CN111611945A - Universal AutoCAD picture frame identification method - Google Patents
Universal AutoCAD picture frame identification method Download PDFInfo
- Publication number
- CN111611945A CN111611945A CN202010448512.XA CN202010448512A CN111611945A CN 111611945 A CN111611945 A CN 111611945A CN 202010448512 A CN202010448512 A CN 202010448512A CN 111611945 A CN111611945 A CN 111611945A
- Authority
- CN
- China
- Prior art keywords
- closed
- straight lines
- vertical
- entities
- rectangles
- 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 19
- 238000012216 screening Methods 0.000 claims abstract description 15
- 238000005191 phase separation Methods 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 238000007429 general method Methods 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 229930091051 Arenine Natural products 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001915 proofreading effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/42—Document-oriented image-based pattern recognition based on the type of document
- G06V30/422—Technical drawings; Geographical maps
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Character Input (AREA)
Abstract
The invention discloses a universal AutoCAD picture frame identification method, which is characterized by traversing the current space of a DWG drawing, screening out all linear entities and character entities, and acquiring all horizontal lines, vertical lines and characters from the linear entities and the character entities; acquiring a vertical intersection position of a horizontal straight line and a vertical straight line; finding out all closed rectangles in the drawing according to the vertical intersection position; screening out all closed rectangles which contain the keywords and have only phase-containing or phase-separating relations with other rectangles; and comprehensively judging and identifying the picture frame according to the relation among the closed rectangles, the area ratio of the inner and outer layer closed rectangles and the number and ordered arrangement of the keywords in the closed rectangles. The method can identify the picture frame more accurately through the keywords and other picture frame key elements, realizes the identification of the general picture frame with different design habits, and has more universality and universality.
Description
Technical Field
The invention relates to a graphic processing technology in the field of computer software, in particular to a universal automatic CAD drawing frame identification method.
Background
The current DWG drawing identification method mostly depends on some conditions preset manually by a user, such as a drawing layer where a drawing frame is located, whether the drawing frame is a block, a block name of the drawing frame, a line composition type of the drawing frame, and the like.
The existing picture frame identification methods are realized mostly based on some technologies of analyzing whether a closed rectangle exists in a DWG drawing and whether the side length of the closed rectangle is consistent with the size of a standard picture frame. The recognition result is that the sketch drawn by the designer is recognized as a frame only by sleeving a rectangle. In this case the identified frame is often a rectangle, not a true frame. Still others are identified on a frame-by-block basis, which requires that frames must be made and named to meet certain specifications.
The invention can more accurately identify the real picture frame based on the picture frame character characteristics and the picture frame composition characteristics, and further conditionally acquire some information of the drawing labels and the drawing areas in the picture frame. Further, the functions of obtaining the picture width of the picture frame, the drawing proportion, the lengthening specification, the position of the drawing label and the like can be realized.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a universal automatic CAD drawing frame identification method, which can identify the drawing frame in the drawing without preprocessing DWG drawing, is suitable for various drawings and improves the drawing processing efficiency.
The invention adopts the following technical scheme. A general method for identifying an AutoCAD drawing frame is characterized by comprising the following steps:
1) traversing the current space of the DWG drawing, screening out all straight line entities and character entities, and acquiring all horizontal straight lines, vertical straight lines and characters from the entities: firstly, acquiring common entities forming a picture frame in all block entities, linear entities, two-dimensional multi-line-segment entities, three-dimensional multi-line-segment entities, single-line character entities, multi-line character entities and attribute character entities; then recursively acquiring the common entities in the block entities and attribute character contents in the attribute block entities; extracting the starting point coordinates and the end point coordinates of the straight line entity, and eliminating non-horizontal and non-vertical straight lines;
2) acquiring the vertical intersection position of the horizontal straight line and the vertical straight line: firstly, sequencing all horizontal straight lines and all vertical straight lines respectively; then, taking a horizontal straight line, carrying out vertical intersection position judgment on the horizontal straight line and a vertical straight line which is possibly intersected, and sequentially comparing the rest straight lines according to the difference of the vertical intersection positions so as to obtain the corresponding vertical intersection positions of all the horizontal straight lines and the vertical straight lines;
3) finding out all closed rectangles in the drawing according to the vertical intersection position: firstly, screening all straight lines with vertical intersection positions of upper left intersection and straight lines with vertical intersection positions of lower right intersection; then, two straight lines with the vertical intersection position as the upper left and two straight lines with the vertical intersection position as the lower right are extracted for judgment, if the lengths of the two horizontal straight lines are equal and the lengths of the two vertical straight lines are equal, the two straight lines with the vertical intersection position as the upper left and the two straight lines with the vertical intersection position as the lower right are judged to form a closed rectangle; otherwise, continuously comparing and judging the two straight lines which are intersected at the right and lower positions of the other vertical intersection positions until all closed rectangles are found out;
4) screening out all closed rectangles which contain the keywords and have only the inclusion or separation relation with other rectangles: firstly, acquiring all keywords from characters, and screening out closed rectangles of related keywords in the closed rectangles by judging the position relationship between the keywords and the closed rectangles; then, the position relation between the closed rectangle and the closed rectangle is judged, the intersection, the phase separation or the phase inclusion relation between the closed rectangle and the closed rectangle is judged, the closed rectangle with the intersection relation is removed, and all possible picture frames are obtained
5) Comprehensively judging and identifying the picture frame according to the relation between the closed rectangles, the area ratio of the inner and outer layer closed rectangles and the number and ordered arrangement of the keywords in the closed rectangles: firstly, respectively taking out a small group from the closed rectangle groups with the keywords in the extracted phase inclusion relationship, thereby obtaining a plurality of closed rectangles with the keywords; then calculating the area ratio of the outermost layer closed rectangle to the next outer layer closed rectangle; if the area ratio is not between 0.8 and 0.98, removing the outermost closed rectangle from the closed rectangle group, calculating the area ratio of the outermost closed rectangle to the next outer closed rectangle, and so on until two closed rectangles with the area ratio between 0.8 and 0.98 are found, and determining the closed rectangles as frames; or calculating the number and position arrangement condition of the keywords, and if the acquired ordered arranged keywords are positioned in a certain closed rectangle, determining the keywords as the picture frame.
Further, the vertical intersection positions are left upper, middle upper, right upper, left middle, middle, right middle, left lower, middle lower and right lower positions, respectively.
Further, the keyword is a text type entity specifically appearing in the drawing frame.
Further, the ordered arrangement is that a plurality of characters are arranged in a row or a column at equal intervals.
The method can more accurately identify the picture frame in the general DWG drawing under the condition that a plurality of pictures exist in the drawing, does not need to set a picture layer or a picture block for the picture frame, does not need to set various picture frame coordinates, and can determine each picture frame in the drawing according to the geometric relationship between the keywords and the rectangle. The picture frame can be identified more accurately through keywords and other picture frame key elements, and the picture frame with special conditions, such as the situation that the characters are scattered into lines, one line of characters is split into a plurality of single characters, the picture frame is formed by unclosed or closed rectangles of more than four lines, and the like, can be identified. The method realizes the identification of the universal picture frames with different design habits, and has higher universality and universality.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic view of a vertical intersection of a horizontal line and a vertical line;
FIG. 3 is a diagram illustrating a frame determination method.
Detailed Description
The invention is described in detail below with reference to the following figures and examples:
as shown in fig. 1, the present invention first traverses entities in the current space of a drawing to obtain all straight line entities and text entities, and the specific method is as follows: traversing the current space of the drawing to find out entities with attributes of straight lines, multi-segment lines, two-dimensional multi-segment lines, three-dimensional multi-segment lines, single-row characters, multi-row characters and attribute characters; the pieces need to be fried to obtain the above-mentioned solid mass; extracting initial point coordinate information of the linear entity and initial point coordinate information of the linear entity in the entity of the multi-line segment type, thereby obtaining a large amount of linear coordinate information, and screening out horizontal straight lines and vertical straight lines from the linear coordinate information; the text type entity acquires text content and bounding box information.
Then, intersecting every two of all horizontal straight lines and vertical straight lines, and solving a vertical intersection relationship, wherein the method specifically comprises the following steps: if the horizontal straight line is vertically intersected with the upper end point of the vertical straight line at the left end point, the relation of the two right angles is the upper left intersection; if the horizontal straight line is vertically intersected with the lower end point of the vertical straight line at any position in the middle, the relationship of the two right angles is that the two right angles are intersected in the middle and the bottom; by analogy, there are nine straight lines of vertical intersection positions (as shown in fig. 2) including upper left, upper middle, upper right, middle left, middle, middle right, lower left, middle lower and lower right; lines that do not intersect any other line are eliminated.
Then, finding out all closed rectangles in the drawing according to the vertical intersection position, specifically, the method is as follows: screening all straight lines with vertical intersection positions as upper left intersection and straight lines with vertical intersection positions as lower right intersection; extracting two straight lines which are intersected at the upper left and two straight lines which are intersected at the vertical intersection position at the lower right for judgment; if the lengths of the two horizontal straight lines are equal and the lengths of the two vertical straight lines are equal, the four straight lines can be judged to just form a closed rectangle; if the closed rectangles can not be formed, the two straight lines which are intersected at the right and lower positions of the other vertical intersection positions are continuously compared and judged until all the closed rectangles are found out.
Then, screening out all closed rectangles which contain the keywords and have only phase-containing or phase-separating relations with other rectangles, wherein the specific method is as follows: firstly, comparing the relation between all keywords and closed rectangles, and screening out the closed rectangles containing the keywords; the keywords are the text type entities specifically appearing in the picture frame, such as "design", "proofreading", "project responsible person", "general engineer", "building", "special registration chapter", and the like; then, acquiring the position relation between the closed rectangle and the closed rectangle to obtain the intersection, phase separation or phase inclusion relation between every two rectangles; if the four vertexes of the rectangle A are all positioned inside the rectangle B, the rectangle A comprises the rectangle B, and otherwise, the rectangle A is comprised of the rectangle B; and after the position relation among the rectangles is obtained, screening to obtain a closed rectangle which is provided with the key word and does not intersect with other rectangles, wherein the closed rectangle and the other closed rectangles only have the inclusion, included or separated relation.
Then, the frame is comprehensively judged and identified, specifically, the method is as follows: and taking out one group from the extracted closed rectangle group with the keywords, wherein only the closed rectangles with the keywords have the relation of inclusion and separation, thereby obtaining a plurality of closed rectangles with the keywords. And acquiring an outermost periphery closed rectangle and a secondary periphery closed rectangle. And calculating the area ratio of the two closed rectangles and the distance between the upper side, the lower side, the left side and the right side. If the area ratio is less than 0.8 or greater than 0.98, the area ratio of the sub-peripheral closed rectangle to the more inner layer rectangle is calculated until two closed rectangles with the area ratio greater than 0.8 and less than 0.98 are found, and the frame can be determined. The two closed rectangles are an outer frame and an inner frame of the picture frame respectively. If only one closed rectangle is skipped, the area ratio is not compared. And calculating the number and position arrangement condition of the keywords, and if the acquired ordered arranged keywords are positioned in a certain closed rectangle, determining the keywords as a picture frame. The characters are arranged in order, namely a plurality of characters are arranged in a row or a column at equal intervals.
As shown in FIG. 3, closed rectangles with keywords are obtained from dwg current space, and only inclusion, included and separated relationships among the closed rectangles are obtained. Wherein rectangle 2 and rectangle 3 form a frame, rectangle 4 and rectangle 5 form a frame, and rectangle 6 forms a frame, for a total of three frames. Specifically, the rectangles 1 to 6 are all closed rectangles containing keywords, but the area ratio of the rectangle 1 does not meet the judgment condition with the inner layer rectangles 2 and 3, and is excluded. The area ratio of the rectangle 2 to the rectangle 3, and the area ratio of the rectangle 4 to the rectangle 5, which both satisfy the frame determination condition, can be determined as a frame. In addition, the rectangle 6 has only one closed rectangle, but the interior of the rectangle has orderly arranged keywords, so that the frame can be judged as a frame according with the frame judgment condition.
While the preferred embodiments and examples of the present invention have been described in detail with reference to the accompanying drawings, it will be understood by those skilled in the art that the invention is not limited to the embodiments and examples, but is capable of modifications in accordance with the principles of the invention and the general knowledge of those skilled in the art without departing from the spirit and the general principles of the invention, and such modifications and variations are considered to be within the scope of the invention.
Claims (4)
1. A general method for identifying an AutoCAD drawing frame is characterized by comprising the following steps:
1) traversing the current space of the DWG drawing, screening out all straight line entities and character entities, and acquiring all horizontal straight lines, vertical straight lines and characters from the entities: firstly, acquiring common entities forming a picture frame in all block entities, linear entities, two-dimensional multi-line-segment entities, three-dimensional multi-line-segment entities, single-line character entities, multi-line character entities and attribute character entities; then recursively acquiring the common entities in the block entities and attribute character contents in the attribute block entities; extracting the starting point coordinates and the end point coordinates of the straight line entity, and eliminating non-horizontal and non-vertical straight lines;
2) acquiring the vertical intersection position of the horizontal straight line and the vertical straight line: firstly, sequencing all horizontal straight lines and all vertical straight lines respectively; then, taking a horizontal straight line, carrying out vertical intersection position judgment on the horizontal straight line and a vertical straight line which is possibly intersected, and sequentially comparing the rest straight lines according to the difference of the vertical intersection positions so as to obtain the corresponding vertical intersection positions of all the horizontal straight lines and the vertical straight lines;
3) finding out all closed rectangles in the drawing according to the vertical intersection position: firstly, screening all straight lines with vertical intersection positions of upper left intersection and straight lines with vertical intersection positions of lower right intersection; then, two straight lines with the vertical intersection position as the upper left and two straight lines with the vertical intersection position as the lower right are extracted for judgment, if the lengths of the two horizontal straight lines are equal and the lengths of the two vertical straight lines are equal, the two straight lines with the vertical intersection position as the upper left and the two straight lines with the vertical intersection position as the lower right are judged to form a closed rectangle; otherwise, continuously comparing and judging the two straight lines which are intersected at the right and lower positions of the other vertical intersection positions until all closed rectangles are found out;
4) screening out all closed rectangles which contain the keywords and have only the inclusion or separation relation with other rectangles: firstly, acquiring all keywords from characters, and screening out closed rectangles of related keywords in the closed rectangles by judging the position relationship between the keywords and the closed rectangles; then, the position relation between the closed rectangle and the closed rectangle is judged, the intersection, the phase separation or the phase inclusion relation between the closed rectangle and the closed rectangle is judged, the closed rectangle with the intersection relation is removed, and all possible picture frames are obtained
5) Comprehensively judging and identifying the picture frame according to the relation between the closed rectangles, the area ratio of the inner and outer layer closed rectangles and the number and ordered arrangement of the keywords in the closed rectangles: firstly, respectively taking out a small group from the closed rectangle groups with the keywords in the extracted phase inclusion relationship, thereby obtaining a plurality of closed rectangles with the keywords; then calculating the area ratio of the outermost layer closed rectangle to the next outer layer closed rectangle; if the area ratio is not between 0.8 and 0.98, removing the outermost closed rectangle from the closed rectangle group, calculating the area ratio of the outermost closed rectangle to the next outer closed rectangle, and so on until two closed rectangles with the area ratio between 0.8 and 0.98 are found, and determining the closed rectangles as frames; or calculating the number and position arrangement condition of the keywords, and if the acquired ordered arranged keywords are positioned in a certain closed rectangle, determining the keywords as the picture frame.
2. The method according to claim 1, wherein the vertical intersection positions are upper left, upper middle, upper right, middle left, middle, middle right, lower left, lower middle and lower right positions, respectively.
3. The method of claim 1, wherein the keyword is a text type entity specifically appearing in the frame.
4. The method according to claim 1, wherein the ordered arrangement comprises a plurality of words arranged equidistantly in a row or a column.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010448512.XA CN111611945A (en) | 2020-05-25 | 2020-05-25 | Universal AutoCAD picture frame identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010448512.XA CN111611945A (en) | 2020-05-25 | 2020-05-25 | Universal AutoCAD picture frame identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111611945A true CN111611945A (en) | 2020-09-01 |
Family
ID=72202404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010448512.XA Pending CN111611945A (en) | 2020-05-25 | 2020-05-25 | Universal AutoCAD picture frame identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111611945A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114155716A (en) * | 2022-02-08 | 2022-03-08 | 深圳市金蚁云数字科技有限公司 | Parking space management method, device, equipment and computer storage medium |
| CN117373052A (en) * | 2023-12-05 | 2024-01-09 | 江西少科智能建造科技有限公司 | CAD drawing frame information extraction method and system |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1930842A1 (en) * | 2006-12-08 | 2008-06-11 | Sony Corporation | Information classification |
| CN101673347A (en) * | 2008-09-12 | 2010-03-17 | 纬衡浩建科技（深圳）有限公司 | Spitting method of electronic drawing file |
| CN103793145A (en) * | 2014-02-21 | 2014-05-14 | 湖南省交通规划勘察设计院 | Drawing frame parameter identification and drawing output method based on autoCAD |
| CN104036060A (en) * | 2013-03-07 | 2014-09-10 | 纬衡浩建科技（深圳）有限公司 | Online auditing method and system for engineering drawing |
| CN105989347A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Intelligent marking method and system of objective questions |
| CN108121867A (en) * | 2017-12-21 | 2018-06-05 | 山东同智伟业软件股份有限公司 | The general quick method for splitting of DWG drawings |
| CN108491885A (en) * | 2018-03-28 | 2018-09-04 | 广东工业大学 | A kind of autoCAD graphic blocks identifying method and devices based on Naive Bayes Classifier |
| CN109255187A (en) * | 2018-09-12 | 2019-01-22 | 深圳市彬讯科技有限公司 | Analysis method and device |
| CN110084179A (en) * | 2019-04-24 | 2019-08-02 | 上海外高桥造船有限公司 | Picture frame recognition methods and system |
| EP3605277A1 (en) * | 2017-03-20 | 2020-02-05 | Beijing Kingsoft Office Software, Inc. | Method and device for quickly inserting recognized word |
| CN110765893A (en) * | 2019-09-30 | 2020-02-07 | 万翼科技有限公司 | Drawing file identification method, electronic equipment and related product |
| CN111126212A (en) * | 2019-12-13 | 2020-05-08 | 广东施富电气实业有限公司 | Matting identification method and system based on distribution box primary system diagram and storage medium |
-
2020
- 2020-05-25 CN CN202010448512.XA patent/CN111611945A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1930842A1 (en) * | 2006-12-08 | 2008-06-11 | Sony Corporation | Information classification |
| CN101673347A (en) * | 2008-09-12 | 2010-03-17 | 纬衡浩建科技（深圳）有限公司 | Spitting method of electronic drawing file |
| CN104036060A (en) * | 2013-03-07 | 2014-09-10 | 纬衡浩建科技（深圳）有限公司 | Online auditing method and system for engineering drawing |
| CN103793145A (en) * | 2014-02-21 | 2014-05-14 | 湖南省交通规划勘察设计院 | Drawing frame parameter identification and drawing output method based on autoCAD |
| CN105989347A (en) * | 2015-02-28 | 2016-10-05 | 科大讯飞股份有限公司 | Intelligent marking method and system of objective questions |
| EP3605277A1 (en) * | 2017-03-20 | 2020-02-05 | Beijing Kingsoft Office Software, Inc. | Method and device for quickly inserting recognized word |
| CN108121867A (en) * | 2017-12-21 | 2018-06-05 | 山东同智伟业软件股份有限公司 | The general quick method for splitting of DWG drawings |
| CN108491885A (en) * | 2018-03-28 | 2018-09-04 | 广东工业大学 | A kind of autoCAD graphic blocks identifying method and devices based on Naive Bayes Classifier |
| CN109255187A (en) * | 2018-09-12 | 2019-01-22 | 深圳市彬讯科技有限公司 | Analysis method and device |
| CN110084179A (en) * | 2019-04-24 | 2019-08-02 | 上海外高桥造船有限公司 | Picture frame recognition methods and system |
| CN110765893A (en) * | 2019-09-30 | 2020-02-07 | 万翼科技有限公司 | Drawing file identification method, electronic equipment and related product |
| CN111126212A (en) * | 2019-12-13 | 2020-05-08 | 广东施富电气实业有限公司 | Matting identification method and system based on distribution box primary system diagram and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 林悦香: "《工程制图与CAD》", 31 December 2016 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114155716A (en) * | 2022-02-08 | 2022-03-08 | 深圳市金蚁云数字科技有限公司 | Parking space management method, device, equipment and computer storage medium |
| CN117373052A (en) * | 2023-12-05 | 2024-01-09 | 江西少科智能建造科技有限公司 | CAD drawing frame information extraction method and system |
| CN117373052B (en) * | 2023-12-05 | 2024-02-23 | 江西少科智能建造科技有限公司 | CAD drawing frame information extraction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10223345B2 (en) | Interactively predicting fields in a form | |
| Li et al. | Rhythmic brushstrokes distinguish van Gogh from his contemporaries: findings via automated brushstroke extraction | |
| CN105868758B (en) | method and device for detecting text area in image and electronic equipment | |
| CN114581742B (en) | Linearity-based connected domain clustering fusion method, device, system and medium | |
| US7246306B2 (en) | Web information presentation structure for web page authoring | |
| CN104850633B (en) | A kind of three-dimensional model searching system and method based on the segmentation of cartographical sketching component | |
| JP3425834B2 (en) | Title extraction apparatus and method from document image | |
| EP1296278A2 (en) | Document analysis system and method | |
| CN111611945A (en) | Universal AutoCAD picture frame identification method | |
| CN114004204A (en) | Table structure reconstruction and character extraction method and system based on computer vision | |
| CN113239818B (en) | Table cross-modal information extraction method based on segmentation and graph convolution neural network | |
| CN111460204A (en) | Subgraph extraction method and device of electronic engineering drawing, storage medium and terminal | |
| CN114863408A (en) | Document content classification method, system, device and computer readable storage medium | |
| CN115223172A (en) | Text extraction method, device and equipment | |
| US9846949B2 (en) | Determine the shape of a representation of an object | |
| JPH0821057B2 (en) | Document image analysis method | |
| CN113435289A (en) | Space division method, device, equipment and storage medium | |
| Montgomery et al. | Mapping the perceptions of non-linguists in Northern England | |
| JP2017211976A (en) | Image processing device and image processing program | |
| CN109101973B (en) | Character recognition method, electronic device and storage medium | |
| CN109117841B (en) | Scene text detection method based on stroke width transformation and convolutional neural network | |
| CN112926581B (en) | Text detection method based on contour modeling | |
| CN112396056B (en) | Method for high-accuracy line division of text image OCR result | |
| JP5146279B2 (en) | Image processing apparatus and program | |
| Randriamasy et al. | Automatic benchmarking scheme for page segmentation |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200901 |
|
| WD01 | Invention patent application deemed withdrawn after publication |