CN111709436A - Marking method and system, and classification method and system for medical image contour - Google Patents
Marking method and system, and classification method and system for medical image contour Download PDFInfo
- Publication number
- CN111709436A CN111709436A CN202010434421.0A CN202010434421A CN111709436A CN 111709436 A CN111709436 A CN 111709436A CN 202010434421 A CN202010434421 A CN 202010434421A CN 111709436 A CN111709436 A CN 111709436A
- Authority
- CN
- China
- Prior art keywords
- contour
- user
- acquiring
- image
- segment
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000012549 training Methods 0.000 claims description 20
- 238000013136 deep learning model Methods 0.000 claims description 17
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000013135 deep learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 210000000623 ulna Anatomy 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of computers, in particular to a marking method and system and a classification method and system for medical image contours. A marking method of medical image contour comprises the following steps: acquiring the outline of a region of interest marked on a medical image by a user; acquiring a category coding value of the contour specified by a user; and filling contour pixels according to the category code values of the contours, generating a label image and storing the label image. The invention has the beneficial effects that: marking the outline of the region of interest on the medical image by acquiring a user; acquiring a category coding value of the contour specified by a user; and filling contour pixels according to the category coding values of the contours, generating a label image and storing the label image so as to realize the rapid marking of the contours on the medical images.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a marking method and system and a classification method and system for medical image contours.
Background
The deep learning of the medical image needs to be carried out by calculating the cross entropy error between the original image and the artificial marked image. At present, the field of medical image deep learning is lack of a mature method for carrying out regional contour marking and category calibration on tissues or organs of interest in medical images, in particular to marking and calibration of multi-classification and repeated classification.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method and a system for labeling a medical image contour, and a method and a system for classifying the medical image contour.
In a first aspect, the present invention provides a method for marking a contour of a medical image, including:
acquiring the outline of a region of interest marked on a medical image by a user;
acquiring a category coding value of the contour specified by a user;
and filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
Preferably, the acquiring the outline of the region of interest marked by the user on the medical image includes:
and acquiring coordinate information of the dragging event in real time, converting the coordinate information into a coordinate in the original image through a coordinate system, and connecting adjacent coordinates to obtain the outline of the region of interest.
Preferably, the tag images are stored in order using a queue data structure.
Preferably, the category code values are stored in the order of the corresponding label images using a queue data structure.
Preferably, when the contour needs to be modified, the new contour segment is drawn, two points of the modified contour, which are respectively closest to the head and the tail of the new contour segment, are obtained, the modified contour is divided into two replaced contour segments through the two points, the new contour segment is replaced with one replaced contour segment, which has a smaller area formed by the new contour segment, and the new contour formed by the new contour segment and the other replaced contour segment is stored.
In a second aspect, the present invention provides a method for classifying medical images, including:
acquiring the outline of a region of interest marked on a medical image by a user;
acquiring a category coding value of the contour specified by a user;
filling contour pixels according to the category coding values of the contours, generating label images and storing the label images;
training by taking the label image as a training sample to obtain a deep learning model;
and classifying the medical images to be classified according to the trained deep learning model.
Preferably, the acquiring the outline of the region of interest marked by the user on the medical image includes:
and acquiring coordinate information of the dragging event in real time, converting the coordinate information into a coordinate in the original image through a coordinate system, and connecting adjacent coordinates to obtain the outline of the region of interest.
Preferably, when the contour needs to be modified, the new contour segment is drawn, two points of the modified contour, which are respectively closest to the head and the tail of the new contour segment, are obtained, the modified contour is divided into two replaced contour segments through the two points, the new contour segment is replaced with one replaced contour segment, which has a smaller area formed by the new contour segment, and the new contour formed by the new contour segment and the other replaced contour segment is stored.
In a third aspect, the present invention provides a system for marking a contour of a medical image, comprising:
the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user;
the category code value acquisition module is used for acquiring a category code value of the contour specified by a user;
and the contour filling module is used for filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
In a fourth aspect, the present invention provides a classification system for medical images, comprising:
the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user;
the category code value acquisition module is used for acquiring a category code value of the contour specified by a user;
the contour filling module is used for performing category pixel filling according to the contour, generating a label image and storing the label image;
the sample training module is used for training by taking the label image as a training sample to obtain a deep learning model;
and the sample classification module is used for classifying the medical images to be classified according to the trained deep learning model.
The invention has the beneficial effects that:
1. marking the outline of the region of interest on the medical image by acquiring a user; acquiring a category coding value of the contour specified by a user; filling contour pixels according to the category coding values of the contours, generating label images and storing the label images so as to realize rapid marking of the contours on the medical images;
2. marking the outline of the region of interest on the medical image by acquiring a user; acquiring a category coding value of the contour specified by a user; filling contour pixels according to the category coding values of the contours, generating label images and storing the label images; training by taking the label image as a training sample to obtain a deep learning model; and classifying the medical images to be classified according to the trained deep learning model so as to realize the rapid and accurate classification of the medical images.
Drawings
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
FIG. 1 is a schematic flow chart of steps S101-S103 according to a first embodiment of the present invention;
FIG. 2 is a flowchart illustrating step S104 according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a system according to an embodiment of the present invention;
FIG. 4 is a flow chart illustrating a first operation of the system according to one embodiment of the present invention;
FIG. 5 is a flow chart illustrating a second operation of the system according to one embodiment of the present invention;
FIG. 6 is a schematic flow chart of a second embodiment of the present invention;
fig. 7 is a schematic structural diagram of a system according to a second embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be further described below with reference to the accompanying drawings, but the present invention is not limited to these embodiments.
Example one
The basic idea of the invention is to realize visual human-computer interaction operation through software programming, so that a user can mark the outline of an interested region on a medical image, assign all pixels in the outline to the coding value of a target class and finally store the coding value as a group of label images for deep learning.
Based on the above basic ideas, an embodiment of the present invention provides a method for marking a medical image contour, as shown in fig. 1, including the following steps:
s101: and acquiring the outline of the region of interest marked on the medical image by the user.
The medical image in this embodiment may be an 8-bit or 16-bit depth medical image
The contour is a curve which is formed by a series of coordinate points (x, y) and contains a target shape and is closed in a two-dimensional space, wherein x represents the coordinate in the x-axis direction, y represents the coordinate in the y-axis direction, and the curve is stored by using a data structure similar to a linked list.
Firstly, coordinate information of a dragging event is acquired in real time, wherein the dragging event can be operated by a mouse, and can also be operated by a finger aiming at a touch control input device. And then converting the coordinate system into the coordinate in the original image, and connecting adjacent coordinates to obtain the outline of the region of interest.
S102: and acquiring a category code value of the outline specified by the user.
The user classifies the drawn outline, for example, the ulna is defined by a category code value of 1, the radius is defined by a category code value of 2, and so on.
S103: and filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
The outlines corresponding to different category code values are filled by different pixels for distinction and for observation.
When the label images are stored, the queue data structure is used for storing the label images in sequence, and the queue data structure is used for storing the category code values in the sequence corresponding to the label images, so that the management, the search and the like of the label images are facilitated.
In one embodiment, as shown in fig. 2, the method further comprises the steps of:
s104: when the contour needs to be modified, drawing a new contour segment and obtaining two points of the modified contour, which are respectively closest to the head and the tail of the new contour segment, dividing the modified contour into two replaced contour segments through the two points, replacing one replaced contour segment with the new contour segment, which has a smaller area formed by the new contour segment, and storing the new contour formed by the new contour segment and the other replaced contour segment.
When the contour does not correspond to the actual image, the contour needs to be modified. And acquiring the track of the new contour segment according to the dragging event so as to determine the head and the tail of the new contour segment. Under normal conditions, the head and the tail of the new contour segment are on the modified contour, so that the head and the tail of the new contour segment, namely two points of the modified contour which are respectively closest to the head and the tail of the new contour segment, replace the new contour segment with a replaced contour segment which has a smaller area with the new contour segment, and save the new contour formed by the new contour segment and the other replaced contour segment. In some cases, the new contour segment is not intersected with the modified contour, the head and tail two points of the new contour segment and the two points of the modified contour which are respectively closest to the head and tail two points of the new contour segment are connected, the new contour segment replaces one replaced contour segment which has a smaller area formed by the new contour segment, and the new contour segment and the new contour formed by the other replaced contour segment are stored.
The embodiment provides a method for marking a medical image contour, and correspondingly, further provides a system for marking a medical image contour in terms of hardware, as shown in fig. 3, including: the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user; the category code value acquisition module is used for acquiring a category code value of the contour specified by a user; and the contour filling module is used for filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
It should be noted that, the contour acquisition module, the category code value acquisition module, and the contour filling module in this embodiment may all be developed based on java language through cross-platform application software to implement marking of the medical image contour on a computer, and a user may input the medical image contour through a mouse.
The operation flow chart of the system is shown in fig. 4, when a user calls a menu by using a right button, clicks and selects a 'mark outline' and presses a left button of a mouse to start outline drawing, the user drags the mouse, software acquires coordinate information in real time by capturing a mouse dragging event, the coordinate information is converted into coordinates in an original image through a coordinate system, coordinate points are recorded into an outline linked list (a one-way circular linked list), when the user marks a required outline, the left button of the mouse is bounced, a dialog box is popped up to require the user to designate a category number, and when the user designates a category number, the software queues the outline and the category, inquires whether the user marks the next outline, and carries out the outline drawing again according to the selection of the user. And exiting the modification mode after the user right clicks. When a user right clicks a call-out menu and clicks 'tag making', the software fills the class pixels according to the new outline, and then generates and stores a tag image.
The operation flow chart of the system contour modification is shown in fig. 5, a user calls a menu with a right button and clicks 'modify appointed contour', the software pops up a contour selection dialog box, the user moves a mouse to be modified to be close to a target contour after selecting the contour to be modified, the software changes the mouse style and indicates that the user can modify, when the user presses a left mouse button and drags the mouse, the software acquires coordinate information in real time by capturing a mouse dragging event, the coordinate information is converted into coordinates in an original image through a coordinate system, the coordinate points are recorded in a modified linked list (another one-way circular linked list), when the user finishes modifying the mouse and pops up, the software searches for the nearest distance point from the modified contour to the head coordinate point and the tail coordinate point of the modified linked list, and inserts the modified linked list between the two points to form a new contour and refreshes the modified linked list.
Example two
On the basis of the first embodiment, a second embodiment of the present invention provides a method for classifying medical images, which is characterized in that, as shown in fig. 6, the method includes the following steps:
s201: acquiring the outline of a region of interest marked on a medical image by a user;
s202: acquiring a category coding value of the contour specified by a user;
s203: filling contour pixels according to the category coding values of the contours, generating label images and storing the label images;
s204: training by taking the label image as a training sample to obtain a deep learning model;
s205: and classifying the medical images to be classified according to the trained deep learning model.
Since steps S201 to S203 have already been described in detail in the first embodiment, they are not described again in this embodiment.
Based on the label images obtained in steps S201 to S203, a part of the label images is used as a training sample, another part is used as a verification sample, and the deep learning model after final training is a verified model. The medical images to be classified are quickly and accurately classified according to the trained deep learning model, and the working efficiency is improved.
The present embodiment provides a method for classifying medical images, and correspondingly, a system for classifying medical images is further provided in terms of hardware, as shown in fig. 7, including: the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user; the category code value acquisition module is used for acquiring a category code value of the contour specified by a user; the contour filling module is used for performing category pixel filling according to the contour, generating a label image and storing the label image; the sample training module is used for training by taking the label image as a training sample to obtain a deep learning model; and the sample classification module is used for classifying the medical images to be classified according to the trained deep learning model.
It should be noted that, in this embodiment, the contour obtaining module, the category code value obtaining module, the contour filling module, the sample training module, and the sample classification module may all perform cross-platform application software development based on java language to implement marking and classifying of medical image contours on a computer, and a user may input the medical image contours through a mouse.
In the man-machine interaction operation, the classification operation can utilize a trained deep learning model to perform one-key intelligent contour presetting, namely only clicking an 'AI preset contour' menu item, software can perform contour marking and category calibration on an unlabelled medical image according to the model to generate a contour queue and a category queue preset by the model, so that a user can add contours, delete contours, modify contours and check contours, and can realize multi-classification and repeated classification, namely, the contour queue and the corresponding classification queue can have multiple classifications, particularly different contours of the same classification can exist, so that the classification calibration of a discontinuous region of the same tissue in the medical image is facilitated.
Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (10)
1. A method for marking a contour of a medical image, comprising:
acquiring the outline of a region of interest marked on a medical image by a user;
acquiring a category coding value of the contour specified by a user;
and filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
2. The method for marking the contour of the medical image as claimed in claim 1, wherein the obtaining the contour of the region of interest marked on the medical image by the user comprises:
and acquiring coordinate information of the dragging event in real time, converting the coordinate information into a coordinate in the original image through a coordinate system, and connecting adjacent coordinates to obtain the outline of the region of interest.
3. The method as claimed in claim 1, wherein the tag images are stored in sequence using a queue data structure.
4. A method as claimed in claim 1, wherein the category code values are stored in the order of the corresponding label images using a queue data structure.
5. The method as claimed in any one of claims 1 to 4, wherein when the contour needs to be modified, the new contour segment is drawn and two points of the modified contour closest to the head and the tail of the new contour segment are obtained, the modified contour is divided into two replaced contour segments by the two points, the new contour segment is replaced with a replaced contour segment having a smaller area with the new contour segment, and the new contour formed by the new contour segment and the other replaced contour segment is saved.
6. A method for classifying a medical image, comprising:
acquiring the outline of a region of interest marked on a medical image by a user;
acquiring a category coding value of the contour specified by a user;
filling contour pixels according to the category coding values of the contours, generating label images and storing the label images;
training by taking the label image as a training sample to obtain a deep learning model;
and classifying the medical images to be classified according to the trained deep learning model.
7. The method for classifying medical images according to claim 6, wherein the step of obtaining the outline of the region of interest marked on the medical image by the user comprises:
and acquiring coordinate information of the dragging event in real time, converting the coordinate information into a coordinate in the original image through a coordinate system, and connecting adjacent coordinates to obtain the outline of the region of interest.
8. The method as claimed in any one of claims 6 to 7, wherein when the contour needs to be modified, the new contour segment is drawn and two points of the modified contour, which are respectively closest to the head and the tail of the new contour segment, are obtained, the modified contour is divided into two replaced contour segments by the two points, the new contour segment is replaced with one replaced contour segment having a smaller area with the new contour segment, and the new contour segment and the other replaced contour segment are saved to form the new contour.
9. A system for marking contours of medical images, comprising:
the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user;
the category code value acquisition module is used for acquiring a category code value of the contour specified by a user;
and the contour filling module is used for filling contour pixels according to the category code values of the contours, generating a label image and storing the label image.
10. A system for classifying medical images, comprising:
the contour acquisition module is used for acquiring the contour of the region of interest marked on the medical image by the user;
the category code value acquisition module is used for acquiring a category code value of the contour specified by a user;
the contour filling module is used for performing category pixel filling according to the contour, generating a label image and storing the label image;
the sample training module is used for training by taking the label image as a training sample to obtain a deep learning model;
and the sample classification module is used for classifying the medical images to be classified according to the trained deep learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010434421.0A CN111709436A (en) | 2020-05-21 | 2020-05-21 | Marking method and system, and classification method and system for medical image contour |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010434421.0A CN111709436A (en) | 2020-05-21 | 2020-05-21 | Marking method and system, and classification method and system for medical image contour |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111709436A true CN111709436A (en) | 2020-09-25 |
Family
ID=72537854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010434421.0A Pending CN111709436A (en) | 2020-05-21 | 2020-05-21 | Marking method and system, and classification method and system for medical image contour |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111709436A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113920088A (en) * | 2021-10-11 | 2022-01-11 | 浙江康源医疗器械有限公司 | Deep learning-based radius density detection method, system and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359208A (en) * | 2018-09-13 | 2019-02-19 | 郑津 | A kind of distributed method and system of precisely lossless mark image instance |
| CN110032315A (en) * | 2019-03-28 | 2019-07-19 | 广州柏视数据科技有限公司 | A kind of target of prophylactic radiotherapy contour line quick modification method |
| CN110264443A (en) * | 2019-05-20 | 2019-09-20 | 平安科技（深圳）有限公司 | Eye fundus image lesion mask method, device and medium based on feature visualization |
| CN110415248A (en) * | 2019-08-29 | 2019-11-05 | 新名医(北京)科技有限公司 | A kind of blood vessel monitoring method, device, equipment and storage medium based on ultrasound |
| CN110599508A (en) * | 2019-08-01 | 2019-12-20 | 平安科技（深圳）有限公司 | Spine image processing method based on artificial intelligence and related equipment |
| CN110675940A (en) * | 2019-08-01 | 2020-01-10 | 平安科技（深圳）有限公司 | Pathological image labeling method and device, computer equipment and storage medium |
| CN110751183A (en) * | 2019-09-24 | 2020-02-04 | 东软集团股份有限公司 | Image data classification model generation method, image data classification method and device |
| CN111127404A (en) * | 2019-12-06 | 2020-05-08 | 广州柏视医疗科技有限公司 | Medical image contour rapid extraction method |
-
2020
- 2020-05-21 CN CN202010434421.0A patent/CN111709436A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359208A (en) * | 2018-09-13 | 2019-02-19 | 郑津 | A kind of distributed method and system of precisely lossless mark image instance |
| CN110032315A (en) * | 2019-03-28 | 2019-07-19 | 广州柏视数据科技有限公司 | A kind of target of prophylactic radiotherapy contour line quick modification method |
| CN110264443A (en) * | 2019-05-20 | 2019-09-20 | 平安科技（深圳）有限公司 | Eye fundus image lesion mask method, device and medium based on feature visualization |
| CN110599508A (en) * | 2019-08-01 | 2019-12-20 | 平安科技（深圳）有限公司 | Spine image processing method based on artificial intelligence and related equipment |
| CN110675940A (en) * | 2019-08-01 | 2020-01-10 | 平安科技（深圳）有限公司 | Pathological image labeling method and device, computer equipment and storage medium |
| CN110415248A (en) * | 2019-08-29 | 2019-11-05 | 新名医(北京)科技有限公司 | A kind of blood vessel monitoring method, device, equipment and storage medium based on ultrasound |
| CN110751183A (en) * | 2019-09-24 | 2020-02-04 | 东软集团股份有限公司 | Image data classification model generation method, image data classification method and device |
| CN111127404A (en) * | 2019-12-06 | 2020-05-08 | 广州柏视医疗科技有限公司 | Medical image contour rapid extraction method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113920088A (en) * | 2021-10-11 | 2022-01-11 | 浙江康源医疗器械有限公司 | Deep learning-based radius density detection method, system and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11797847B2 (en) | Selecting instances of detected objects in images utilizing object detection models | |
| US12020414B2 (en) | Utilizing deep neural networks to automatically select instances of detected objects in images | |
| US11972569B2 (en) | Segmenting objects in digital images utilizing a multi-object segmentation model framework | |
| CN110349652B (en) | Medical data analysis system fusing structured image data | |
| US11436436B2 (en) | Data augmentation system, data augmentation method, and information storage medium | |
| JP5437934B2 (en) | System for artificially labeling pixels in an image with artificial assistance and ground true labeler user interface display | |
| CN111242691A (en) | Method and device for generating advertisement poster, storage medium and terminal equipment | |
| GB2586678A (en) | Utilizing multiple object detection models to automatically select user-requested objects in images | |
| CN109063720A (en) | Handwritten word training sample acquisition methods, device, computer equipment and storage medium | |
| CN114730241A (en) | Gesture stroke recognition in touch user interface input | |
| CN111709436A (en) | Marking method and system, and classification method and system for medical image contour | |
| CN109858402B (en) | Image detection method, device, terminal and storage medium | |
| CN116610304B (en) | Page code generation method, device, equipment and storage medium | |
| CN108681731A (en) | A kind of thyroid cancer ultrasound picture automatic marking method and system | |
| JPH1166283A (en) | Correlation extracting method for image feature quantity and image retrieval device | |
| Pohudina et al. | Method for identifying and counting objects | |
| US20230071291A1 (en) | System and method for a precise semantic segmentation | |
| US11893313B2 (en) | Interactive object selection | |
| Boroujerdi et al. | Deep interactive region segmentation and captioning | |
| Beuvens et al. | UsiGesture: An environment for integrating pen-based interaction in user interface development | |
| CN113987170A (en) | Multi-label text classification method based on convolutional neural network | |
| CN111968111A (en) | Method and device for identifying visceral organs or artifacts of CT (computed tomography) image | |
| Koenig et al. | NEURAL-UML: Intelligent Recognition System of Structural Elements in UML Class Diagram | |
| Kazangirler et al. | UIBee: An improved deep instance segmentation and classification of UI elements in wireframes | |
| CN111553894A (en) | Medical image segmentation model training method, medium and electronic 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200925 |