CN108805959A - A kind of image labeling method and system - Google Patents
A kind of image labeling method and system Download PDFInfo
- Publication number
- CN108805959A CN108805959A CN201810392754.4A CN201810392754A CN108805959A CN 108805959 A CN108805959 A CN 108805959A CN 201810392754 A CN201810392754 A CN 201810392754A CN 108805959 A CN108805959 A CN 108805959A
- Authority
- CN
- China
- Prior art keywords
- image
- mark
- tab area
- characteristic point
- rectangle frame
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
The invention discloses a kind of image labeling method and system, which includes：Obtain the notation methods that user chooses；The image that user selectes is opened, and according to notation methods, image is labeled, to obtain including the mark image of tab area；In the case of the mark inaccuracy of tab area, to marking image into edlin.The present invention can carry out secondary editor, to keep mark more accurate by a kind of disclosed image labeling method and system to the data marked.
Description
Technical field
The present invention relates to technical field of image processing, it particularly relates to a kind of image labeling method and system.
Background technology
With flourishing for AI (Artificial Intelligence, artificial intelligence) field, intelligent algorithm pair
The quality requirement of sample data is also higher and higher, and the annotation tool that previous mark person uses when being labeled to sample is simultaneously endless
Kind, making sample data accuracy rate, there are deviations.
For the problems in the relevant technologies, currently no effective solution has been proposed.
Invention content
For the problems in the relevant technologies, a kind of image labeling method of present invention proposition and system, which raises sample numbers
According to accuracy.
The technical proposal of the invention is realized in this way：
According to an aspect of the invention, there is provided a kind of image labeling method.
The image labeling method includes：Obtain the notation methods that user chooses；The image that user selectes is opened, and according to mark
Note mode, is labeled image, to obtain including the mark image of tab area；In the inaccurate feelings of the mark of tab area
Under condition, to marking image into edlin.
According to one embodiment of present invention, notation methods include：Characteristic point mark, rectangle frame mark.
According to one embodiment of present invention, image labeling method further comprises：Obtain the characteristic point mark that user chooses
The notation methods of note；The image that user selectes is opened, and is marked according to characteristic point, image is labeled, to obtain comprising mark
Note the mark image in region；In the case of the mark inaccuracy of tab area, the characteristic point in tab area is pulled
Modification, with to marking image into edlin.
According to one embodiment of present invention, the notation methods marked by characteristic point, are labeled face.
According to one embodiment of present invention, image labeling method further comprises：Obtain the rectangle collimation mark that user chooses
The notation methods of note；The image that user selectes is opened, and is marked according to rectangle frame, image is labeled, to obtain comprising mark
Note the mark image in region；In the case of the mark inaccuracy of tab area, the rectangle frame in tab area is pulled
Modification, with to marking image into edlin.
According to one embodiment of present invention, by the notation methods of rectangle frame, vehicle is labeled.
According to another aspect of the present invention, a kind of image labeling system is provided.
The image labeling system includes：Acquisition module, the notation methods for obtaining user's selection；Labeling module is opened,
It is used to open the image that user selectes, and according to notation methods, image is labeled, to obtain including the mark of tab area
Image；Editor module is used in the case of the mark inaccuracy of tab area, to marking image into edlin.
According to one embodiment of present invention, notation methods include：Characteristic point mark, rectangle frame mark.
According to one embodiment of present invention, image labeling system further comprises：First acquisition submodule, for obtaining
The notation methods for the characteristic point mark that user chooses；First opens mark submodule, is used to open the image that user selectes, and root
It is marked according to characteristic point, image is labeled, to obtain including the mark image of tab area；First editor's submodule, is used for
In the case of the mark inaccuracy of tab area, dragging modification is carried out to the characteristic point in tab area, with to marking image
Into edlin.
According to one embodiment of present invention, image labeling system further comprises：Second acquisition submodule, for obtaining
The notation methods for the rectangle frame mark that user chooses；Second opens mark submodule, is used to open the image that user selectes, and root
It is marked according to rectangle frame, image is labeled, to obtain including the mark image of tab area；Second editor's submodule, is used for
In the case of the mark inaccuracy of tab area, dragging modification is carried out to the rectangle frame in tab area, with to marking image
Into edlin.
The beneficial technical effect of the present invention lies in：
1, image labeling method through the invention and system can carry out secondary editor, such as square to the data marked
Shape frame can carry out drag operation etc. again after having deviation to rectangle frame, to keep mark more accurate；
2, image labeling system features in convenient of the invention, menu is very clear, can meet the mark person's of different levels
Requirement is conducive to improve working efficiency faster.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of image labeling method according to the ... of the embodiment of the present invention；
Fig. 2 is the block diagram of image labeling system according to the ... of the embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained belong to what the present invention protected
Range.
According to an embodiment of the invention, a kind of image labeling method is provided.
As shown in Figure 1, image labeling method according to the ... of the embodiment of the present invention includes：Step S101 obtains what user chose
Notation methods；Step S103 opens the image that user selectes, and according to notation methods, is labeled to image, to be included
The mark image of tab area；Step S105 compiles mark image in the case of the mark inaccuracy of tab area
Volume.
By means of the above-mentioned technical proposal of the present invention, secondary editor can be carried out to the data marked, as rectangle frame has
Drag operation etc. again can be carried out to rectangle frame, after deviation to keep mark more accurate.
According to one embodiment of present invention, notation methods include：Characteristic point mark, rectangle frame mark.
In this embodiment, the present invention has been provided by two kinds of notation methods of characteristic point mark and rectangle frame mark
Kind annotation tool can carry out secondary volume by any of the above-described kind of notation methods after mark is inaccurate to the data of mark
Collect equal operation.In addition, the mark to face can be achieved in characteristic point mark, for example, according to one embodiment of present invention, can lead to
The mode for crossing characteristic point mark, points out the face (binocular, nose or so corners of the mouth) of face, meanwhile, in the labeling position of characteristic point
In the case of inaccuracy, also characteristic point can be carried out to pull mobile modification；The mark to vehicle, example can be achieved in rectangle frame mark
Such as, according to one embodiment of present invention, it is inaccurate in the mark of rectangle frame after first passage rectangle frame is labeled vehicle
In the case of, it mouse can be used to choose drag operation to keep mark more accurate rectangle frame, to be marked based on both the above
Mode and the operation updated, greatly improve the accuracy of sample data.
In addition, it will be understood by those of skill in the art that although it illustrates characteristic point mark and square in the present embodiment
The specifically used mode of shape collimation mark note, but those skilled in the art can set it according to actual demand, for example, according to
One embodiment of the present of invention can also be labeled face by way of rectangle frame mark, and the present invention does not limit this
It is fixed.
In addition, those skilled in the art can also choose other notation methods and be labeled according to actual demand, this hair
It is bright that this is not limited.
According to one embodiment of present invention, image labeling method further comprises：Obtain the characteristic point mark that user chooses
The notation methods of note；The image that user selectes is opened, and is marked according to characteristic point, image is labeled, to obtain comprising mark
Note the mark image in region；In the case of the mark inaccuracy of tab area, the characteristic point in tab area is pulled
Modification, with to marking image into edlin.
In this embodiment, first, user's selected characteristic point from the pre-stored a variety of notation methods of system marks
Notation methods, then, user open the file where target image, and target image is made to be shown in first window, to
Target image in window is labeled, then, after frame selects the rough region of face, the region (or target area) of frame choosing
It can be shown enlarged in second window on the right side of first window, the mark of the characteristic point of face is then carried out with mouse, i.e., it is logical
The notation methods for crossing characteristic point mark, are labeled target image, then in the situation of the labeling position of characteristic point inaccuracy
Under, also characteristic point can be carried out to pull mobile modification, and annotation results are shown.
According to one embodiment of present invention, image labeling method further comprises：Obtain the rectangle collimation mark that user chooses
The notation methods of note；The image that user selectes is opened, and is marked according to rectangle frame, image is labeled, to obtain comprising mark
Note the mark image in region；In the case of the mark inaccuracy of tab area, the rectangle frame in tab area is pulled
Modification, with to marking image into edlin.
In this embodiment, first, user chooses rectangle frame mark from the pre-stored a variety of notation methods of system
Notation methods, then, user open the file where target image, and target image is made to be shown in first window, to
Target image in window is labeled, then, after selecting the rough region of face with rectangle circle, frame choosing region (or
Target area) first window can be shown enlarged in right side the second window in, subsequent feelings inaccurate in the mark of rectangle frame
Under condition, mouse can be used to choose drag operation to keep mark more accurate rectangle frame, and annotation results are shown.
According to an embodiment of the invention, a kind of image labeling system is additionally provided.
As shown in Fig. 2, image labeling system according to the ... of the embodiment of the present invention includes：Acquisition module 21, for obtaining user
The notation methods of selection；Open labeling module 22, be used to open the image that user selectes, and according to notation methods, to image into
Rower is noted, to obtain including the mark image of tab area；Editor module 23, the feelings inaccurate for the mark in tab area
Under condition, to marking image into edlin.
According to one embodiment of present invention, notation methods include：Characteristic point mark, rectangle frame mark.
According to one embodiment of present invention, image labeling system further comprises：First acquisition submodule (not shown),
Notation methods for the characteristic point mark for obtaining user's selection；First opens mark submodule (not shown), is used to open use
The selected image in family, and marked according to characteristic point, image is labeled, to obtain including the mark image of tab area；The
One editor's submodule (not shown), is used in the case of the mark inaccuracy of tab area, to the characteristic point in tab area
Dragging modification is carried out, with to marking image into edlin.
According to one embodiment of present invention, image labeling system further comprises：Second acquisition submodule (not shown),
Notation methods for the rectangle frame mark for obtaining user's selection；Second opens mark submodule (not shown), is used to open use
The selected image in family, and marked according to rectangle frame, image is labeled, to obtain including the mark image of tab area；The
Two editor's submodule (not shown), are used in the case of the mark inaccuracy of tab area, to the rectangle frame in tab area
Dragging modification is carried out, with to marking image into edlin.
In conclusion by means of the above-mentioned technical proposal of the present invention, image labeling method through the invention and system,
The data that have marked can be carried out with secondary editor, such as rectangle frame have deviation after can carry out drag operation again to rectangle frame,
To keep mark more accurate.In addition, the image labeling system features in convenient of the present invention, menu is very clear, can meet difference
The requirement of the mark person of level is conducive to improve working efficiency faster.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention god.
Claims (10)
1. a kind of image labeling method, which is characterized in that including：
Obtain the notation methods that user chooses；
The image that the user selectes is opened, and according to the notation methods, described image is labeled, to obtain comprising mark
Note the mark image in region；
In the case of the mark inaccuracy of the tab area, to the mark image into edlin.
2. image labeling method according to claim 1, which is characterized in that the notation methods include：Characteristic point mark,
Rectangle frame marks.
3. image labeling method according to claim 2, which is characterized in that described image mask method further comprises：
Obtain the notation methods for the characteristic point mark that the user chooses；
The image that the user selectes is opened, and is marked according to the characteristic point, described image is labeled, to be included
The mark image of the tab area；
In the case of the mark inaccuracy of the tab area, dragging modification is carried out to the characteristic point in the tab area,
With to the mark image into edlin.
4. image labeling method according to claim 3, which is characterized in that the mark side marked by the characteristic point
Formula is labeled face.
5. image labeling method according to claim 2, which is characterized in that described image mask method further comprises：
Obtain the notation methods for the rectangle frame mark that the user chooses；
The image that the user selectes is opened, and is marked according to the rectangle frame, described image is labeled, to be included
The mark image of the tab area；
In the case of the mark inaccuracy of the tab area, dragging modification is carried out to the rectangle frame in the tab area,
With to the mark image into edlin.
6. image labeling method according to claim 5, which is characterized in that right by the notation methods of the rectangle frame
Vehicle is labeled.
7. a kind of image labeling system, which is characterized in that including：
Acquisition module, the notation methods for obtaining user's selection；
Labeling module is opened, is used to open the image that the user selectes, and according to the notation methods, carried out to described image
Mark, to obtain including the mark image of tab area；
Editor module, in the case of the mark inaccuracy of the tab area, to the mark image into edlin.
8. image labeling system according to claim 7, which is characterized in that the notation methods include：Characteristic point mark,
Rectangle frame marks.
9. image labeling system according to claim 8, which is characterized in that described image labeling system further comprises：
First acquisition submodule, the notation methods for obtaining the characteristic point mark that the user chooses；
First opens mark submodule, the image that the user selectes is used to open, and mark according to the characteristic point, to described
Image is labeled, to obtain including the mark image of the tab area；
First editor's submodule, is used in the case of the mark inaccuracy of the tab area, in the tab area
Characteristic point carries out dragging modification, with to the mark image into edlin.
10. image labeling system according to claim 8, which is characterized in that described image labeling system further comprises：
Second acquisition submodule, the notation methods for obtaining the rectangle frame mark that the user chooses；
Second opens mark submodule, the image that the user selectes is used to open, and mark according to the rectangle frame, to described
Image is labeled, to obtain including the mark image of the tab area；
Second editor's submodule, is used in the case of the mark inaccuracy of the tab area, in the tab area
Rectangle frame carries out dragging modification, with to the mark image into edlin.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810392754.4A CN108805959A (en) | 2018-04-27 | 2018-04-27 | A kind of image labeling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810392754.4A CN108805959A (en) | 2018-04-27 | 2018-04-27 | A kind of image labeling method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108805959A true CN108805959A (en) | 2018-11-13 |
Family
ID=64093972
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810392754.4A Withdrawn CN108805959A (en) | 2018-04-27 | 2018-04-27 | A kind of image labeling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108805959A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784381A (en) * | 2018-12-27 | 2019-05-21 | 广州华多网络科技有限公司 | Markup information processing method, device and electronic equipment |
| CN109840832A (en) * | 2019-01-30 | 2019-06-04 | 准动网络科技(厦门)有限公司 | Commodity image mask method, device, electronic equipment and system |
| CN110598705A (en) * | 2019-09-27 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Semantic annotation method and device for image |
| CN111260613A (en) * | 2020-01-10 | 2020-06-09 | 丽水正阳电力建设有限公司 | Image labeling method based on machine learning model |
| CN111583209A (en) * | 2020-04-29 | 2020-08-25 | 上海杏脉信息科技有限公司 | Brain perfusion image feature point selection method, medium and electronic equipment |
| CN113052369A (en) * | 2021-03-15 | 2021-06-29 | 北京农业智能装备技术研究中心 | Intelligent agricultural machinery operation management method and system |
| CN113435447A (en) * | 2021-07-26 | 2021-09-24 | 杭州海康威视数字技术股份有限公司 | Image annotation method, device and system |
| WO2021217543A1 (en) * | 2020-04-30 | 2021-11-04 | 华为技术有限公司 | Image annotation method, apparatus, device and medium |
-
2018
- 2018-04-27 CN CN201810392754.4A patent/CN108805959A/en not_active Withdrawn
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784381A (en) * | 2018-12-27 | 2019-05-21 | 广州华多网络科技有限公司 | Markup information processing method, device and electronic equipment |
| CN109840832A (en) * | 2019-01-30 | 2019-06-04 | 准动网络科技(厦门)有限公司 | Commodity image mask method, device, electronic equipment and system |
| CN110598705A (en) * | 2019-09-27 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Semantic annotation method and device for image |
| CN111260613A (en) * | 2020-01-10 | 2020-06-09 | 丽水正阳电力建设有限公司 | Image labeling method based on machine learning model |
| CN111583209A (en) * | 2020-04-29 | 2020-08-25 | 上海杏脉信息科技有限公司 | Brain perfusion image feature point selection method, medium and electronic equipment |
| CN111583209B (en) * | 2020-04-29 | 2021-11-02 | 上海杏脉信息科技有限公司 | Brain perfusion image feature point selection method, medium and electronic equipment |
| WO2021217543A1 (en) * | 2020-04-30 | 2021-11-04 | 华为技术有限公司 | Image annotation method, apparatus, device and medium |
| CN113052369A (en) * | 2021-03-15 | 2021-06-29 | 北京农业智能装备技术研究中心 | Intelligent agricultural machinery operation management method and system |
| CN113052369B (en) * | 2021-03-15 | 2024-05-10 | 北京农业智能装备技术研究中心 | Intelligent agricultural machinery operation management method and system |
| CN113435447A (en) * | 2021-07-26 | 2021-09-24 | 杭州海康威视数字技术股份有限公司 | Image annotation method, device and system |
| CN113435447B (en) * | 2021-07-26 | 2023-08-04 | 杭州海康威视数字技术股份有限公司 | Image labeling method, device and image labeling system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108805959A (en) | A kind of image labeling method and system | |
| CN101601072B (en) | Coherent image selection and modification | |
| CN110472467A (en) | The detection method for transport hub critical object based on YOLO v3 | |
| CN109145713A (en) | A kind of Small object semantic segmentation method of combining target detection | |
| JP4918383B2 (en) | Outline detecting apparatus, outline detecting method and program thereof | |
| CN110766694B (en) | Interactive segmentation method of three-dimensional medical image | |
| CN113240691A (en) | Medical image segmentation method based on U-shaped network | |
| JP2007111531A (en) | Method for improving image consistency, signal, computer readable medium and system | |
| CN104463843B (en) | Interactive image segmentation method of Android system | |
| Bartesaghi et al. | An energy-based three-dimensional segmentation approach for the quantitative interpretation of electron tomograms | |
| CN104504745A (en) | Identification photo generation method based on image segmentation and image matting | |
| CN109377509A (en) | Method, apparatus, storage medium and the equipment of image, semantic segmentation mark | |
| CN108564528A (en) | A kind of portrait photo automatic background weakening method based on conspicuousness detection | |
| CN110503705B (en) | Image labeling method and device | |
| CN109800788A (en) | A method of reducing image segmentation data set mark | |
| CN110826555A (en) | Man-machine cooperative image target detection data semi-automatic labeling method | |
| CN106651797A (en) | Determination method and apparatus for effective region of signal lamp | |
| CN107036956A (en) | Nucleus statistical method and device | |
| CN100383811C (en) | Method for extracting tongue body from tongue images | |
| Tang et al. | Leaf extraction from complicated background | |
| CN113313700B (en) | X-ray image interactive segmentation method based on deep learning | |
| CN109241893A (en) | Road selection method, device and readable storage medium storing program for executing based on artificial intelligence technology | |
| CN103985113A (en) | Tongue image segmentation method | |
| CN115796306B (en) | Training of permanent tooth maturity grading model and permanent tooth maturity grading method | |
| CN113361530A (en) | Image semantic accurate segmentation and optimization method using interaction means |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181113 |
|
| WW01 | Invention patent application withdrawn after publication |