CN110211134A - A kind of image partition method and device, electronic equipment and storage medium - Google Patents
A kind of image partition method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110211134A CN110211134A CN201910464349.3A CN201910464349A CN110211134A CN 110211134 A CN110211134 A CN 110211134A CN 201910464349 A CN201910464349 A CN 201910464349A CN 110211134 A CN110211134 A CN 110211134A
- Authority
- CN
- China
- Prior art keywords
- image
- image segmentation
- segmentation result
- interactive operation
- obtains
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
This disclosure relates to a kind of image partition method and device, electronic equipment and storage medium, wherein the described method includes: carrying out image segmentation according to the regional location of image where object to be split to described image, obtaining image segmentation result；Exist in described image segmentation result and obtains the first interactive operation in wrong subregional situation；First interactive operation is responded, obtains carrying out correction processing to described image segmentation result according to the annotation results for the subregional annotation results of mistake.Using the disclosure, the accuracy of image segmentation result can be improved, to position in time to lesion.
Description
Technical field
This disclosure relates to technical field of computer vision more particularly to a kind of image partition method and device, electronic equipment
And storage medium.
Background technique
Deep learning is fast-developing, obtains prominent achievement in image segmentation field.Image segmentation skill based on deep learning
Art, takes to mark obtaining annotation results by hand, does not have enough stability, thus obtained image point to annotation results accuracy
It is not high to cut result accuracy.There is no effective solution in the related technology.
Summary of the invention
The present disclosure proposes a kind of image Segmentation Technology schemes.
According to the one side of the disclosure, a kind of image partition method is provided, which comprises
According to the regional location of image where object to be split, image segmentation is carried out to described image, obtains image segmentation
As a result；
Exist in described image segmentation result and obtains the first interactive operation in wrong subregional situation；
First interactive operation is responded, is obtained for the subregional annotation results of mistake, according to the annotation results
Correction processing is carried out to described image segmentation result.
Using the disclosure, by the regional location of image where object to be split, image segmentation is carried out to described image, is obtained
To image segmentation result, available first interactive operation responds first interactive operation, obtains for the wrong subregion
Annotation results, automatic marking is realized by interactive operation.The annotation results obtained according to automatic marking obtain segmentation
Image segmentation result is corrected automatically, so as to improve the accuracy of image segmentation result, to determine in time lesion
Position.
It is described according to image where object to be split in the case that described image is 2D image in possible implementation
Regional location, to described image carry out image segmentation, obtain image segmentation result, comprising:
The extracting parameter of the corresponding regional location is determined according to the second interactive operation；
Rectangle frame is obtained according to the extracting parameter, image segmentation is carried out to the 2D image according to the rectangle frame, is obtained
To 2D image segmentation result.
Using the disclosure, extracting parameter is obtained according to the second interactive operation, the rectangle frame obtained according to extracting parameter is to 2D
Image carries out image segmentation and intuitively obtains 2D image segmentation result using visualization effect.
It is described that image segmentation is carried out to the 2D image according to the rectangle frame in possible implementation, obtain 2D figure
As segmentation result, comprising:
2D image segmentation network is inputted by the 2D image and according to the rectangular area that the rectangle frame obtains, output obtains
The 2D image segmentation result.
Using the disclosure, 2D image segmentation result is obtained by the processing of 2D image segmentation network, more than artificial treatment
Add accurately, the accuracy of image segmentation result can be improved.
In possible implementation, it is described exist in described image segmentation result obtain first in wrong subregional situation
Interactive operation, comprising:
The mistake subregion is that the target object mistake as prospect is divided into background, obtains first interactive operation；
Response first interactive operation, obtains for the subregional annotation results of mistake, comprising:
First interactive operation is parsed according to preset operation description information, obtains first interactive operation
First annotation results of corresponding description, first annotation results are identified with the first mark information, are marked by described first
The target object mistake as prospect is divided into background described in note information representation.
Using the disclosure, the first interactive operation is obtained according to wrong subregional judgement, is realized according to the first interactive operation
Automatic marking obtains the first annotation results of the corresponding description of the first interactive operation, is obtained by the first annotation results to segmentation
Image segmentation result is corrected automatically, so as to improve the accuracy of image segmentation result.
In possible implementation, first interactive operation includes: the input operation of left mouse button, the first specified touching
Any one in control operation, the first specified slide.
Using the disclosure, the first interactive operation is visual operation, enables automatic marking more convenient and easy-to-use.
In possible implementation, correction processing is carried out to described image segmentation result according to the annotation results, comprising:
2D image, 2D image segmentation result and the first annotation results are inputted into 2D image rectification network, output obtains 2D figure
As segmentation result.
Using the disclosure, 2D image segmentation result is obtained by 2D image rectification network, it is more more accurate than artificial treatment,
The accuracy of image segmentation result can be improved.
In possible implementation, it is described exist in described image segmentation result obtain first in wrong subregional situation
Interactive operation, comprising:
The mistake subregion is that the part mistake for belonging to background is divided into the target object as prospect, obtains described first and hands over
Interoperability；
Response first interactive operation, obtains for the subregional annotation results of mistake, comprising:
First interactive operation is parsed according to preset operation description information, obtains first interactive operation
Second annotation results of corresponding description, second annotation results are identified with the second mark information, are marked by described second
The part mistake for belonging to background is divided into the target object as prospect described in note information representation.
Using the disclosure, the first interactive operation is obtained according to wrong subregional judgement, is realized according to the first interactive operation
Automatic marking obtains the second annotation results of the corresponding description of the first interactive operation, is obtained by the second annotation results to segmentation
Image segmentation result is corrected automatically, so as to improve the accuracy of image segmentation result.
In possible implementation, first interactive operation includes: the input operation of right mouse button, the second specified touching
Any one in control operation, the second specified slide.
Using the disclosure, the first interactive operation is visual operation, enables automatic marking more convenient and easy-to-use.
In possible implementation, correction processing is carried out to described image segmentation result according to the annotation results, comprising:
2D image, 2D image segmentation result and second annotation results are inputted into 2D image rectification network, output obtains
2D image segmentation result.
Using the disclosure, 2D image segmentation result is obtained by 2D image rectification network, it is more more accurate than artificial treatment,
The accuracy of image segmentation result can be improved.
In possible implementation, during carrying out correction processing to described image segmentation result, pass through 2D image
Network is corrected to carry out handling targeted image data being the pixel data demarcated by two-dimensional coordinate system.
Using the disclosure, targeted image data is the pixel data demarcated by two-dimensional coordinate system, is come to 2D image
It says, by the available 2D image segmentation result of image rectification network, the accuracy of 2D image segmentation result can be improved.
According to the one side of the disclosure, a kind of image segmentation device is provided, described device includes:
Divide module, for the regional location according to image where object to be split, image segmentation carried out to described image,
Obtain image segmentation result；
Operation obtains module, and the first interaction is obtained in wrong subregional situation for existing in described image segmentation result
Operation；
Respond module is operated, for responding first interactive operation, is obtained for the subregional annotation results of mistake,
Correction processing is carried out to described image segmentation result according to the annotation results.
In possible implementation, in the case that described image is 2D image, the segmentation module is further used for:
The extracting parameter of the corresponding regional location is determined according to the second interactive operation；
Rectangle frame is obtained according to the extracting parameter, image segmentation is carried out to the 2D image according to the rectangle frame, is obtained
To 2D image segmentation result.
In possible implementation, the segmentation module is further used for:
2D image segmentation network is inputted by the 2D image and according to the rectangular area that the rectangle frame obtains, output obtains
The 2D image segmentation result.
In possible implementation, the operation obtains module, is further used for:
The mistake subregion is that the target object mistake as prospect is divided into background, obtains first interactive operation；
The operation respond module, is further used for:
First interactive operation is parsed according to preset operation description information, obtains first interactive operation
First annotation results of corresponding description, first annotation results are identified with the first mark information, are marked by described first
The target object mistake as prospect is divided into background described in note information representation.
In possible implementation, the operation respond module is further used for:
2D image, 2D image segmentation result and first annotation results are inputted into 2D image rectification network, output obtains
The 2D image segmentation result.
In possible implementation, the operation obtains module, is further used for:
The mistake subregion is that the part mistake for belonging to background is divided into the target object as prospect, obtains described first and hands over
Interoperability；
The operation respond module, is further used for:
First interactive operation is parsed according to preset operation description information, obtains first interactive operation
Second annotation results of corresponding description, second annotation results are identified with the second mark information, are marked by described second
The part mistake for belonging to background is divided into the target object as prospect described in note information representation.
In possible implementation, the operation respond module is further used for:
2D image, 2D image segmentation result and second annotation results are inputted into 2D image rectification network, output obtains
2D image segmentation result.
In possible implementation, during carrying out correction processing to described image segmentation result, pass through 2D image
Network is corrected to carry out handling targeted image data being the pixel data demarcated by two-dimensional coordinate system.
According to the one side of the disclosure, a kind of electronic equipment is provided, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to: execute above-mentioned image partition method.
According to the one side of the disclosure, a kind of computer readable storage medium is provided, computer program is stored thereon with
Instruction, the computer program instructions realize above-mentioned image partition method when being executed by processor.
In the embodiments of the present disclosure, according to the regional location of image where object to be split, to described image, (such as 2D schemes
Picture) image segmentation is carried out, obtain image segmentation result (such as 2D image segmentation result)；There is mistake in described image segmentation result
The first interactive operation is obtained in subregional situation；First interactive operation is responded, is obtained for the subregional mark of mistake
Note is as a result, carry out correction processing to described image segmentation result according to the annotation results.Using the disclosure, pass through above-mentioned interaction
Operation to realize automatic marking, can according to the obtained annotation results of automatic marking to the obtained image segmentation result of segmentation into
Row is automatic to be corrected, so as to improve the accuracy of image segmentation result, to position in time to lesion.
It should be understood that above general description and following detailed description is only exemplary and explanatory, rather than
Limit the disclosure.
According to below with reference to the accompanying drawings to detailed description of illustrative embodiments, the other feature and aspect of the disclosure will become
It is clear.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and those figures show meet this public affairs
The embodiment opened, and together with specification it is used to illustrate the technical solution of the disclosure.
Fig. 1 shows the flow chart of the image partition method according to the embodiment of the present disclosure.
Fig. 2 shows the flow charts according to the image partition method of the embodiment of the present disclosure.
Fig. 3 is shown according to the disclosure for image segmentation flow chart of the image based on automatic interaction.
Fig. 4 shows the block diagram of the image segmentation device according to the embodiment of the present disclosure.
Fig. 5 shows the block diagram of the electronic equipment according to the embodiment of the present disclosure.
Fig. 6 shows the block diagram of the electronic equipment according to the embodiment of the present disclosure.
Specific embodiment
Various exemplary embodiments, feature and the aspect of the disclosure are described in detail below with reference to attached drawing.It is identical in attached drawing
Appended drawing reference indicate element functionally identical or similar.Although the various aspects of embodiment are shown in the attached drawings, remove
It non-specifically points out, it is not necessary to attached drawing drawn to scale.
Dedicated word " exemplary " means " being used as example, embodiment or illustrative " herein.Here as " exemplary "
Illustrated any embodiment should not necessarily be construed as preferred or advantageous over other embodiments.
The terms "and/or", only a kind of incidence relation for describing affiliated partner, indicates that there may be three kinds of passes
System, for example, A and/or B, can indicate: individualism A exists simultaneously A and B, these three situations of individualism B.In addition, herein
Middle term "at least one" indicate a variety of in any one or more at least two any combination, it may for example comprise A,
B, at least one of C can indicate to include any one or more elements selected from the set that A, B and C are constituted.
In addition, giving numerous details in specific embodiment below to better illustrate the disclosure.
It will be appreciated by those skilled in the art that without certain details, the disclosure equally be can be implemented.In some instances, for
Method, means, element and circuit well known to those skilled in the art are not described in detail, in order to highlight the purport of the disclosure.
One application direction of image procossing is: the processing by optimizing medical image obtains clearer lesion, with side
Doctor is helped accurately to understand the whole disease condition of patient.The purpose of medical image segmentation is that will have certain in medical image
The partial segmentation of a little particular meanings (such as focal part) comes out, and extracts correlated characteristic.This is the key that medical image analysis link
Technology plays increasing effect in medical imaging.Image segmentation is not only to extract particular tissues in imaged image
The indispensable means of quantitative information, while being also the Visual Implementation and pretreated step and premise.Image after segmentation
Be widely used in the quantitative analysis such as tissue volume, diagnosis, the positioning of pathological tissues, the study of anatomical structure, treatment planning,
In the application scenarios such as the correction of local bulk effect and computer guidance operation of functional imaging data.In medical imaging field, diagnosis
The acquisition and image interpretation that image is depended on assessment, with the fast development of medical imaging equipment and universal, imaging technique packet
Magnetic resonance imaging (MR), computed tomography (CT), ultrasound, positron emission computerized tomography (PET) etc. are included, with speed faster
Degree and higher resolution ratio collect a large amount of medical imaging data, but the image processing function of related medical image is but still by efficiency
Lowly, subjectivity is strong and is easy to produce the progress of the manual type of fatigue, and particularly with three-dimensional image data, doctor is for organ
It is successively carried out with the segmentation need of work of lesion, image processing efficiency is low.
It can carry out image segmentation using deep learning to come out, it is substantially the classification of pixel scale to image segmentation, i.e.,
Judge each pixel generic on image.Convolutional neural networks realize output figure using mode of learning end to end
As the classification results of Pixel-level.Convolutional neural networks model based on U-Net structure uses coding-decoding structure and jump
Jump connection structure, significantly more efficient extraction characteristics of image, and feature is handed on, encoded pixels classification results, in image point
It cuts and obtains preferable effect in task.However, being taken based on voxel (pixel) segmentation in image segmentation process and based on area
The method of regional partition, there are threshold value choose it is difficult, it is sensitive to picture noise the problems such as.Image segmentation is carried out based on deep learning
In the related technology, only simply the dividing method based on deep learning is applied in medical image segmentation field, these is asked
Topic not can solve.It is simple to rely on depth since result of the medical application to image segmentation has compared with strict requirements
Method annotation results (mark by hand) accuracy of habit does not have enough stability；In addition, deep learning method lacks to special
Image type adaptability, it is desirable to reach preferable segmentation effect on particular kind of medical image, generally require a large amount of
Finely the data of mark segmentation are trained.These problems to directly apply to deep learning into actual medical image
Lack feasibility in segmentation task.
The disclosure is a kind of Interactive Segmentation scheme based on deep learning, the mesh to be split that can be provided according to labeler
The segmentation task to multiple types segmentation object is completed in substantially rectangular region where marking；If dividing as a result, it has been found that occurring wrong
Accidentally, then labeler can region to segmentation errors according to pre-defined rule interact formula mark, obtained annotation results are defeated
Enter the convolutional neural networks for correcting mistake, is entangled automatically to be realized to the wrong subregion found in image segmentation result
Just, by replacing the manual of the relevant technologies to mark, the disclosure improves annotating efficiency by the interactive mode cutting scheme, to mention
The high Stability and veracity of image segmentation.The segmentation of multiple types segmentation object can also be solved the problems, such as simultaneously.
Fig. 1 shows the flow chart of the image partition method according to the embodiment of the present disclosure, which is applied to figure
As segmenting device, for example, image segmentation device can be executed by terminal device or server or other processing equipments, wherein eventually
End equipment can for user equipment (UE, User Equipment), mobile device, cellular phone, wireless phone, at individual digital
Manage (PDA, Personal Digital Assistant), handheld device, calculating equipment, mobile unit, wearable device etc..?
In some possible implementations, which can be computer-readable by storing in processor calling memory
The mode of instruction is realized.As shown in Figure 1, the process includes:
Step S101, according to the regional location of image where object to be split, image segmentation is carried out to described image, is obtained
Image segmentation result.
In one example, in the case that described image is 2D image, rectangle frame, the rectangle region provided such as labeler can be used
Domain is formed by rectangle frame and chooses to object to be split in present image.Wherein, the rectangular area is to be split for characterizing
The regional location of 2D image where object.Object to be split can also claim for the lesion in medical image, the regional location
For the position where focal area.
In one example, image segmentation is carried out to the 2D image, it can be using trained model, such as 2D image point
It cuts network model and carries out image segmentation.
Step S102, exist in described image segmentation result and obtain the first interactive operation in wrong subregional situation.
Step S103, first interactive operation is responded, is obtained for the subregional annotation results of mistake, according to described
Annotation results carry out correction processing to described image segmentation result.
In one example, correction processing is carried out to the 2D image, it can be using trained model, as 2D image entangles
Positive network model carries out image rectification processing.
The disclosure is used, if there is mistake in image segmentation result (such as 2D image segmentation result), without mark by hand
Note, labeler can region to segmentation errors according to pre-defined rule interact formula mark, obtained annotation results are inputted
For correcting the convolutional neural networks of mistake, entangled automatically to be realized to the wrong subregion found in image segmentation result
Just, the cutting scheme of this interactive mode improves annotating efficiency, to improve the Stability and veracity of image segmentation.
Fig. 2 shows the flow chart according to the image partition method of the embodiment of the present disclosure, which is applied to figure
As segmenting device, for example, image segmentation device can be executed by terminal device or server or other processing equipments, wherein eventually
End equipment can for user equipment (UE, User Equipment), mobile device, cellular phone, wireless phone, at individual digital
Manage (PDA, Personal Digital Assistant), handheld device, calculating equipment, mobile unit, wearable device etc..?
In some possible implementations, which can be computer-readable by storing in processor calling memory
The mode of instruction is realized.As shown in Fig. 2, the process includes:
Step S201, the extracting parameter of 2D image-region position where determining corresponding object to be split.
In one example, it can determine that the extraction of the corresponding regional location is joined according to interactive operation (such as the second interactive operation)
Number.In approximate region position of the object to be split provided according to user's (labeler) in 2D image, that is, pass through the second interaction
Available target area after obtained rectangle frame is divided in the picture is operated, so as to carry out image point to 2D image
It cuts.
It should be pointed out that the execution that above-mentioned first interactive operation and second interactive operation do not represent interactive operation is suitable
Sequence is only respectively referred to " first ", " second " for different interactive operations.It first passes through the first step and square is obtained based on interactive operation
Shape frame carries out " 2D image segmentation ", obtains " thick " segmentation result, and then " the point mark " by second step based on interactive operation is right
Wrong subregion in " 2D image segmentation " is labeled, and is corrected according to annotation results to wrong subregion.
Step S202, rectangle frame is obtained according to the extracting parameter, figure is carried out to the 2D image according to the rectangle frame
As segmentation, 2D image segmentation result is obtained.
In the possible implementation of the disclosure, inputted by the 2D image and according to the rectangular area that the rectangle frame obtains
2D image segmentation network, output obtain the 2D image segmentation result.
Step S203, exist in the 2D image segmentation result and obtain the first interactive operation in wrong subregional situation.
Step S204, first interactive operation is responded, is obtained for the subregional annotation results of mistake, according to described
Annotation results carry out correction processing to the 2D image segmentation result.
For wrong subregion there are two types of situation, a kind of situation is: the part for belonging to prospect is divided into background by mistake.Wherein,
The part for belonging to prospect is that user thinks target object to be processed, the i.e. specific location of lesion；Another situation is that: back will be belonged to
The part mistake of scape is divided into the target object as prospect.It is described in detail below:
Situation one: existing in 2D image segmentation result in wrong subregional situation, which is that will be used as prospect
Target object mistake be divided into background, obtain first interactive operation.Respond the first interactive operation, by the first interactive operation according to
Preset operation description information is parsed, and the first annotation results of the corresponding description of the first interactive operation are obtained.First mark
As a result it is identified, is characterized by the first mark information described by conduct with the first mark information (can such as use Green Marker point)
The target object mistake of prospect is divided into background.Wherein, which includes: the input operation of left mouse button, specified the
Any one in one touch control operation, the first specified slide.The present disclosure is not limited to be left mouse button operation, it is also possible to
Screen touch control operation (for example, clicking in current region finger, touch-control is primary, can be above-mentioned first touch control operation), may be used also
It can be that track slide (for example, in the cursor of current region finger or mouse side sliding from left to right etc., is ok
For above-mentioned first slide).
In this case, 2D image, 2D image segmentation result and the first annotation results can be inputted into 2D image rectification net
Network, output obtain 2D image segmentation result.
Situation two: existing in 2D image segmentation result in wrong subregional situation, which is that will belong to background
Part mistake be divided into the target object as prospect, obtain first interactive operation.First interactive operation is responded, by institute
It states the first interactive operation to be parsed according to preset operation description information, obtains the of the corresponding description of first interactive operation
Two annotation results, second annotation results with the second mark information (be such as different from the one other identification of above-mentioned " Green Marker point ",
Such as red-label point) it is identified, the part mistake that will belong to background is characterized by second mark information and is divided into conduct
The target object of prospect.Wherein, first interactive operation include: the input operation of right mouse button, the second specified touch control operation,
Any one in the second specified slide.The present disclosure is not limited to be right mouse button operation, it is also possible to screen touch-control behaviour
Make (can be above-mentioned second touch control operation if touch-control is secondary for example, double-clicking in current region), it is also possible to track sliding
Operation (can be above-mentioned second slide in current region side sliding from right to left etc.).
In this case, 2D image, 2D image segmentation result and the second annotation results can be inputted into 2D image rectification net
Network, output obtain 2D image segmentation result.
In the possible implementation of the disclosure, the 2D image rectification network in the process of processing, for figure
As data are the pixel datas demarcated by two-dimensional coordinate system, i.e., the pixel demarcated by horizontal axis x in two-dimensional coordinate system and longitudinal axis y
Data.
In the possible implementation of the disclosure, using 2D image rectification network, the wrong subregion in image is corrected
During, be the correction to the local updating carried out in local a certain range wrong in image and be dynamic update, rather than
This image is all replaced.The purpose of local updating: being that the multiple pixels for being included are handled to mark point near zone,
Obtain mark point pixel how far belongs in image wants the region of local updating apart from this by differentiating the relationship between pixel.Such as
What differentiates the relationship between pixel to carry out concrete mode used by image segmentation, and can be take index geodesic distance as guidance
Image segmentation (egd-guide graph cut) algorithm.For the correction of local updating, using egd-guide graph cut
Method is completed to be split in the example that local updating is corrected using image segmentation (graphcut), needs to scheme one
As being first converted into graph structure, using each pixel in image as the node in graph structure, by the distance between pixel as figure
The weight on side in structure.Due to image segmentation (graphcut) need one measurement pixel distance method come to all sides into
Row initialization, local updating can not be achieved the effect that using traditional graphcut, therefore, it is necessary to index geodesic distance (such as
The geodesic distance of exponential damping) it first all sides is initialized (is obtained not with the mode of the geodesic distance of exponential damping under guidance
The distance of disconnected decaying, to give graphcut algorithm initialization side right weight), local updating then just may be implemented, local updating
Range is dynamic.That is, first there is index ranging guide, the optional regional scope of distance mark point is obtained, is then based on
This optional regional scope application graphcut algorithm advanced optimizes, and to realize image segmentation, is finally reached local updating
Effect.For geodesic distance, geodesic distance is a metric form for measuring distance between pixel in image, passes through geodetic
Away from distance spatially between two pixels can not only being measured, also comprising distance semantically.Distance semantically is come
It says, such as two pixels on an image, spatially may be adjacent, but belong to cat in semantically first pixel,
Second pixel belongs to that blanket of cat recumbency, wishes to find a kind of measurement mode this when to define: the two pixels
Though spatially close to " semantic distance " be actually it is far, need just to be able to achieve by finding distance semantically
This measurement.
Using example:
Fig. 3 is shown according to the disclosure for the image segmentation flow chart of image (such as 2D image) based on automatic interaction.Including
Following content:
One, the labeler for participating in interactive segmentation provides object to be split and is scheming by way of drawing rectangle frame in the picture
As in approximate location (rectangular area 21 as shown in Figure 3), the location information of rectangular area 21 can be encoded with it is to be split
Image 22 is input in 2D image segmentation network 23 together and is split, will by the 2D image segmentation network 23 of the first step
Export segmentation result 24.
2D image segmentation network 23 can be the convolutional neural networks of U-Net structure.By through 2D image segmentation network 23
Rectangular area that original image (image 22 to be split) and labeler provide is received as inputting, complete to lesion in rectangular area or
The identification and segmentation of the objects such as organ.
Two, exported segmentation result 24 is split through 2D image segmentation network 23 for the first step, if labeler
Satisfied to the segmentation result, then segmentation terminates, if dissatisfied, labeler carries out click mark to the region of erroneous segmentation.
Wherein, labeler can carry out the mark of two categories, the first indicates the portion for belonging to prospect for the first mark point 261 in Fig. 3
Divide and background is divided by mistake；It is for second that the second mark point 262 indicates that the part for belonging to background by erroneous segmentation is prospect in Fig. 3
Situation.
Three, the segmentation result 24 exported by original image (image 22 to be split), currently is input to together with wrong minute mark note
In the 2D image rectification network 25 of second step, the result divided by 25 Duis of 2D image rectification network is modified, and output is corrected
Segmentation result 27 afterwards after any time by 2D image rectification network 25 and labeler interact, is realized to stablize and quickly be cured
Learn the correction processing of image segmentation.
2D image rectification network 25 can use neural network/figure neural network.It is received by 2D image rectification network 25
The wrong minute mark note that original image (image 22 to be split), the segmentation result 24 of first step output and labeler provide is complete as input
The local correction of pairs of previous step segmentation result, or local updating is completed using the methods of egd-guide graph cut
Correction procedure.
It will be understood by those skilled in the art that each step writes sequence simultaneously in the above method of specific embodiment
It does not mean that stringent execution sequence and any restriction is constituted to implementation process, the specific execution sequence of each step should be with its function
It can be determined with possible internal logic.
Above-mentioned each embodiment of the method that the disclosure refers to can phase each other without prejudice to principle logic
The embodiment formed after combining is mutually combined, as space is limited, the disclosure repeats no more.
In addition, the disclosure additionally provides image segmentation device, electronic equipment, computer readable storage medium, program, it is above-mentioned
It can be used to realize any image partition method that the disclosure provides, corresponding technical solution and description and referring to method part
It is corresponding to record, it repeats no more.
Fig. 4 shows the block diagram of the image segmentation device according to the embodiment of the present disclosure, as shown in figure 4, the embodiment of the present disclosure
Image segmentation device, comprising: segmentation module 31, for the regional location according to image where object to be split, to described image
Image segmentation is carried out, image segmentation result is obtained；Operation obtains module 32, for there is wrong point in described image segmentation result
The first interactive operation is obtained in the case where region；Operation respond module 33 is directed to for responding first interactive operation
The subregional annotation results of mistake carry out correction processing to described image segmentation result according to the annotation results.
In the possible implementation of the disclosure, in the case that described image is 2D image, the segmentation module is further used
In: the extracting parameter of the corresponding regional location is determined according to the second interactive operation；Rectangle frame is obtained according to the extracting parameter,
Image segmentation is carried out to the 2D image according to the rectangle frame, obtains 2D image segmentation result.
In the possible implementation of the disclosure, the segmentation module is further used for: by the 2D image and according to described
The rectangular area that rectangle frame obtains inputs 2D image segmentation network, and output obtains the 2D image segmentation result.
In the possible implementation of the disclosure, the operation obtains module, is further used for: the mistake subregion is that will make
It is divided into background for the target object mistake of prospect, obtains first interactive operation；The operation respond module, is further used for:
First interactive operation is parsed according to preset operation description information, obtains the corresponding description of first interactive operation
The first annotation results, first annotation results are identified with the first mark information, pass through the first mark information table
It levies and described the target object mistake as prospect is divided into background.First interactive operation include: left mouse button input operation,
Any one in specified the first touch control operation, the first specified slide.
In the possible implementation of the disclosure, the operation respond module is further used for: by the 2D image, described
2D image segmentation result and first annotation results input 2D image rectification network, and output obtains the 2D image segmentation knot
Fruit.
In the possible implementation of the disclosure, the operation obtains module, is further used for: the mistake subregion is that will belong to
It is divided into the target object as prospect in the part mistake of background, obtains first interactive operation；The operation respond module, into
One step is used for: first interactive operation being parsed according to preset operation description information, obtains the first interaction behaviour
Make the second annotation results of corresponding description, second annotation results are identified with the second mark information, pass through described second
Mark information characterizes the part mistake that will belong to background and is divided into target object as prospect.The first interactive operation packet
It includes: the input operation of right mouse button, the second specified touch control operation, any one in the second specified slide.
In the possible implementation of the disclosure, the operation respond module is further used for: by the 2D image, described
2D image segmentation result and second annotation results input 2D image rectification network, and output obtains the 2D image segmentation knot
Fruit.
In the possible implementation of the disclosure, the 2D image rectification network in the process of processing, for figure
As data are the pixel datas demarcated by two-dimensional coordinate system.
In some embodiments, the embodiment of the present disclosure provides the function that has of device or comprising module can be used for holding
The method of row embodiment of the method description above, specific implementation are referred to the description of embodiment of the method above, for sake of simplicity, this
In repeat no more.
The embodiment of the present disclosure also proposes a kind of computer readable storage medium, is stored thereon with computer program instructions, institute
It states when computer program instructions are executed by processor and realizes the above method.Computer readable storage medium can be non-volatile meter
Calculation machine readable storage medium storing program for executing.
The embodiment of the present disclosure also proposes a kind of electronic equipment, comprising: processor；For storage processor executable instruction
Memory；Wherein, the processor is configured to the above method.
The equipment that electronic equipment may be provided as terminal, server or other forms.
Fig. 5 is the block diagram of a kind of electronic equipment 800 shown according to an exemplary embodiment.For example, electronic equipment 800 can
To be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, Medical Devices are good for
Body equipment, the terminals such as personal digital assistant.
Referring to Fig. 5, electronic equipment 800 may include following one or more components: processing component 802, memory 804,
Power supply module 806, multimedia component 808, audio component 810, the interface 812 of input/output (I/O), sensor module 814,
And communication component 816.
The integrated operation of the usual controlling electronic devices 800 of processing component 802, such as with display, call, data are logical
Letter, camera operation and record operate associated operation.Processing component 802 may include one or more processors 820 to hold
Row instruction, to perform all or part of the steps of the methods described above.In addition, processing component 802 may include one or more moulds
Block, convenient for the interaction between processing component 802 and other assemblies.For example, processing component 802 may include multi-media module, with
Facilitate the interaction between multimedia component 808 and processing component 802.
Memory 804 is configured as storing various types of data to support the operation in electronic equipment 800.These data
Example include any application or method for being operated on electronic equipment 800 instruction, contact data, telephone directory
Data, message, picture, video etc..Memory 804 can by any kind of volatibility or non-volatile memory device or it
Combination realize, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable
Except programmable read only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, fastly
Flash memory, disk or CD.
Power supply module 806 provides electric power for the various assemblies of electronic equipment 800.Power supply module 806 may include power supply pipe
Reason system, one or more power supplys and other with for electronic equipment 800 generate, manage, and distribute the associated component of electric power.
Multimedia component 808 includes the screen of one output interface of offer between the electronic equipment 800 and user.
In some embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch surface
Plate, screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touches
Sensor is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding
The boundary of movement, but also detect duration and pressure associated with the touch or slide operation.In some embodiments,
Multimedia component 808 includes a front camera and/or rear camera.When electronic equipment 800 is in operation mode, as clapped
When taking the photograph mode or video mode, front camera and/or rear camera can receive external multi-medium data.It is each preposition
Camera and rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 810 is configured as output and/or input audio signal.For example, audio component 810 includes a Mike
Wind (MIC), when electronic equipment 800 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone
It is configured as receiving external audio signal.The received audio signal can be further stored in memory 804 or via logical
Believe that component 816 is sent.In some embodiments, audio component 810 further includes a loudspeaker, is used for output audio signal.
I/O interface 812 provides interface between processing component 802 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lock
Determine button.
Sensor module 814 includes one or more sensors, for providing the state of various aspects for electronic equipment 800
Assessment.For example, sensor module 814 can detecte the state that opens/closes of electronic equipment 800, the relative positioning of component, example
As the component be electronic equipment 800 display and keypad, sensor module 814 can also detect electronic equipment 800 or
The position change of 800 1 components of electronic equipment, the existence or non-existence that user contacts with electronic equipment 800, electronic equipment 800
The temperature change of orientation or acceleration/deceleration and electronic equipment 800.Sensor module 814 may include proximity sensor, be configured
For detecting the presence of nearby objects without any physical contact.Sensor module 814 can also include optical sensor,
Such as CMOS or ccd image sensor, for being used in imaging applications.In some embodiments, which may be used also
To include acceleration transducer, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 816 is configured to facilitate the communication of wired or wireless way between electronic equipment 800 and other equipment.
Electronic equipment 800 can access the wireless network based on communication standard, such as WiFi, 2G or 3G or their combination.Show at one
In example property embodiment, communication component 816 receives broadcast singal or broadcast from external broadcasting management system via broadcast channel
Relevant information.In one exemplary embodiment, the communication component 816 further includes near-field communication (NFC) module, short to promote
Cheng Tongxin.For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band can be based in NFC module
(UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, electronic equipment 800 can be by one or more application specific integrated circuit (ASIC), number
Word signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
In the exemplary embodiment, a kind of non-volatile computer readable storage medium storing program for executing is additionally provided, for example including calculating
The memory 804 of machine program instruction, above-mentioned computer program instructions can be executed by the processor 820 of electronic equipment 800 to complete
The above method.
Fig. 6 is the block diagram of a kind of electronic equipment 900 shown according to an exemplary embodiment.For example, electronic equipment 900 can
To be provided as a server.Referring to Fig. 6, it further comprises one or more that electronic equipment 900, which includes processing component 922,
Processor, and the memory resource as representated by memory 932, for store can by the instruction of the execution of processing component 922,
Such as application program.The application program stored in memory 932 may include it is one or more each correspond to one
The module of group instruction.In addition, processing component 922 is configured as executing instruction, to execute the above method.
Electronic equipment 900 can also include that a power supply module 926 is configured as executing the power supply pipe of electronic equipment 900
Reason, a wired or wireless network interface 950 are configured as electronic equipment 900 being connected to network and an input and output (I/
O) interface 958.Electronic equipment 900 can be operated based on the operating system for being stored in memory 932, such as Windows
ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
In the exemplary embodiment, a kind of non-volatile computer readable storage medium storing program for executing is additionally provided, for example including calculating
The memory 932 of machine program instruction, above-mentioned computer program instructions can be executed by the processing component 922 of electronic equipment 900 with complete
At the above method.
The disclosure can be system, method and/or computer program product.Computer program product may include computer
Readable storage medium storing program for executing, containing for making processor realize the computer-readable program instructions of various aspects of the disclosure.
Computer readable storage medium, which can be, can keep and store the tangible of the instruction used by instruction execution equipment
Equipment.Computer readable storage medium for example can be-- but it is not limited to-- storage device electric, magnetic storage apparatus, optical storage
Equipment, electric magnetic storage apparatus, semiconductor memory apparatus or above-mentioned any appropriate combination.Computer readable storage medium
More specific example (non exhaustive list) includes: portable computer diskette, hard disk, random access memory (RAM), read-only deposits
It is reservoir (ROM), erasable programmable read only memory (EPROM or flash memory), static random access memory (SRAM), portable
Compact disk read-only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanical coding equipment, for example thereon
It is stored with punch card or groove internal projection structure and the above-mentioned any appropriate combination of instruction.Calculating used herein above
Machine readable storage medium storing program for executing is not interpreted that instantaneous signal itself, the electromagnetic wave of such as radio wave or other Free propagations lead to
It crosses the electromagnetic wave (for example, the light pulse for passing through fiber optic cables) of waveguide or the propagation of other transmission mediums or is transmitted by electric wire
Electric signal.
Computer-readable program instructions as described herein can be downloaded to from computer readable storage medium it is each calculate/
Processing equipment, or outer computer or outer is downloaded to by network, such as internet, local area network, wide area network and/or wireless network
Portion stores equipment.Network may include copper transmission cable, optical fiber transmission, wireless transmission, router, firewall, interchanger, gateway
Computer and/or Edge Server.Adapter or network interface in each calculating/processing equipment are received from network to be counted
Calculation machine readable program instructions, and the computer-readable program instructions are forwarded, for the meter being stored in each calculating/processing equipment
In calculation machine readable storage medium storing program for executing.
Computer program instructions for executing disclosure operation can be assembly instruction, instruction set architecture (ISA) instructs,
Machine instruction, machine-dependent instructions, microcode, firmware instructions, condition setup data or with one or more programming languages
The source code or object code that any combination is write, the programming language include the programming language-of object-oriented such as
Smalltalk, C++ etc., and conventional procedural programming languages-such as " C " language or similar programming language.Computer
Readable program instructions can be executed fully on the user computer, partly execute on the user computer, be only as one
Vertical software package executes, part executes on the remote computer or completely in remote computer on the user computer for part
Or it is executed on server.In situations involving remote computers, remote computer can pass through network-packet of any kind
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).In some embodiments, by utilizing computer-readable program instructions
Status information carry out personalized customization electronic circuit, such as programmable logic circuit, field programmable gate array (FPGA) or can
Programmed logic array (PLA) (PLA), the electronic circuit can execute computer-readable program instructions, to realize each side of the disclosure
Face.
Referring herein to according to the flow chart of the method, apparatus (system) of the embodiment of the present disclosure and computer program product and/
Or block diagram describes various aspects of the disclosure.It should be appreciated that flowchart and or block diagram each box and flow chart and/
Or in block diagram each box combination, can be realized by computer-readable program instructions.
These computer-readable program instructions can be supplied to general purpose computer, special purpose computer or other programmable datas
The processor of processing unit, so that a kind of machine is produced, so that these instructions are passing through computer or other programmable datas
When the processor of processing unit executes, function specified in one or more boxes in implementation flow chart and/or block diagram is produced
The device of energy/movement.These computer-readable program instructions can also be stored in a computer-readable storage medium, these refer to
It enables so that computer, programmable data processing unit and/or other equipment work in a specific way, thus, it is stored with instruction
Computer-readable medium then includes a manufacture comprising in one or more boxes in implementation flow chart and/or block diagram
The instruction of the various aspects of defined function action.
Computer-readable program instructions can also be loaded into computer, other programmable data processing units or other
In equipment, so that series of operation steps are executed in computer, other programmable data processing units or other equipment, to produce
Raw computer implemented process, so that executed in computer, other programmable data processing units or other equipment
Instruct function action specified in one or more boxes in implementation flow chart and/or block diagram.
The flow chart and block diagram in the drawings show system, method and the computer journeys according to multiple embodiments of the disclosure
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
One module of table, program segment or a part of instruction, the module, program segment or a part of instruction include one or more use
The executable instruction of the logic function as defined in realizing.In some implementations as replacements, function marked in the box
It can occur in a different order than that indicated in the drawings.For example, two continuous boxes can actually be held substantially in parallel
Row, they can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that block diagram and/or
The combination of each box in flow chart and the box in block diagram and or flow chart, can the function as defined in executing or dynamic
The dedicated hardware based system made is realized, or can be realized using a combination of dedicated hardware and computer instructions.
The presently disclosed embodiments is described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes are obvious for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or technological improvement to technology in market for best explaining each embodiment, or make the art
Other those of ordinary skill can understand each embodiment disclosed herein.
Claims (10)
1. a kind of image partition method, which is characterized in that the described method includes:
According to the regional location of image where object to be split, image segmentation is carried out to described image, obtains image segmentation result；
Exist in described image segmentation result and obtains the first interactive operation in wrong subregional situation；
First interactive operation is responded, is obtained for the subregional annotation results of mistake, according to the annotation results to institute
It states image segmentation result and carries out correction processing.
2. the method according to claim 1, wherein described image be 2D image in the case where, the basis to
The regional location of image where cutting object carries out image segmentation to described image, obtains image segmentation result, comprising:
The extracting parameter of the corresponding regional location is determined according to the second interactive operation；
Rectangle frame is obtained according to the extracting parameter, image segmentation is carried out to the 2D image according to the rectangle frame, obtains 2D
Image segmentation result.
3. according to the method described in claim 2, it is characterized in that, described carry out figure to the 2D image according to the rectangle frame
As segmentation, 2D image segmentation result is obtained, comprising:
2D image segmentation network is inputted by the 2D image and according to the rectangular area that the rectangle frame obtains, output obtains described
2D image segmentation result.
4. method according to claim 1-3, which is characterized in that described to exist in described image segmentation result
The first interactive operation is obtained in wrong subregional situation, comprising:
The mistake subregion is that the target object mistake as prospect is divided into background, obtains first interactive operation；
Response first interactive operation, obtains for the subregional annotation results of mistake, comprising:
First interactive operation is parsed according to preset operation description information, it is corresponding to obtain first interactive operation
First annotation results of description, first annotation results are identified with the first mark information, are believed by first label
Breath characterization is described to be divided into background for the target object mistake as prospect.
5. according to the method described in claim 4, it is characterized in that, first interactive operation includes: the input of left mouse button
Operation, the first specified touch control operation, any one in the first specified slide.
6. according to the method described in claim 4, it is characterized in that, according to the annotation results to described image segmentation result into
Row correction processing, comprising:
2D image, 2D image segmentation result and the first annotation results are inputted into 2D image rectification network, output obtains 2D image point
Cut result.
7. method according to claim 1-3, which is characterized in that described to exist in described image segmentation result
The first interactive operation is obtained in wrong subregional situation, comprising:
The mistake subregion is that the part mistake for belonging to background is divided into the target object as prospect, obtains the first interaction behaviour
Make；
Response first interactive operation, obtains for the subregional annotation results of mistake, comprising:
First interactive operation is parsed according to preset operation description information, it is corresponding to obtain first interactive operation
Second annotation results of description, second annotation results are identified with the second mark information, are believed by second label
Breath characterizes the part mistake that will belong to background and is divided into the target object as prospect.
8. a kind of image segmentation device, which is characterized in that described device includes:
Divide module, for the regional location according to image where object to be split, image segmentation is carried out to described image, is obtained
Image segmentation result；
Operation obtains module, and the first interaction behaviour is obtained in wrong subregional situation for existing in described image segmentation result
Make；
Respond module is operated, for responding first interactive operation, is obtained for the subregional annotation results of mistake, according to
The annotation results carry out correction processing to described image segmentation result.
9. a kind of electronic equipment characterized by comprising
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to: perform claim require any one of 1 to 7 described in method.
10. a kind of computer readable storage medium, is stored thereon with computer program instructions, which is characterized in that the computer
Method described in any one of claim 1 to 7 is realized when program instruction is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910464349.3A CN110211134B (en) | 2019-05-30 | 2019-05-30 | Image segmentation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910464349.3A CN110211134B (en) | 2019-05-30 | 2019-05-30 | Image segmentation method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110211134A true CN110211134A (en) | 2019-09-06 |
| CN110211134B CN110211134B (en) | 2021-11-05 |
Family
ID=67789802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910464349.3A Active CN110211134B (en) | 2019-05-30 | 2019-05-30 | Image segmentation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110211134B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751659A (en) * | 2019-09-27 | 2020-02-04 | 北京小米移动软件有限公司 | Image segmentation method and device, terminal and storage medium |
| CN111259184A (en) * | 2020-02-27 | 2020-06-09 | 厦门大学 | Image automatic labeling system and method for new retail |
| CN112070763A (en) * | 2020-09-18 | 2020-12-11 | 上海商汤智能科技有限公司 | Image data processing method and device, electronic equipment and storage medium |
| CN112418205A (en) * | 2020-11-19 | 2021-02-26 | 上海交通大学 | Interactive image segmentation method and system based on focusing on wrongly segmented areas |
| CN112925938A (en) * | 2021-01-28 | 2021-06-08 | 上海商汤智能科技有限公司 | Image annotation method and device, electronic equipment and storage medium |
| CN113313700A (en) * | 2021-06-09 | 2021-08-27 | 浙江大学 | X-ray image interactive segmentation method based on deep learning |
| CN113837194A (en) * | 2021-09-23 | 2021-12-24 | 北京百度网讯科技有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103295219A (en) * | 2012-03-02 | 2013-09-11 | 北京数码视讯科技股份有限公司 | Method and device for segmenting image |
| CN103559719A (en) * | 2013-11-20 | 2014-02-05 | 电子科技大学 | Interactive graph cutting method |
| CN103971415A (en) * | 2014-05-23 | 2014-08-06 | 南京大学 | Online marking method for three-dimensional model component |
| CN104899851A (en) * | 2014-03-03 | 2015-09-09 | 天津医科大学 | Lung nodule image segmentation method |
| CN106780536A (en) * | 2017-01-13 | 2017-05-31 | 深圳市唯特视科技有限公司 | A kind of shape based on object mask network perceives example dividing method |
| CN109087327A (en) * | 2018-07-13 | 2018-12-25 | 天津大学 | A kind of thyroid nodule ultrasonic image division method cascading full convolutional neural networks |
| CN109255791A (en) * | 2018-07-19 | 2019-01-22 | 杭州电子科技大学 | A kind of shape collaboration dividing method based on figure convolutional neural networks |
| CN109598728A (en) * | 2018-11-30 | 2019-04-09 | 腾讯科技（深圳）有限公司 | Image partition method, device, diagnostic system and storage medium |
| CN109727251A (en) * | 2018-12-29 | 2019-05-07 | 上海联影智能医疗科技有限公司 | The system that lung conditions are divided a kind of quantitatively, method and apparatus |
| CN109801260A (en) * | 2018-12-20 | 2019-05-24 | 北京海益同展信息科技有限公司 | The recognition methods of livestock number and device |
-
2019
- 2019-05-30 CN CN201910464349.3A patent/CN110211134B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103295219A (en) * | 2012-03-02 | 2013-09-11 | 北京数码视讯科技股份有限公司 | Method and device for segmenting image |
| CN103559719A (en) * | 2013-11-20 | 2014-02-05 | 电子科技大学 | Interactive graph cutting method |
| CN104899851A (en) * | 2014-03-03 | 2015-09-09 | 天津医科大学 | Lung nodule image segmentation method |
| CN103971415A (en) * | 2014-05-23 | 2014-08-06 | 南京大学 | Online marking method for three-dimensional model component |
| CN106780536A (en) * | 2017-01-13 | 2017-05-31 | 深圳市唯特视科技有限公司 | A kind of shape based on object mask network perceives example dividing method |
| CN109087327A (en) * | 2018-07-13 | 2018-12-25 | 天津大学 | A kind of thyroid nodule ultrasonic image division method cascading full convolutional neural networks |
| CN109255791A (en) * | 2018-07-19 | 2019-01-22 | 杭州电子科技大学 | A kind of shape collaboration dividing method based on figure convolutional neural networks |
| CN109598728A (en) * | 2018-11-30 | 2019-04-09 | 腾讯科技（深圳）有限公司 | Image partition method, device, diagnostic system and storage medium |
| CN109801260A (en) * | 2018-12-20 | 2019-05-24 | 北京海益同展信息科技有限公司 | The recognition methods of livestock number and device |
| CN109727251A (en) * | 2018-12-29 | 2019-05-07 | 上海联影智能医疗科技有限公司 | The system that lung conditions are divided a kind of quantitatively, method and apparatus |
Non-Patent Citations (2)
| Title |
|---|
| GUOTAI WANG , MARIA A. ZULUAGA ET AL.: ""DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentation"", 《IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 * |
| OLAF RONNEBERGER, PHILIPP FISCHER, AND THOMAS BROX: ""U-Net: Convolutional Networks for Biomedical Image Segmentation"", 《ARXIV》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751659A (en) * | 2019-09-27 | 2020-02-04 | 北京小米移动软件有限公司 | Image segmentation method and device, terminal and storage medium |
| CN110751659B (en) * | 2019-09-27 | 2022-06-10 | 北京小米移动软件有限公司 | Image segmentation method and device, terminal and storage medium |
| CN111259184A (en) * | 2020-02-27 | 2020-06-09 | 厦门大学 | Image automatic labeling system and method for new retail |
| CN111259184B (en) * | 2020-02-27 | 2022-03-08 | 厦门大学 | Image automatic labeling system and method for new retail |
| CN112070763A (en) * | 2020-09-18 | 2020-12-11 | 上海商汤智能科技有限公司 | Image data processing method and device, electronic equipment and storage medium |
| CN112418205A (en) * | 2020-11-19 | 2021-02-26 | 上海交通大学 | Interactive image segmentation method and system based on focusing on wrongly segmented areas |
| CN112925938A (en) * | 2021-01-28 | 2021-06-08 | 上海商汤智能科技有限公司 | Image annotation method and device, electronic equipment and storage medium |
| CN113313700A (en) * | 2021-06-09 | 2021-08-27 | 浙江大学 | X-ray image interactive segmentation method based on deep learning |
| CN113837194A (en) * | 2021-09-23 | 2021-12-24 | 北京百度网讯科技有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
| CN113837194B (en) * | 2021-09-23 | 2024-05-10 | 北京百度网讯科技有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110211134B (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110211134A (en) | A kind of image partition method and device, electronic equipment and storage medium | |
| CN110210487A (en) | A kind of image partition method and device, electronic equipment and storage medium | |
| TWI713054B (en) | Image segmentation method and device, electronic equipment and storage medium | |
| TWI755175B (en) | Image segmentation method, electronic device and storage medium | |
| US20210158533A1 (en) | Image processing method and apparatus, and storage medium | |
| US20220180521A1 (en) | Image processing method and apparatus, and electronic device, storage medium and computer program | |
| US9733713B2 (en) | Laser beam based gesture control interface for mobile devices | |
| RU2708027C1 (en) | Method of transmitting motion of a subject from a video to an animated character | |
| TW202044113A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN107533372A (en) | The hybird environment of attachment data is shown | |
| CN112541928A (en) | Network training method and device, image segmentation method and device and electronic equipment | |
| WO2021259391A2 (en) | Image processing method and apparatus, and electronic device and storage medium | |
| TWI765404B (en) | Interactive display method for image positioning, electronic device and computer-readable storage medium | |
| CN112115894B (en) | Training method and device of hand key point detection model and electronic equipment | |
| CN106803777A (en) | Method for information display and device | |
| KR20220149503A (en) | Image capturing method and apparatus, electronic device and computer readable storage medium | |
| CN110019960A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN109543536A (en) | Image identification method and device, electronic equipment and storage medium | |
| CN110532956A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN110648762A (en) | Method and device for generating lesion area identification model and method and device for identifying lesion area | |
| CN111597922A (en) | Cell image recognition method, system, device, equipment and medium | |
| CN106980880A (en) | The method and device of images match | |
| CN115170464A (en) | Lung image processing method and device, electronic equipment and storage medium | |
| CN114343687A (en) | Adrenal gland functional tumor typing positioning equipment, device and storage medium | |
| CN106060707A (en) | Reverberation processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |