中关村科金技术有限公司 | Risk mining method, device, storage medium and equipment based on position and image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112070144B (en) | 2024-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021169723A1 (en) | Image recognition method and apparatus, electronic device, and storage medium | |
| JP6932360B2 (en) | Object search method, device and server | |
| CN109598250B (en) | Feature extraction method, device, electronic equipment and computer readable medium | |
| CN110490225B (en) | Scene-based image classification method, device, system and storage medium | |
| US20220343146A1 (en) | Method and system for temporal graph neural network acceleration | |
| CN110704189A (en) | Memory recovery method and device and electronic equipment | |
| CN111258649B (en) | Processor, chip and electronic equipment | |
| CN110955390B (en) | Data processing method, device, electronic equipment and storage medium | |
| CN112070144B (en) | Image clustering method, device, electronic equipment and storage medium | |
| CN109447489B (en) | Selection method and device of crowd sensing user, electronic equipment and storage medium | |
| WO2022007596A1 (en) | Image retrieval system, method and apparatus | |
| CN111310834A (en) | Data processing method and device, processor, electronic equipment and storage medium | |
| US11995890B2 (en) | Method and apparatus for tensor processing | |
| CN110286981A (en) | The display methods and display system of the use state of virtual cloud desktop server | |
| CN110674086A (en) | Data merging method and device, electronic equipment and storage medium | |
| CN114365111A (en) | Method and device for updating label data, electronic equipment and storage medium | |
| CN112069337A (en) | Picture processing method and device, electronic equipment and storage medium | |
| CN113989559A (en) | Method, device and equipment for determining probability threshold of classification model and storage medium | |
| CN112214627B (en) | Search method, readable storage medium, and electronic device | |
| CN113238852A (en) | Task allocation method and device, electronic equipment and storage medium | |
| CN114048847A (en) | Method, device and equipment for caching graph neural network data and storage medium | |
| CN113065597A (en) | Clustering method, device, equipment and storage medium | |
| CN112597161A (en) | Data processing method, electronic device and readable storage medium | |
| CN110727405A (en) | Data processing method and device, electronic equipment and computer readable medium | |
| CN107735800B (en) | Image processing method and related 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 |