CN116340562A - Image labeling method and device, electronic equipment and storage medium - Google Patents

Image labeling method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN116340562A
CN116340562A CN202310136172.0A CN202310136172A CN116340562A CN 116340562 A CN116340562 A CN 116340562A CN 202310136172 A CN202310136172 A CN 202310136172A CN 116340562 A CN116340562 A CN 116340562A
Authority
CN
China
Prior art keywords
labeling
image
target
target account
pushed
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310136172.0A
Other languages
Chinese (zh)
Inventor
安超
沈智慧
辛安民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikrobot Co Ltd
Original Assignee
Hangzhou Hikrobot Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikrobot Co Ltd filed Critical Hangzhou Hikrobot Co Ltd
Priority to CN202310136172.0A priority Critical patent/CN116340562A/en
Publication of CN116340562A publication Critical patent/CN116340562A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/5866Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53Querying
    • G06F16/535Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53Querying
    • G06F16/538Presentation of query results
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Image Analysis (AREA)

Abstract

The embodiment of the application provides an image labeling method, an image labeling device, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: firstly, determining working groups to which a plurality of target accounts which are logged in parallel respectively belong, then determining a labeling object corresponding to the working group to which the target account belongs according to a first corresponding relation between the working groups and the labeling object, generating a labeling task of the target account, then determining a designated labeling object of the target account, retrieving images to be labeled of the target account based on the designated labeling object, determining images to be pushed of each target account, pushing the images to be pushed of each target account to target terminal equipment, enabling the target terminal equipment to display the images to be pushed, and finally obtaining labeling information of the images to be pushed of the target account. By applying the embodiment of the application, the accuracy of image annotation is improved under the condition of ensuring the image annotation efficiency.

Description

Image labeling method and device, electronic equipment and storage medium
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to an image labeling method, an image labeling device, an electronic device, and a storage medium.
Background
With the rapid development of computer technology, image processing technology is increasingly applied to various fields, and the demand for image processing models is also increasing, and in general, sample images for training the image processing models can be obtained by labeling image data. For example, category labeling is performed on items in the image to obtain a sample image for training the image classification model, category labeling is performed on vehicles in the image to obtain a sample image for the vehicle classification model, and the like.
In the related art, in order to improve the efficiency of image labeling, the same staff is usually used for labeling various data of an image, however, because different labeling modes may be adopted for labeling different data of an image and different labeling contents need to be added, the situation of labeling errors easily occurs, which results in lower accuracy of the obtained sample image and further results in lower accuracy of the trained image processing model.
Based on the above, how to improve the accuracy of image labeling under the condition of ensuring the image labeling efficiency is a technical problem to be solved currently.
Disclosure of Invention
An embodiment of the application aims to provide an image labeling method, an image labeling device, electronic equipment and a storage medium, which are used for improving the accuracy of image labeling under the condition of ensuring the image labeling efficiency. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present application provides an image labeling method, where the method includes:
determining a working group to which a plurality of target account numbers logged in parallel respectively belong;
for each target account, determining a labeling object corresponding to the work group to which the target account belongs according to a first corresponding relation between the work group and the labeling object, and generating a labeling task of the target account based on the determined labeling object; the labeling task of each target account comprises the following steps: labeling objects corresponding to the work groups to which the target account belongs;
determining a designated labeling object of each target account, and searching an image to be labeled of the target account based on the designated labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
Determining an image to be pushed of each target account according to the image to be marked of each target account;
pushing the image to be pushed of each target account to the target terminal equipment logged in with the target account, so that the target terminal equipment displays the received image to be pushed;
for each target account, acquiring labeling information of an image to be pushed of the target account; the annotation information is sent by the target terminal equipment, and the annotation information of the specified annotation object in the image to be pushed of the target account is related to the target account.
Optionally, in a specific implementation manner, the determining, according to the image to be annotated of each target account, the image to be pushed of each target account includes: determining the image to be marked of each target account as an image to be pushed of the target account; or determining the sub-image of the specified labeling object of each target account in the images to be labeled of the target account as the images to be pushed of the target account.
Optionally, in a specific implementation manner, pushing the image to be pushed of each target account to the target terminal device includes: pushing a specified number of residual images to the target terminal equipment aiming at each target account; wherein, the residual image is: an image to be pushed, which is not pushed to the target terminal equipment, of the target account; the method further comprises the steps of: returning to the step of pushing the specified number of residual images to the target terminal equipment after the labeling information of all pushed images to be pushed of the target account is obtained for each target account until the target account terminates login, or the residual images are not existed; or, for each target account, after the labeling information of the preset number of images to be pushed of the target account is obtained, pushing the preset number of residual images to the target terminal equipment until the target account terminates login, or the residual images are not existed.
Optionally, in a specific implementation manner, for each target account, determining, according to a first correspondence between a relevant working group and a labeling object, the labeling object corresponding to the working group to which the target account belongs includes: for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme; the labeling scheme comprises the following steps: a first correspondence relationship between the work group and the labeling object, and a subordinate relationship between the account number and the work group.
Optionally, in a specific implementation manner, the labeling scheme further includes: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode; before pushing the image to be annotated corresponding to each target account to the target terminal device, the method further comprises: according to the second corresponding relation in the labeling scheme, determining candidate labeling contents and labeling modes of the designated labeling objects of each target account, wherein the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account; pushing the image to be pushed of each target account to the target terminal device, so that the target terminal device displays the received image to be pushed, including: pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode; the obtaining, for each target account, labeling information of an image to be pushed of the target account includes: and acquiring labeling information of the image to be pushed of the target account selected from candidate labeling contents of the target account according to the labeling mode of the target account aiming at each target account.
Optionally, in a specific implementation manner, the method further includes: and adding new entries related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting the current entries related to the object relation between the working group and the labeling object, and returning to the step of determining the multiple target account numbers which are logged in parallel.
Optionally, in a specific implementation manner, the method further includes: in the labeling scheme, adding new entries related to the affiliation of the account and the working group, and/or deleting current entries related to the affiliation of the account and the working group, and returning to the step of determining a plurality of target accounts which are logged in parallel.
Optionally, in a specific implementation manner, the method further includes: determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image with target integration information; and sending each target image carrying the target integration information to the designated equipment.
In a second aspect, an embodiment of the present application provides an image labeling apparatus, including:
The work group determining module is used for determining work groups to which a plurality of target account numbers logged in parallel respectively belong;
the task generating module is configured to determine, for each target account, a labeling object corresponding to a work group to which the target account belongs according to a first correspondence between the work group and the labeling object, and generate a labeling task of the target account based on the determined labeling object, where the labeling task of each target account includes: labeling objects corresponding to the work groups to which the target account belongs;
the image retrieval module is used for determining a specified labeling object of each target account and retrieving an image to be labeled of the target account based on the specified labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
The image determining module is used for determining an image to be pushed of each target account according to the image to be annotated of each target account;
the image pushing module is used for pushing the image to be pushed of each target account to the target terminal equipment so that the target terminal equipment displays the received image to be pushed;
the information acquisition module is used for acquiring labeling information of the image to be pushed of each target account; the annotation information is sent by the target terminal equipment, and the annotation information of the specified annotation object in the image to be pushed of the target account is related to the target account.
Optionally, in a specific implementation manner, the image determining module is specifically configured to: determining the image to be marked of each target account as an image to be pushed of the target account; or determining the sub-image of the target labeling object corresponding to the target account in the image to be labeled of each target account as the image to be pushed of the target account.
Optionally, in a specific implementation manner, the image pushing module is specifically configured to: pushing a specified number of residual images to target terminal equipment aiming at each target account; wherein, the residual image is: an image to be pushed, which is not pushed to target terminal equipment, of the target account; the apparatus further comprises:
The residual image pushing module is used for triggering the image pushing module after the labeling information of all pushed images to be pushed of each target account is obtained, until the target account stops logging, or the residual images are not existed; or, for each target account, after the labeling information of the preset number of images to be pushed of the target account is obtained, pushing the preset number of residual images to the target terminal device until the target account terminates login, or the residual images do not exist.
Optionally, in a specific implementation manner, the task generating module is specifically configured to: for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme; the labeling scheme comprises the following steps: a first correspondence relationship between the work group and the labeling object, and a subordinate relationship between the account number and the work group.
Optionally, in a specific implementation manner, the labeling scheme further includes: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode; the apparatus further comprises:
The content mode determining module is used for determining candidate labeling contents and labeling modes of the designated labeling objects of each target account according to the second corresponding relation in the labeling scheme before pushing the images to be labeled corresponding to each target account to the target terminal equipment, and the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account;
the image pushing module is specifically configured to: pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode;
the information acquisition module is specifically configured to: and acquiring labeling information of the image to be pushed of the target account selected from candidate labeling contents of the target account according to the labeling mode of the target account aiming at each target account.
Optionally, in a specific implementation manner, the apparatus further includes:
and the item adding and deleting first module is used for adding new items related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting current items related to the object relation between the working group and the labeling object, and returning to the step of determining the multiple target account numbers which are logged in parallel.
Optionally, in a specific implementation manner, the apparatus further includes:
and the second item adding and deleting module is used for adding new items related to the affiliation of the account and the working group in the labeling scheme, and/or deleting the current items related to the affiliation of the account and the working group, and returning to the step of determining the multiple target accounts which are logged in parallel.
Optionally, in a specific implementation manner, the apparatus further includes:
the information integration module is used for determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image carrying target integration information;
and the image sending module is used for sending each target image carrying the target integration information to the designated equipment.
In a third aspect, an embodiment of the present application provides an electronic device, including:
a memory for storing a computer program;
and the processor is used for realizing the steps of any image marking method provided in the first aspect when executing the program stored in the memory.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having stored therein a computer program which, when executed by a processor, implements the steps of any one of the image labeling methods provided in the first aspect above.
In a fifth aspect, embodiments of the present application also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the image annotation method according to any of the first aspects described above.
The beneficial effects of the embodiment of the application are that:
the technical scheme provided by the embodiment of the application is applied, firstly, the working groups to which the multiple target accounts which are logged in parallel respectively belong are determined, then, for each target account, the labeling object corresponding to the working group to which the target account belongs is determined according to the first corresponding relation between the working group and the labeling object, the labeling task of the target account is generated based on the determined labeling object, then, for each target account, the designated labeling object of the target account is determined, the image to be labeled of the target account is searched based on the designated labeling object, then, the image to be pushed of each target account is determined according to the image to be labeled of each target account, further, the image to be pushed of each target account is pushed to the target terminal equipment logged with the target account, so that the terminal equipment displays the received image to be pushed, and finally, the labeling information of the image to be pushed of the target account is obtained for each target account.
Based on the method, for a plurality of objects to be marked, the objects to be marked are distributed to different working groups, so that a first corresponding relation between the working groups and the marked objects is established, a plurality of account numbers for image marking are distributed in the working groups, the working groups to which the account numbers belong are determined after the staff logs in the account numbers, the marked objects are determined according to the first corresponding relation, and marking of the marked objects is completed. Therefore, when labeling is carried out, different staff can log in the account in parallel, and accordingly corresponding images to be labeled can be labeled in parallel. Because a plurality of accounts are logged in parallel, a plurality of labeling objects of a plurality of images can be labeled simultaneously, the image labeling efficiency can be ensured, and because different labeling objects are distributed by accounts of different staff, the staff distributed by each labeling object does not need to pay attention to all the labeling objects, so that the staff is not easy to mistake when labeling the images, the image labeling accuracy can be improved under the condition of ensuring the image labeling efficiency, the image sample accuracy can be improved, and the subsequent image model training accuracy can be improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other embodiments may also be obtained according to these drawings to those skilled in the art.
Fig. 1 is a flowchart of a first image labeling method according to an embodiment of the present application.
Fig. 2a is a flowchart of a second image labeling method according to an embodiment of the present application.
Fig. 2b is a flowchart of a third image labeling method according to an embodiment of the present application.
Fig. 3 is a flowchart of a fourth image labeling method according to an embodiment of the present application.
Fig. 4 is a flowchart of a fifth image labeling method according to an embodiment of the present application.
Fig. 5 is a schematic structural diagram of an image labeling device according to an embodiment of the present application.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. Based on the embodiments herein, a person of ordinary skill in the art would be able to obtain all other embodiments based on the disclosure herein, which are within the scope of the disclosure herein.
In the related art, in order to improve the efficiency of image labeling, the same staff is usually used for labeling various data of an image, however, because different labeling modes may be adopted for labeling different data of an image and different labeling contents need to be added, the situation of labeling errors easily occurs, which results in lower accuracy of the obtained sample image and further results in lower accuracy of the trained image processing model. Based on the above, how to improve the accuracy of image labeling under the condition of ensuring the image labeling efficiency is a technical problem to be solved currently.
In order to solve the technical problems, an embodiment of the present application provides an image labeling method.
The method can be suitable for various application scenes needing image annotation, for example, sample images used for image model training are obtained by annotating the images. The method can be applied to various types of electronic equipment such as servers, desktop computers, mobile phones and the like, and is hereinafter referred to as electronic equipment. Based on this, the embodiment of the present invention does not limit the application scenario and execution subject of the method.
The electronic equipment can execute the image labeling method through the installed client with the image labeling function; the image labeling method can also be executed by a self-configured module with an image labeling function. When the electronic equipment executes the image marking method through the client, the module for realizing the image marking method can be used as a plug-in of the client; the client may also be a client dedicated to performing the object detection method. This is reasonable.
The image labeling method provided by the embodiment of the application can comprise the following steps:
determining a working group to which a plurality of target account numbers logged in parallel respectively belong;
for each target account, determining a labeling object corresponding to the work group to which the target account belongs according to a first corresponding relation between the work group and the labeling object, and generating a labeling task of the target account based on the determined labeling object; the labeling task of each target account comprises the following steps: labeling objects corresponding to the work groups to which the target account belongs;
determining a designated labeling object of each target account, and searching an image to be labeled of the target account based on the designated labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
Determining an image to be pushed of each target account according to the image to be marked of each target account;
pushing the image to be pushed of each target account to the target terminal equipment so that the target terminal equipment displays the received image to be pushed;
for each target account, acquiring labeling information of an image to be pushed of the target account; the annotation information is sent by the target terminal equipment, and the annotation information of the specified annotation object in the image to be pushed of the target account is related to the target account.
The technical scheme provided by the embodiment of the application is applied, firstly, the working groups to which the multiple target accounts which are logged in parallel respectively belong are determined, then, for each target account, the labeling object corresponding to the working group to which the target account belongs is determined according to the first corresponding relation between the working group and the labeling object, the labeling task of the target account is generated based on the determined labeling object, then, for each target account, the designated labeling object of the target account is determined, the image to be labeled of the target account is searched based on the designated labeling object, then, the image to be pushed of each target account is determined according to the image to be labeled of each target account, further, the image to be pushed of each target account is pushed to the target terminal equipment logged with the target account, so that the terminal equipment displays the received image to be pushed, and finally, the labeling information of the image to be pushed of the target account is obtained for each target account.
Based on the method, for a plurality of objects to be marked, the objects to be marked are distributed to different working groups, so that a first corresponding relation between the working groups and the marked objects is established, a plurality of account numbers for image marking are distributed in the working groups, the working groups to which the account numbers belong are determined after the staff logs in the account numbers, the marked objects are determined according to the first corresponding relation, and marking of the marked objects is completed. Therefore, when labeling is carried out, different staff can log in the account in parallel, and accordingly corresponding images to be labeled can be labeled in parallel. Because a plurality of accounts are logged in parallel, a plurality of labeling objects of a plurality of images can be labeled simultaneously, the image labeling efficiency can be ensured, and because different labeling objects are distributed by accounts of different staff, the staff distributed by each labeling object does not need to pay attention to all the labeling objects, so that the staff is not easy to mistake when labeling the images, the image labeling accuracy can be improved under the condition of ensuring the image labeling efficiency, the image sample accuracy can be improved, and the subsequent image model training accuracy can be improved.
Before a specific description of an image labeling method provided by the embodiment of the present invention is given, first a description is given of a first correspondence related to the method:
in general, for an image, multiple objects in the image may be labeled, so that there may be multiple labeled objects in an image to be labeled, e.g., a face, an object, a vehicle, etc. Therefore, a plurality of labeling objects in the image to be labeled can be determined according to the labeling requirement, and staff for labeling the labeling objects can be allocated to the labeling objects.
Determining a labeling object, and distributing a working group for image labeling for each labeling object, wherein the image labeling is completed by a worker through logging in an account distributed in the working group on a terminal device, so that the worker marks the determined labeling object according to a first corresponding relation between the working group and the labeling object after determining the working group to which the account belongs through logging in the corresponding account in the working group; the account number logged in by the staff may be a fixed account number or a randomly determined account number, which is not particularly limited in this application.
Because the number of the images to be marked is large, in order to improve the image marking efficiency, the marking objects for marking the images are required to be divided into a plurality of working groups, the marking objects corresponding to each working group are the same object, each worker can mark the images of the marking objects corresponding to each worker through logging in the account numbers in the working groups, each marking object corresponds to at least one working group, and therefore a first corresponding relation between the working groups and the marking objects can be established. The first correspondence characterizes marking authority in the work group, and when a user logs in an account belonging to the work group containing the first correspondence, the user only marks the marking authority of an object corresponding to the account, but cannot mark other objects which are not corresponding to the account.
For example, two annotation objects exist: the method comprises the steps that the labeling objects of eyes and a mouth are respectively the eyes and the mouth, the account number allocated by the first working group is A, the account number allocated by the second working group is B, so that the eyes and the account number A have a first corresponding relation, the mouth and the account number B have a first corresponding relation, a worker C logs in the account number A on the electronic equipment, so that the eyes can be labeled, and the worker D logs in the account number B on the electronic equipment, so that the mouth can be labeled.
Optionally, a labeling scheme including a first correspondence relation between the working group and the labeling object may be created, where a corresponding manner of the first correspondence relation between the working group and the labeling object in the labeling scheme may be that an entry table corresponding to the working group and the labeling object one to one is created; or a database for building the corresponding relation between the working group and the labeling object.
Optionally, because candidate labeling contents and labeling modes corresponding to different labeling objects can be different, the labeling scheme can further include: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode; the determining process of the second corresponding relation is as follows:
selecting proper candidate labeling contents for each labeling object according to the labeling objects, and determining the most proper labeling mode according to the difference of the candidate labeling contents so as to determine a second corresponding relation among the labeling objects, the candidate labeling contents and the labeling modes;
the candidate labeling content comprises the following components: labeling one or more of the shape, color, number and size of the objects; the labeling mode comprises the following steps: one or more of the hook label and the text label are selected;
in general, the presentation form of the second correspondence relationship between the labeling object, the candidate labeling content and the labeling mode corresponding to the account may be represented as a table corresponding to the labeling object, the candidate labeling content and the labeling mode one by one, which are established in the labeling scheme, or may be a database including the labeling object, the candidate labeling content and the labeling mode, which is established in the labeling scheme, and the candidate labeling content and the labeling mode may be searched according to the determined corresponding labeling object of the account, which is reasonable.
For example, the existing labeling objects are the types of the vehicles and the colors of the vehicles, the candidate labeling contents are the types 1, 2 and 3 of the vehicles and the colors a, b and c of the vehicles, and the labeling modes are the hook labeling and the text labeling; the labeling objects of the third staff and the fourth staff are the type of the traffic tool and the color of the traffic tool respectively, the candidate labeling content of the third staff is the type 1, the type 2 and the type 3 of the traffic tool, the candidate labeling content of the fourth staff is the color a, the color B and the color c of the traffic tool, and the most suitable labeling modes for the candidate labeling content are the hook labeling, so that the type of the traffic tool has a second corresponding relation with the type 1, the type 2, the type 3 and the hook labeling, the color of the traffic tool has a second corresponding relation with the color a, the color B, the color c and the hook labeling, the third staff logs in the account A on the electronic equipment, the type of the traffic tool can be labeled by the hook labeling, and the fourth staff logs in the account B on the electronic equipment, and the color of the traffic tool can be labeled by the hook labeling.
Next, based on the above description, a specific description is given of an image labeling method provided in the embodiments of the present application with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart in an image labeling method provided in an embodiment of the present application, as shown in fig. 1, the method includes steps S101 to S106:
step S101: determining a working group to which a plurality of target account numbers logged in parallel respectively belong;
after the target account numbers are distributed to each working group, a plurality of workers can log in the respective distributed target account numbers in parallel, and the electronic equipment can detect that a plurality of account numbers are logged in parallel; therefore, after determining a plurality of parallel logged-in target account numbers, the electronic device can determine the working group to which each target account number belongs.
When the target account numbers are distributed, all the target account numbers can be distributed to each work group evenly; the number of the target accounts contained in each working group can be distributed according to actual demands, for example, when the labeling difficulty of labeling objects corresponding to some working groups is relatively high, the number of the target accounts in the working groups can be increased; for another example, when the annotation completion time limit in a portion of the workgroup is relatively short, it may be reasonable to increase the number of target accounts in the workgroup.
In addition, the same target account number may be allocated to one work group or may be allocated to a plurality of work groups, which is not particularly limited in this application.
Step S102: for each target account, determining a labeling object corresponding to the work group to which the target account belongs according to a first corresponding relation between the work group and the labeling object, and generating a labeling task of the target account based on the determined labeling object;
the labeling task of each target account comprises the following steps: labeling objects corresponding to the work groups to which the target account belongs;
the first correspondence between the working groups and the labeling objects is a one-to-one correspondence, that is, each working group corresponds to only one labeling object.
After the target account establishes the labeling objects based on the first corresponding relation, a labeling task can be generated, and the labeling task comprises all labeling objects required to be labeled by the target account, so that a worker using the target account can clearly and intuitively determine own work tasks and select the labeling objects required to be labeled at first;
optionally, the labeling task may include one or more of labeling objects, time limit requirements of each labeling object, and unlabeled images of each labeling object.
Step S103: determining a designated labeling object of each target account, and searching an image to be labeled of the target account based on the designated labeling object;
Wherein, the appointed label object is: the received object selection instruction indicates the object in the labeling task of the target account, and the image to be labeled is: the method comprises the steps that a specified labeling object contained in a preset image is not labeled, is not pushed out of target terminal equipment, and an object indicated by a received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: and logging in the terminal equipment with the target account number.
For example, the labeling objects corresponding to the working groups to which the target account number a and the target account number B belong include eyes, when the worker 1 logs in the target account number a at the terminal device 1, and the designated labeling object indicated by the received object selection instruction is the eyes, and when the worker 2 logs in the target account number B at the terminal device 2, and the designated labeling object indicated by the received object selection instruction is the eyes, then the image to be labeled, which is the eyes, of the same labeling object can be randomly pushed to any terminal device on which the target account number a or the target account number B is logged. However, for the image to be annotated, when it is pushed to the electronic device with the target account a, it cannot be pushed to the electronic device with the target account B any more, and vice versa.
For example, when the staff member 1 logs in the target account a according to the specified labeling object indicated by the received object selection instruction, and the staff member 2 logs in the target account B according to the specified labeling object indicated by the received object selection instruction, then the same image to be labeled can be pushed to the terminal device with the target account a to label the eyes or to the terminal device with the target account B to label the hair. At this time, the to-be-annotated image pushed to the target account number a may not show hair, and the to-be-annotated image pushed to the target account number B may not show eyes.
Optionally, when each target account only belongs to any working group and the working group has only one labeling object, that is, when the target accounts are in one-to-one correspondence with the working group and the labeling objects, the same labeling image containing the labeling object is pushed to a terminal device registered by any target account in the working group and is not pushed to a terminal device registered by other target accounts in the working group.
After the staff using the target account number obtains the labeling task, the staff can intuitively know all labeling objects which need to be labeled; when only one labeling object exists in the labeling task list, labeling the labeling object by default; when a plurality of labeling objects exist in the labeling task, a worker can select one labeling object which is labeled preferentially, and after the selection of the labeling object is determined, only the image to be labeled of the labeling object is searched.
Optionally, after the staff determines the labeling object, the terminal device logged in by the target account used by the staff only displays the image to be labeled retrieved for the labeling object; the relevant information interfaces of the rest marked objects are moved to a small window, and after marking the marked image of the marked object is completed, one marked object can be determined again to search the marked image.
In general, a plurality of images to be labeled can be preset during image labeling, so that after the target labeling object corresponding to each target account is determined, the image to be labeled by the staff corresponding to the target account can be selected from the preset images, so as to complete image labeling.
The electronic equipment can search the image to be marked of each target account in the preset image, so that the image to be pushed can be pushed to each target account conveniently; in order to ensure that a worker corresponding to the target account can label a target label object required to be labeled based on the first corresponding relation, the target label object determined by the target account according to the first corresponding relation is searched in a preset image to be in an unlabeled state, and meanwhile, in order to avoid repeated labeling of the same target label object, the target label object is ensured not to be pushed to other accounts which can label the target label object when the preset image is searched.
Therefore, for each account, the image to be annotated retrieved by the electronic device is the image which is not annotated by the included target annotation object corresponding to the target account and is not pushed to the terminal device which logs in the other accounts except the target account of the same target annotation object corresponding to the target account.
Step S104: determining an image to be pushed of each target account according to the image to be marked of each target account;
after the image to be annotated corresponding to each target account is retrieved, pushing the image to the electronic equipment of the target account corresponding to the image is hoped to enable staff corresponding to the target account to carry out subsequent annotation, so that the image to be pushed of the target account is determined in the image to be annotated of the target account.
Optionally, in one specific implementation, as shown in fig. 2a, step S104 is described above: according to the image to be annotated of each target account, determining the image to be pushed of each target account may include the following step S201:
step S201: determining the image to be marked of each target account as an image to be pushed of the target account;
the image to be pushed, which is determined to be the target account, is a complete image to be annotated of the target account; the image to be annotated not only includes a target annotation object corresponding to the target account number, but also includes other annotation objects except the target annotation object, and because the authority of the account number is limited by the first corresponding relation, a worker of the account number can only annotate the image of the target annotation object determined by the account number based on the first corresponding relation, and the operable range is limited to the target annotation object.
Optionally, as shown in fig. 2b, the step S104 may include the following step S202:
step S202: and determining the sub-image of the designated labeling object of each target account included in the images to be labeled of the target account as the images to be pushed of the target account.
The sub-image of the target labeling object may be obtained by intercepting a preset image before retrieving the image to be labeled of each target account, and intercepting the preset image into a plurality of sub-images according to the labeling object;
the sub-image of the target labeling object may further be that after the target login account determines the target labeling object corresponding to the target login account, all preset images including the target labeling object are intercepted, and only the image including the target labeling object corresponding to the target login account is reserved as the sub-image.
Based on the method, the labeling range of the image to be labeled, which is acquired by the terminal equipment with the target account, can be ensured to be only the target labeling object corresponding to the target account, so that the labeling object which does not accord with the first corresponding relation with the target labeling account can be effectively prevented from being not labeled, and the accuracy of image labeling is improved.
Step S105: pushing the image to be pushed of each target account to the target terminal equipment so that the target terminal equipment displays the received image to be pushed;
step S106: for each target account, acquiring labeling information of an image to be pushed of the target account;
The labeling information is the labeling information of a specified labeling object in the image to be pushed of the target account, which is sent by the target terminal device, and the specified labeling object is the target labeling object corresponding to the target account.
After determining the image to be pushed of each target account, in order to enable a worker to label a target labeling object, the image to be pushed can be pushed to a target terminal logged in by an account used by the worker, when the target terminal equipment receives the image to be pushed, the image to be pushed can be displayed in the target terminal equipment, and a labeling result obtained after labeling the labeling object in the image to be pushed by the worker is labeling information of the labeling object.
The electronic device can directly push all the images to be pushed corresponding to each target account to the terminal device on which the target account is registered; the images to be pushed can be divided into a plurality of batches according to the designated quantity and pushed to the terminal equipment logged in with the target account; for clarity of the line, step S105 will be described in detail later.
Optionally, in a specific implementation manner, step S105 is described above: pushing the image to be pushed of each target account to the target terminal device may include the following step B1:
Step B1: pushing a specified number of residual images to the target terminal equipment for each target account until the target account stops logging in, or the residual images do not exist;
among these, the residual image is: an image to be pushed, which is not pushed to target terminal equipment, of the target account;
in this specific implementation manner, before the first push, the residual image is the determined image to be pushed to the target terminal device.
When the images to be annotated corresponding to each target account number are pushed to the target terminal equipment, the appointed number of residual images can be pushed to the target terminal equipment, and after the target account number is completed and the images of the appointed number of residual images are annotated, the electronic equipment can determine the number of the residual images so as to facilitate the continuous pushing of the appointed number of residual images to the target terminal equipment.
When the number of the residual images is larger than the specified number, pushing the specified number of the residual images to the target terminal equipment; when the residual images exist and the number of the residual images is smaller than the number of the specified number of the residual images, pushing all the residual images to the target terminal equipment and stopping pushing after the pushing; and stopping pushing the residual image to the target terminal equipment when the residual image does not exist.
In the above process, when the target account is detected to log out, the electronic device stops pushing images to the target terminal device, and for the residual images which are not pushed in the electronic device, the residual images can be restored to be pushed when the target account logs in next time, or when other accounts which have a first corresponding relation with the labeling object in the residual images and are out of the target account are detected to log in the terminal device, the residual images which are not pushed can be pushed to a new account by a method of pushing the residual images.
That is, in the electronic device, the number of the remaining images may be determined, and the number of the remaining images pushed to the terminal device in the next batch may be determined according to the determination result.
For example, if the image to be annotated corresponding to a target account is 100, 30 pieces of residual images can be pushed to the target terminal device for the target account, when the target account finishes annotating the images of the 30 pieces of residual images, the 30 pieces of residual annotated images are pushed to the target account until the last 10 pieces of residual images are not pushed, at this time, the 10 pieces of images are all pushed to the target terminal device,
Based on the method, a certain number of images to be pushed can be pushed to the target terminal equipment, and the situation that the terminal equipment is blocked due to the fact that all the images to be marked are pushed to the terminal equipment at one time is avoided, so that the image marking efficiency is reduced or the image loss occurs is avoided.
Optionally, the step S105 may further include the following step B2:
step B2: and pushing the preset number of residual images to target terminal equipment after the labeling information of the pushed preset number of images to be pushed of each target account is obtained for each target account until the target account stops logging, or the residual images are absent.
In this embodiment, when the to-be-annotated image corresponding to each target account is pushed to the target terminal device, after the target account completes the image annotation of the remaining images with the preset number according to the annotation information of the preset number of to-be-pushed images of the target account, the preset number of remaining images are pushed to the terminal device logged in with the target account again until the target account terminates logging, or the number of remaining images is smaller than the number of preset images.
For example, if the number of images to be marked corresponding to a target account is 50, 10 images to be marked are pushed to the target account for the first time, after marking information of 2 pushed images of the target account is obtained, 2 remaining images are pushed to the target terminal device again, so that the number of the remaining images in the terminal device of the target account is ensured to be always kept at 10 until the target account stops logging or the number of the remaining images is less than 2, and at the moment, all the remaining images of the remaining number are pushed to the target terminal device.
In addition, for the residual images of different target account numbers, the number of the residual images pushed to the terminal equipment can be the same, and can also be set to different preset numbers according to the difficulty level of the corresponding target labeling objects of the different residual images, which is reasonable.
Based on the method, a certain number of images to be pushed can be pushed to the target terminal equipment, and the situation that the terminal equipment is blocked due to the fact that all the images to be marked are pushed to the terminal equipment at one time is avoided, so that the image marking efficiency is reduced or the image loss occurs is avoided.
The technical scheme provided by the embodiment of the application is applied, firstly, the working groups to which the multiple target accounts which are logged in parallel respectively belong are determined, then, for each target account, the labeling object corresponding to the working group to which the target account belongs is determined according to the first corresponding relation between the working groups and the labeling object, the labeling task of the target account is generated based on the determined labeling object, then, for each target account, the designated labeling object of the target account is determined, the image to be labeled of the target account is searched based on the designated labeling object, then, the image to be pushed of each target account is determined according to the image to be labeled of each target account, further, the image to be pushed of each target account is pushed to the target terminal equipment logged in the target account, so that the target terminal equipment displays the received image to be pushed, and finally, the labeling information of the image to be pushed of the target account is obtained.
Optionally, in a specific implementation manner, step S101 is described above: for each target account, determining the labeling object corresponding to the working group to which the target account belongs according to the first corresponding relation between the working group and the labeling object may include the following step C1:
Step C1: for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme;
the labeling scheme comprises the following steps: a first correspondence relationship between the work group and the labeling object, and a subordinate relationship between the account number and the work group.
Optionally, the labeling scheme further includes: and the second corresponding relation among the labeling objects, the candidate labeling contents and the labeling modes.
Based on this, in an optional specific implementation manner, step S104 described above: before pushing the image to be annotated corresponding to each target account to the target terminal device, the method further comprises step D1:
step D1: according to the second corresponding relation in the labeling scheme, determining candidate labeling contents and labeling modes of the designated labeling objects of each target account, wherein the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account;
the second correspondence is preset in the labeling scheme;
after the second corresponding relation is established, the staff can carry out image annotation on the corresponding target annotation object according to the determined candidate annotation content and the annotation mode.
After the labeling objects of the account numbers used by the staff are determined, candidate labeling contents and labeling modes of each labeling object can be determined for each labeling object based on the second corresponding relation.
Optionally, step S104 is as follows: pushing the image to be pushed of each target account to the target terminal device, so that the target terminal device displays the received image to be pushed, and the method may include the following step D2:
step D2: pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode;
the candidate labeling contents and labeling modes can be obtained by configuring proper candidate labeling contents and labeling modes according to the type and labeling requirements of the labeling objects after the labeling objects of the images to be labeled are determined by a preset labeling scheme, and adding the configured candidate labeling contents and labeling modes to the labeling objects of the images to be labeled.
Therefore, candidate labeling contents and labeling modes of the target labeling objects can be directly displayed in the images to be labeled of the target account, and the images to be labeled can be intuitively labeled, so that the efficiency of image labeling is improved; in addition, the candidate labeling content and labeling mode of the target labeling object are displayed in the image to be labeled of the target account, so that the follow-up reference and the integration of the labeling information are facilitated.
Optionally, step S105 is as follows: for each target account, obtaining the labeling information of the image to be pushed of the target account may include the following step D3:
step D3: aiming at each target account, acquiring the annotation information of the image to be pushed of the target account selected from candidate annotation contents of the target account according to the annotation mode of the target account;
after the target account obtains the corresponding image to be pushed, selecting the labeling information of the image to be pushed from the candidate labeling contents of the target account, and finishing labeling of the image to be pushed according to the labeling information.
Based on the method, the labeling of the image to be pushed is completed according to the labeling information selected in the candidate labeling content of the target account, and the situations of slow labeling speed and wrong labeling caused by undefined labeling information can be effectively avoided, so that the accuracy and efficiency of image labeling are improved.
Optionally, in a specific implementation manner, as shown in fig. 3, the method for labeling an image provided in the embodiment of the present application further includes the following step S301:
step S301: and adding new entries related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting the current entries related to the object relation between the working group and the labeling object, and returning to the step of determining a plurality of target account numbers which are logged in parallel.
After an image to be annotated is annotated by an annotation scheme, a new image annotation, i.e., a complementary annotation, may need to be performed on the image. The method comprises the steps that a new labeling object is required to be determined, and a first corresponding relation between a working group and the labeling object in a labeling scheme is required to be determined, so that after a worker logs in an account and determines the working group to which the account belongs, the new labeling object can be determined according to the new first corresponding relation, and the target image is subjected to the labeling.
Similarly, after all the images of some marked objects are marked, the marked objects can be deleted, so that the waste of resource space is avoided.
Therefore, when the newly added labeling object exists in the image to be labeled, the first corresponding relation can be updated in time, so that the image labeling of the newly added labeling object can be conveniently carried out subsequently.
Optionally, in a specific implementation manner, the method for labeling an image provided in the embodiment of the present application further includes the following step E1:
step E1: in the labeling scheme, new entries about the affiliation of the account and the workgroup are added, and/or current entries about the affiliation of the account and the workgroup are deleted, and the step of determining multiple target accounts logged in parallel is returned.
When the labeling objects are increased, new entries of the account numbers and the working groups can be added in the labeling scheme, so that after the new labeling objects are added, more staff can be added to use the account numbers, and therefore more efficient labeling of the labeling objects is completed. In addition, when some accounts are not used for image annotation work, the accounts can be deleted, so that the situation that the accounts incapable of executing annotation work are used for image annotation is avoided.
Optionally, in a specific implementation manner, as shown in fig. 4, the method for labeling an image provided in the embodiment of the present application further includes the following steps S401 to S402:
step S401: determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image with target integration information;
in order to perform pattern model training by using the sample image subjected to image annotation, the annotation information of the image subjected to information annotation needs to be integrated so as to facilitate subsequent sample training; the labeling information may include only the labeling information of the image that is primarily labeled, may include only the labeling information of the image that is secondarily labeled, and may include two labeling information of the primary label and the secondary label.
Step S402: and sending each target image carrying the target integration information to the designated equipment.
The designating device may be a terminal device that registers an account number for labeling the target image, or may be another terminal device or an electronic device designated according to the requirement.
In this specific implementation manner, the target labeling information of each target image with the image labeling completed may be integrated; or the target labeling information required by the part is selected to be integrated according to the actual requirement, so that each target image carrying the target integration information is obtained.
Based on the method, after each target image carrying the target integration information is obtained, the target images are sent to the appointed equipment, so that the target images can be conveniently checked according to the target integration information carried by the target images, further the target images needing to be modified for image labeling can be found in time, and the preset labeling scheme is updated, and therefore the accuracy and the efficiency of image labeling are improved.
Corresponding to the image labeling method provided in the embodiment of the present application, the embodiment of the present application further provides an image labeling device, as shown in fig. 5, where the device includes:
A work group determining module 501, configured to determine work groups to which a plurality of target account numbers logged in parallel respectively belong;
the task generating module 502 is configured to determine, for each target account, a labeling object corresponding to a work group to which the target account belongs according to a first correspondence between the work group and the labeling object, and generate a labeling task of the target account based on the determined labeling object, where the labeling task of each target account includes: labeling objects corresponding to the work groups to which the target account belongs;
an image retrieving module 503, configured to determine, for each target account, a specified labeling object of the target account, and retrieve an image to be labeled of the target account based on the specified labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
The image determining module 504 is configured to determine, according to the image to be annotated of each target account, an image to be pushed of each target account;
the image pushing module 505 is configured to push an image to be pushed of each target account to the target terminal device, so that the target terminal device displays the received image to be pushed;
the information obtaining module 506 is configured to obtain, for each target account, labeling information of an image to be pushed of the target account; the labeling information is sent by the target terminal equipment, and the labeling information of the specified labeling object in the image to be pushed of the target account is related to the labeling information.
The technical scheme provided by the embodiment of the application is applied, firstly, the working groups to which the multiple target accounts which are logged in parallel respectively belong are determined, then, for each target account, the labeling object corresponding to the working group to which the target account belongs is determined according to the first corresponding relation between the working group and the labeling object, the labeling task of the target account is generated based on the determined labeling object, then, for each target account, the designated labeling object of the target account is determined, the image to be labeled of the target account is searched based on the designated labeling object, then, the image to be pushed of each target account is determined according to the image to be labeled of each target account, further, the image to be pushed of each target account is pushed to the target terminal equipment logged with the target account, so that the terminal equipment displays the received image to be pushed, and finally, the labeling information of the image to be pushed of the target account is obtained for each target account.
Based on the method, for a plurality of objects to be marked, the objects to be marked are distributed to different working groups, so that a first corresponding relation between the working groups and the marked objects is established, a plurality of account numbers for image marking are distributed in the working groups, the working groups to which the account numbers belong are determined after the staff logs in the account numbers, the marked objects are determined according to the first corresponding relation, and marking of the marked objects is completed. Therefore, when labeling is carried out, different staff can log in the account in parallel, and accordingly corresponding images to be labeled can be labeled in parallel. Because a plurality of accounts are logged in parallel, a plurality of labeling objects of a plurality of images can be labeled simultaneously, the image labeling efficiency can be ensured, and because different labeling objects are distributed by accounts of different staff, the staff distributed by each labeling object does not need to pay attention to all the labeling objects, so that the staff is not easy to mistake when labeling the images, the image labeling accuracy can be improved under the condition of ensuring the image labeling efficiency, the image sample accuracy can be improved, and the subsequent image model training accuracy can be improved.
Optionally, in a specific implementation manner, the image determining module 504 is specifically configured to:
determining the image to be marked of each target account as an image to be pushed of the target account; or alternatively, the process may be performed,
and determining the sub-image of the target labeling object corresponding to each target account number in the image to be labeled of each target account number as an image to be pushed of the target account number.
Optionally, in a specific implementation manner, the image pushing module 505 is specifically configured to:
pushing a specified number of residual images to the terminal equipment aiming at each target account; wherein, the residual image is: an image to be pushed, which is not pushed to the target terminal equipment, of the target account;
the apparatus further comprises:
the residual image pushing module is used for triggering the image pushing module after the labeling information of all pushed images to be pushed of each target account is obtained, until the target account stops logging, or the residual images are not existed; or, for each target account, after the labeling information of the preset number of images to be pushed of the target account is obtained, pushing the preset number of residual images to the target terminal equipment until the target account terminates login, or the residual images are not existed.
Optionally, in a specific implementation manner, the task generating module 502 is specifically configured to:
for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme;
the labeling scheme comprises the following steps: a first correspondence relationship between the work group and the labeling object, and a subordinate relationship between the account number and the work group.
Optionally, in a specific implementation manner, the labeling scheme further includes: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode;
the apparatus further comprises:
the content mode determining module is used for determining candidate labeling contents and labeling modes of the designated labeling objects of each target account according to the second corresponding relation in the labeling scheme before pushing the images to be labeled corresponding to each target account to the target terminal equipment, and the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account;
the image pushing module 505 is specifically configured to:
pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode;
The information obtaining module 506 is specifically configured to:
and acquiring labeling information of the image to be pushed of the target account selected from candidate labeling contents of the target account according to the labeling mode of the target account aiming at each target account.
Optionally, in a specific implementation manner, the apparatus further includes:
and the item adding and deleting first module is used for adding new items related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting current items related to the object relation between the working group and the labeling object, and returning to the step of determining the multiple target account numbers which are logged in parallel.
Optionally, in a specific implementation manner, the apparatus further includes:
item adding and deleting second module: and the step of adding new entries about the affiliation of the account number and the working group in the labeling scheme, and/or deleting the current entries about the affiliation of the account number and the working group, and returning to the step of determining a plurality of target account numbers which are logged in parallel.
Optionally, in a specific implementation manner, the apparatus further includes:
the information integration module is used for determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image carrying target integration information;
And the image sending module is used for sending each target image carrying the target integration information to the designated equipment.
It should be noted that, in this embodiment, the images to be annotated are all from the public dataset.
The embodiment of the application also provides an electronic device, as shown in fig. 6, including:
a memory 601 for storing a computer program;
the processor 602 is configured to implement any one of the image labeling methods provided in the embodiments of the present application when executing the program stored in the memory 601.
Alternatively, the communication bus mentioned above for the electronic device may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, or the like. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.
The communication interface is used for communication between the electronic device and other devices.
The Memory may include random access Memory (Random Access Memory, RAM) or may include Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
And the electronic device may further comprise a communication bus and/or a communication interface, through which the processor 602, the communication interface, and the memory 601 communicate with each other.
In yet another embodiment provided herein, there is also provided a computer readable storage medium having stored therein a computer program which when executed by a processor implements the steps of any of the image annotation methods described above.
In yet another embodiment provided herein, there is also provided a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the image annotation methods of the above embodiments.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), etc.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the present application. Any modifications, equivalent substitutions, improvements, etc. that are within the spirit and principles of the present application are intended to be included within the scope of the present application.

Claims (12)

1. A method of image annotation, the method comprising:
determining a working group to which a plurality of target account numbers logged in parallel respectively belong;
for each target account, determining a labeling object corresponding to the work group to which the target account belongs according to a first corresponding relation between the work group and the labeling object, and generating a labeling task of the target account based on the determined labeling object; the labeling task of each target account comprises the following steps: labeling objects corresponding to the work groups to which the target account belongs;
determining a designated labeling object of each target account, and searching an image to be labeled of the target account based on the designated labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
Determining an image to be pushed of each target account according to the image to be marked of each target account;
pushing the image to be pushed of each target account to the target terminal equipment so that the target terminal equipment displays the received image to be pushed;
for each target account, acquiring labeling information of an image to be pushed of the target account; the annotation information is sent by the target terminal equipment, and the annotation information of the specified annotation object in the image to be pushed of the target account is related to the target account.
2. The method according to claim 1, wherein the determining the image to be pushed of each target account according to the image to be annotated of each target account includes:
determining the image to be marked of each target account as an image to be pushed of the target account; or alternatively, the process may be performed,
and determining the sub-image of the designated labeling object of each target account included in the images to be labeled of the target account as the images to be pushed of the target account.
3. The method according to claim 1, wherein pushing the image to be pushed of each target account to the target terminal device includes:
Pushing a specified number of residual images to the target terminal equipment aiming at each target account; wherein, the residual image is: an image to be pushed, which is not pushed to the target terminal equipment, of the target account;
the method further comprises the steps of:
returning to the step of pushing the specified number of residual images to the target terminal equipment after the labeling information of all pushed images to be pushed of the target account is obtained for each target account until the target account terminates login, or the residual images are not existed;
or alternatively, the process may be performed,
and pushing the preset number of residual images to target terminal equipment after the labeling information of the pushed preset number of images to be pushed of each target account is obtained, until the target account terminates login, or the residual images are absent.
4. The method according to claim 1, wherein for each target account, determining the labeling object corresponding to the work group to which the target account belongs according to the first correspondence between the work group and the labeling object comprises:
for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme;
The labeling scheme comprises the following steps: a first correspondence relationship between the work group and the labeling object, and a subordinate relationship between the account number and the work group.
5. The method of claim 4, wherein the labeling scheme further comprises: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode;
before pushing the image to be annotated corresponding to each target account to the target terminal device, the method further comprises:
according to the second corresponding relation in the labeling scheme, determining candidate labeling contents and labeling modes of the designated labeling objects of each target account, wherein the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account;
pushing the image to be pushed of each target account to the target terminal device, so that the target terminal device displays the received image to be pushed, including:
pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode;
the obtaining, for each target account, labeling information of an image to be pushed of the target account includes:
And acquiring labeling information of the image to be pushed of the target account selected from candidate labeling contents of the target account according to the labeling mode of the target account aiming at each target account.
6. The method according to claim 1, wherein the method further comprises:
and adding new entries related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting the current entries related to the object relation between the working group and the labeling object, and returning to the step of determining the multiple target account numbers which are logged in parallel.
7. The method according to claim 4, wherein the method further comprises:
in the labeling scheme, adding new entries related to the affiliation of the account and the working group, and/or deleting current entries related to the affiliation of the account and the working group, and returning to the step of determining a plurality of target accounts which are logged in parallel.
8. The method according to any one of claims 1-7, further comprising:
determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image with target integration information;
And sending each target image carrying the target integration information to the designated equipment.
9. An image annotation device, the device comprising:
the work group determining module is used for determining work groups to which a plurality of target account numbers logged in parallel respectively belong;
the task generating module is configured to determine, for each target account, a labeling object corresponding to a work group to which the target account belongs according to a first correspondence between the work group and the labeling object, and generate a labeling task of the target account based on the determined labeling object, where the labeling task of each target account includes: labeling objects corresponding to the work groups to which the target account belongs;
the image retrieval module is used for determining a specified labeling object of each target account and retrieving an image to be labeled of the target account based on the specified labeling object; wherein, the specified labeling object is: the received object selection instruction is an object indicated in the labeling task of the target account, and the image to be labeled is: the specified labeling object contained in the preset image is not labeled, and is not pushed out of the target terminal equipment, and the object indicated by the received object selection instruction is an image of the terminal equipment of the specified labeling object, wherein the target terminal equipment is: a terminal device logged in with the target account;
The image determining module is used for determining an image to be pushed of each target account according to the image to be annotated of each target account;
the image pushing module is used for pushing the image to be pushed of each target account to the target terminal equipment so that the target terminal equipment displays the received image to be pushed;
the information acquisition module is used for acquiring labeling information of the image to be pushed of each target account; the annotation information is sent by the target terminal equipment, and the annotation information of the specified annotation object in the image to be pushed of the target account is related to the target account.
10. The apparatus of claim 9, wherein the image determination module is specifically configured to:
determining the image to be marked of each target account as an image to be pushed of the target account; or alternatively, the process may be performed,
determining a sub-image of a target labeling object corresponding to each target account as a to-be-pushed image of the target account, wherein the to-be-labeled image of each target account comprises the sub-image of the target labeling object corresponding to the target account;
and/or the number of the groups of groups,
the image pushing module is specifically configured to:
pushing a specified number of residual images to the target terminal equipment aiming at each target account; wherein, the residual image is: an image to be pushed, which is not pushed to the target terminal equipment, of the target account;
The apparatus further comprises:
the residual image pushing module is used for triggering the image pushing module after the labeling information of all pushed images to be pushed of each target account is obtained, until the target account stops logging, or the residual images are not existed; or, for each target account, after marking information of a preset number of pushed images of the target account is obtained, pushing the preset number of residual images to the target terminal equipment until the target account terminates login, or the residual images do not exist;
and/or the number of the groups of groups,
the task generating module is specifically configured to:
for each target account, searching a working group to which the target account belongs in a pre-acquired labeling scheme, and searching a labeling object corresponding to the working group to which the target account belongs in the labeling scheme;
the labeling scheme comprises the following steps: a first corresponding relation between the work group and the labeling object, and a subordinate relation between the account number and the work group;
and/or the number of the groups of groups,
the labeling scheme further comprises: a second corresponding relation of the labeling object, the candidate labeling content and the labeling mode;
The apparatus further comprises:
the content mode determining module is used for determining candidate labeling contents and labeling modes of the designated labeling objects of each target account according to the second corresponding relation in the labeling scheme before pushing the images to be labeled corresponding to each target account to the target terminal equipment, and the candidate labeling contents and the labeling modes are used as the candidate labeling contents and the labeling modes of the target account;
the image pushing module is specifically configured to:
pushing the candidate labeling content and the labeling mode of each target account to target terminal equipment, wherein the candidate labeling content and the labeling mode of each target account are carried in the candidate labeling content and the labeling mode of each target account, so that the target terminal equipment displays the received candidate labeling content and the received candidate labeling mode;
the information acquisition module is specifically configured to:
aiming at each target account, acquiring the annotation information of the image to be pushed of the target account selected from candidate annotation contents of the target account according to the annotation mode of the target account;
and/or the number of the groups of groups,
the apparatus further comprises:
the first module is used for adding new items related to the corresponding relation between the working group and the labeling object in the first corresponding relation, and/or deleting the current items related to the object relation between the working group and the labeling object, and returning to the step of determining a plurality of target accounts which are logged in parallel;
And/or the number of the groups of groups,
the apparatus further comprises:
a second module for adding and deleting items, which is used for adding new items related to the subordination relation between the account and the working group and/or deleting the current items related to the subordination relation between the account and the working group in the labeling scheme, and returning to the step of determining a plurality of target accounts which are logged in parallel;
and/or the number of the groups of groups,
the apparatus further comprises:
the information integration module is used for determining target labeling information of each target image with information labeling in the preset image, and integrating the target labeling information of each target image to obtain each target image carrying target integration information;
and the image sending module is used for sending each target image carrying the target integration information to the designated equipment.
11. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the method of any of claims 1-8 when executing a program stored on a memory.
12. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-8.
CN202310136172.0A 2023-02-07 2023-02-07 Image labeling method and device, electronic equipment and storage medium Pending CN116340562A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310136172.0A CN116340562A (en) 2023-02-07 2023-02-07 Image labeling method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310136172.0A CN116340562A (en) 2023-02-07 2023-02-07 Image labeling method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN116340562A true CN116340562A (en) 2023-06-27

Family

ID=86878199

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310136172.0A Pending CN116340562A (en) 2023-02-07 2023-02-07 Image labeling method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN116340562A (en)

Similar Documents

Publication Publication Date Title
CN108984388B (en) Method and terminal equipment for generating automatic test case
WO2019019769A1 (en) Service function implementation method, device, computer apparatus and storage medium
CN110287696B (en) Detection method, device and equipment for rebound shell process
US11176169B2 (en) Recommending visual and execution templates to enable automation of control and data exploration across systems
US20150089415A1 (en) Method of processing big data, apparatus performing the same and storage media storing the same
CN113467868B (en) Method and device for creating equipment resources
CN110659206A (en) Simulation architecture establishing method, device, medium and electronic equipment based on microservice
US20200205001A1 (en) Electronic apparatus and method of executing application program
CN113204345A (en) Page generation method and device, electronic equipment and storage medium
CN111061740A (en) Data synchronization method, equipment and storage medium
CN111309624B (en) Test method, test device, test equipment and storage medium
CN112000911A (en) Page management method and device, electronic equipment and storage medium
CN111522626B (en) Virtual machine list generation method and device and electronic equipment
CN111488386B (en) Data query method and device
CN108228611B (en) Document information copying method and device
CN109189688B (en) Test case script generation method and device and electronic equipment
CN106302011A (en) Method of testings based on multiterminal and terminal
CN116340562A (en) Image labeling method and device, electronic equipment and storage medium
CN113010208B (en) Version information generation method, device, equipment and storage medium
JP2010072876A (en) Rule creation program, rule creation method, and rule creation device
CN111124579B (en) Special effect rendering method and device, electronic equipment and storage medium
US9275358B1 (en) System, method, and computer program for automatically creating and submitting defect information associated with defects identified during a software development lifecycle to a defect tracking system
US8799318B2 (en) Function module leveraging fuzzy search capability
CN117573561B (en) Automatic test system, method, electronic equipment and storage medium
CN112445790B (en) Report data storage method, device, equipment and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination