CN108765428A - A kind of target object extracting method based on click interaction - Google Patents
A kind of target object extracting method based on click interaction Download PDFInfo
- Publication number
- CN108765428A CN108765428A CN201810478354.5A CN201810478354A CN108765428A CN 108765428 A CN108765428 A CN 108765428A CN 201810478354 A CN201810478354 A CN 201810478354A CN 108765428 A CN108765428 A CN 108765428A
- Authority
- CN
- China
- Prior art keywords
- image
- user
- target object
- click
- interaction
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
- G06F3/04845—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range for image manipulation, e.g. dragging, rotation, expansion or change of colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of based on the target object extracting method for clicking interaction, referred to as clicks image cutting (Click-cut).In order to simplify the interbehavior in Interactive Segmentation so that interactive image segmentation is simpler automatic, and the invention discloses a kind of new interactive image segmentation methods.In order to select objective area in image, user need to only input a bit in desired zone, realizing user, some interaction can divide the Click-cut methods of image foreground part, and this method divides the image into the foreground area interacted comprising user and remaining background area.This method substantially can be divided into two parts, and image division first there emerged a several superzone domains automatically, user's desired zone is determined further according to user's interaction point using area selection strategy.The present invention has easy to operate, has excellent performance, and the target object of extraction is clearly outlined, can meet it is following to various what's news the needs of the advantages that, can be applied to various occasions.
Description
Technical field
The present invention relates to the method for carrying out area-of-interest (target object) and extracting is interacted in a kind of image by user, more
Relate particularly to a kind of image segmentation identification that user is extracted by foreground area in one click behavior progress image of input
Method belongs to technical field of image processing.
Background technology
Image segmentation is the technology and process for dividing the image into the region of each tool characteristic and extracting interesting target, it is
By the committed step of image procossing to image analysis.The research of image Segmentation Technology has the history of decades, but people so far
General method, which can not be found, can be suitable for all types of images.Automatic cutting techniques are most common modes, automatically
Segmentation is exactly to develop a kind of method image segmentation is allowed to go out various profiles automatically, does not need user and intervenes.In recent years, it was also proposed that
Many automatic segmentation algorithms, however the segmentation result that automatic cutting techniques obtain is not correct enough, user cannot according to oneself
Wish extracts target area, since performance limitation cannot be satisfied user demand.
In order to solve the problems, such as this, there is the interactive image segmentation method for thering is user to interfere, this mode avoids machine
Device cannot be divided automatically well, and interactive image segmentation incorporates user's interactive action and arrived among image segmentation, with one kind
There is the dividing method of supervision, foreground area significant in image can be efficiently extracted.Typical interactive means include with one
Paintbrush respectively draws strokes at foreground and background and draws box etc. around foreground.But these interactive approaches are required for
Different foreground and background paintbrushes is inputted, user is needed to estimate the distribution in foreground and background region well.
Since existing interactive segmentation method is required for complicated paintbrush interactive information, such as patent No.
" 201310548279.2 " a kind of interactive image segmentation method etc., or need to input the choosing of rectangle circle, such as
" 201310587120.1 " a kind of interactive image segmentation method etc..All these technologies using very inconvenient, how
It is further simplified customer interaction information, can obtain wanting the target area of extraction as user only needs input a bit, at present also
There is no such method to come out.
The present invention proposes a kind of new interactive segmentation method, and in order to select objective area in image, user only need to be
Desired zone inputs a bit, and image can be divided by realizing a click interaction, and the present invention is defined as the side Click-cut
Method.
Invention content
In order to solve the problems, such as that interactive information is more complex in existing interactive segmentation method, it is a primary object of the present invention to
A kind of very simple interactive image segmentation method is provided, this method only needs user to be inputted a bit in required selection region,
The foreground area comprising customer interaction information and remaining background area can be divided the image into.
In order to achieve the above objectives, the technical side of a kind of interactive image segmentation method based on user's interaction point of the invention
Case is：Assuming that a given secondary image to be processed, first divides image, then interacts carry out foreground according to user automatically
It extracts, realizes and a little carry out image interactive segmentation (Click-cut).
The technology of the present invention realization includes 2 steps, first carries out image initial and divides automatically：It, can be first right first according to a sub-picture
It is pre-processed, and is such as carried out super-pixel segmentation, how to be combined the color characteristic and textural characteristics of image, is extracted respectively each super
Color characteristic and textural characteristics of pixel etc. build new super-pixel point similarity matrix, then use a kind of automatic society
Group's division methods, divide an image into several community areas, and after the automatic segmentation for completing image, inventive method is completed to this
The first step.Next the regional choice strategy using invention is needed：User only needs thinking selected region input a bit, area
Domain selection strategy merges near zone where user's interaction point, exports user's desired zone, divides the image into foreground
And background parts, it has been finally reached and has only just obtained foreground or the purpose of target object with any this interbehavior.
The beneficial effects of the invention are as follows：
Interbehavior in user's Interactive Segmentation is reduced to simplest behavior by the present invention, i.e. user only needs to input
A bit, can abstract image well target area, and target object profile information well can be obtained.Based on the present invention's
Method can develop Click-cut image interaction process softwares, or integrate Click-cut's in conventional images processing software
Image interactive process function.
Description of the drawings
Fig. 1 is total frame diagram of the present invention.
Fig. 2 is the Image Automatic Segmentation partial function figure of the present invention.
Specific implementation mode
The present invention is further described in detail below in conjunction with the drawings and specific embodiments, but is not used to the present invention
Restriction.
Such as Fig. 1 and Fig. 2, specific implementation of the invention：In Fig. 1, it gives original image (1) and is used as Interactive Segmentation algorithm
Input, which can be the direct picture pick-up device frame frame image that collects static images or extracted from video file,
Then the automatic segmentation of the image is carried out, picture format is also unlimited, can be the formats (2) such as jpg or bmp.One sub-picture can be pressed
The plain self-similarity that takes pictures is divided into the seldom region of number, and a general sub-picture is finally divided into more than ten of region (3).User
It can think that the target area input of extraction a bit, or by being divided on the image in several regions automatically inputs in original image
A bit (4).Then according to regional choice strategy, the region such as B1 where some positions of user has been obtained, all neighbours of B1 are found
Region T is occupied, if some neighbours region Tj and the region of the similarity minimum of all adjacent areas are B1, illustrates Tj and B1 most
Similar, then the regions Tj are also added in foreground area in B.This strategy is that recurrence executes, until the region in set B is all searched
Rope is excessively primary.It finally obtains the region in B and is listed in foreground area, the target object that need to as extract, remaining region is arranged
For background area (6).
In the embodiment depicted in figure 2, some ready-made technologies, such as super-pixel can be used in original image pre-segmentation, this step
Divide (1).The pixel for including in one image is very more, and network division is directly carried out using pixel as node of graph to be needed
Want longer processing time.Therefore, over-segmentation, such as superpixel segmentation method are carried out to image using initial segmentation method.By
After initial segmentation, many super-pixel zonules of a figure are obtained, each region can be taken as a node of network, be
The automatic segmentation that super-pixel region is carried out using the method that figure is cut, is needed the connection relation for building each region, needs to use
Certain characteristics describe each region, such as color characteristic, side, textural characteristics, and geometric properties, wherein color characteristic and texture are special
Sign energy validity indicates the characteristic (2) of target.The invention comprehensively utilizes both features, and it is every to form image by combination
The relevant connection relationship in a super-pixel region is to get to a unified similarity matrix (3).Will include thousands of super-pixel into
Row reprocessing, that is, the network comprising thousands of a nodes is divided, it divides an image into only comprising tens or several
A region.The number divided due to not knowing needs, it is therefore desirable to by a kind of Automated Partition Method of figure, such as division side of corporations
Method (4).After carrying out corporations' division, a sub-picture is divided into region more less than super-pixel region quantity, due to only
It has carried out a corporations to divide, may include still at least tens superzones domain, repeat network partition process (5), work as area
When domain number no longer changes or has reached ideal division region, automatic division method (6) is terminated.
The series of detailed descriptions listed above only for the present invention feasible embodiment specifically
Bright, they are all without departing from equivalent implementations made by technical spirit of the present invention not to limit the scope of the invention
Or change should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of based on the target object extracting method for clicking interaction, which is characterized in that be made of two parts, first automatically by image
Division there emerged a several regions, and the number in region is much smaller than number of pixels, and interbehavior is clicked further according to one input by user,
It determines the coordinate position clicked and is partitioned into the target object needed for user, i.e. foreground part.
2. according to claim 1 a kind of based on the target object extracting method for clicking interaction, which is characterized in that specific mistake
Journey：It according to a given sub-picture, is first pre-processed, pre-segmentation and feature extraction, according to the feature construction of extraction at similar
Matrix N * N are spent, the similarity matrix of N*N is divided into K nonoverlapping image-regions, wherein K automatically using Automated Partition Method
Much smaller than N；User inputs a click interbehavior, by the location of pixels of user's interaction point, according to the region in this method
Selection strategy merges near zone where user's interaction point, and the targeted object region extracted needed for output user will scheme
As being divided into foreground and background part.
3. according to claim 1 a kind of based on the target object extracting method for clicking interaction, which is characterized in that described to carry
It is Click-cut to take method, i.e. image a bit (Click), i.e., is directly divided into target prospect and background two by user's input
Divide (cut).
4. according to claim 1 a kind of based on the target object extracting method for clicking interaction, which is characterized in that if figure
It, can be simultaneously every when user thinks to extract these target objects simultaneously as in when target object disjunct there are several independences
Input a little carries out Click-cut on a target object, while obtaining extracting these target objects.
5. according to claim 2 a kind of based on the target object extracting method for clicking interaction, which is characterized in that described
Image segmentation is not limited to certain picture format, and any still image is all suitable for Click-cut.
6. according to claim 2 a kind of based on the target object extracting method for clicking interaction, which is characterized in that the spy
Sign extraction includes color characteristic and textural characteristics.
7. it is according to claim 2 it is a kind of based on click interaction target object extracting method, which is characterized in that it is described from
Dynamic division methods are automatic group dividing method.
8. a kind of based on the target object extracting method for clicking interaction according to claim 1-7 any one of them, feature exists
In target object extracting method can be used in developing at Click-cut image interactions in the image clicked based on user
The software of type is managed, or integrates the image interactive process function of Click-cut in conventional images processing software.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2017110682674 | 2017-10-25 | ||
| CN201711068267 | 2017-10-25 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108765428A true CN108765428A (en) | 2018-11-06 |
Family
ID=64007256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810478354.5A Pending CN108765428A (en) | 2017-10-25 | 2018-05-18 | A kind of target object extracting method based on click interaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108765428A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783732A (en) * | 2019-01-10 | 2019-05-21 | 珠海金山网络游戏科技有限公司 | A kind of micro- end construction method of game based on click density |
| CN115293996A (en) * | 2022-10-08 | 2022-11-04 | 杭州群核信息技术有限公司 | Image toning method, device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329763A (en) * | 2007-06-22 | 2008-12-24 | 西门子公司 | Method for segmenting structures in image data records and image processing unit |
| CN103116630A (en) * | 2013-02-03 | 2013-05-22 | 苏州市职业大学 | Color image search method based on scale invariant feature transform (sift) seed region growing |
| CN103914822A (en) * | 2012-12-31 | 2014-07-09 | 清华大学 | Interactive video foreground object extraction method based on super pixel segmentation |
| CN105701810A (en) * | 2016-01-12 | 2016-06-22 | 湖南中航天目测控技术有限公司 | Unmanned aerial vehicle aerial image electronic delineating method based on click type image segmentation |
| CN105809673A (en) * | 2016-03-03 | 2016-07-27 | 上海大学 | SURF (Speeded-Up Robust Features) algorithm and maximal similarity region merging based video foreground segmentation method |
| CN106127791A (en) * | 2016-07-06 | 2016-11-16 | 武汉大学 | A kind of contour of building line drawing method of aviation remote sensing image |
-
2018
- 2018-05-18 CN CN201810478354.5A patent/CN108765428A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329763A (en) * | 2007-06-22 | 2008-12-24 | 西门子公司 | Method for segmenting structures in image data records and image processing unit |
| CN103914822A (en) * | 2012-12-31 | 2014-07-09 | 清华大学 | Interactive video foreground object extraction method based on super pixel segmentation |
| CN103116630A (en) * | 2013-02-03 | 2013-05-22 | 苏州市职业大学 | Color image search method based on scale invariant feature transform (sift) seed region growing |
| CN105701810A (en) * | 2016-01-12 | 2016-06-22 | 湖南中航天目测控技术有限公司 | Unmanned aerial vehicle aerial image electronic delineating method based on click type image segmentation |
| CN105809673A (en) * | 2016-03-03 | 2016-07-27 | 上海大学 | SURF (Speeded-Up Robust Features) algorithm and maximal similarity region merging based video foreground segmentation method |
| CN106127791A (en) * | 2016-07-06 | 2016-11-16 | 武汉大学 | A kind of contour of building line drawing method of aviation remote sensing image |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783732A (en) * | 2019-01-10 | 2019-05-21 | 珠海金山网络游戏科技有限公司 | A kind of micro- end construction method of game based on click density |
| CN109783732B (en) * | 2019-01-10 | 2023-07-04 | 珠海金山数字网络科技有限公司 | Game micro-terminal construction method based on click density |
| CN115293996A (en) * | 2022-10-08 | 2022-11-04 | 杭州群核信息技术有限公司 | Image toning method, device and storage medium |
| WO2024074060A1 (en) * | 2022-10-08 | 2024-04-11 | 杭州群核信息技术有限公司 | Image toning method and apparatus and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Caesar et al. | Coco-stuff: Thing and stuff classes in context | |
| Li et al. | Reconstructing building mass models from UAV images | |
| CN110210387B (en) | Method, system and device for detecting insulator target based on knowledge graph | |
| CN110188635A (en) | A kind of plant pest recognition methods based on attention mechanism and multi-level convolution feature | |
| CN102592132A (en) | System and method for classifying digital image data | |
| CN105096259A (en) | Depth value restoration method and system for depth image | |
| CN104732506A (en) | Character picture color style converting method based on face semantic analysis | |
| CN108447064B (en) | Picture processing method and device | |
| CN104166983A (en) | Motion object real time extraction method of Vibe improvement algorithm based on combination of graph cut | |
| Peng et al. | Iterated graph cuts for image segmentation | |
| Liu et al. | Unsupervised image segmentation based on analysis of binary partition tree for salient object extraction | |
| CN102915541A (en) | Multi-scale image segmenting method | |
| CN104462526B (en) | More people towards high-resolution remote sensing image cooperate with rapid vector method online | |
| CN106203454A (en) | The method and device that certificate format is analyzed | |
| CN106127817A (en) | A kind of image binaryzation method based on passage | |
| CN108765428A (en) | A kind of target object extracting method based on click interaction | |
| CN113784171A (en) | Video data processing method, device, computer system and readable storage medium | |
| Zhou et al. | An efficient two-stage region merging method for interactive image segmentation | |
| EP3053137B1 (en) | Method and apparatus for generating superpixel clusters | |
| CN111210434A (en) | Image replacement method and system based on sky identification | |
| CN104036272A (en) | Text detection method and electronic device | |
| CN117746130A (en) | Weak supervision deep learning classification method based on remote sensing image punctiform semantic tags | |
| Herchenbach et al. | Segmentation and classification of THCs on PCBAs | |
| CN106228553A (en) | High-resolution remote sensing image shadow Detection apparatus and method | |
| Happ et al. | Towards distributed region growing image segmentation based on MapReduce |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |