CN107092864A - Drawing Reading text method and system based on clustering - Google Patents

Drawing Reading text method and system based on clustering Download PDF

Info

Publication number
CN107092864A
CN107092864A CN201710187205.9A CN201710187205A CN107092864A CN 107092864 A CN107092864 A CN 107092864A CN 201710187205 A CN201710187205 A CN 201710187205A CN 107092864 A CN107092864 A CN 107092864A
Authority
CN
China
Prior art keywords
coordinate
text box
text
clustering
value
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
Application number
CN201710187205.9A
Other languages
Chinese (zh)
Other versions
CN107092864B (en
Inventor
席斌
李明
王兴强
彭成超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Excellent Translation Information Technology Ltd By Share Ltd
Original Assignee
Chengdu Excellent Translation Information Technology Ltd By Share Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chengdu Excellent Translation Information Technology Ltd By Share Ltd filed Critical Chengdu Excellent Translation Information Technology Ltd By Share Ltd
Priority to CN201710187205.9A priority Critical patent/CN107092864B/en
Publication of CN107092864A publication Critical patent/CN107092864A/en
Application granted granted Critical
Publication of CN107092864B publication Critical patent/CN107092864B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/40Document-oriented image-based pattern recognition
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • G06F18/232Non-hierarchical techniques
    • G06F18/2321Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/22Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • G06V10/225Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10Character recognition

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Character Input (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention discloses the drawing Reading text method based on clustering, comprise the following steps:S1:Text box on drawing is subjected to angle classification according to its angle;S2:Extract the coordinate characteristic value of the text box of same angular type;S3:Clustering is carried out to the text box of same angular type so that the similar text box of coordinate characteristic value is gathered in same class, and text box is ranked up according to cluster result;S4:Text box after sequence is subjected to word output according to the angular type of text frame.The invention also discloses the drawing Reading text system based on clustering using the inventive method.Drawing Reading text method and system of the invention based on clustering, by the way that text box is clustered, so that the close text box of content is gathered same class, then output character, avoid continuous content and be resolved to and be separated by very remote place so that extract drawing text and translated more convenient.

Description

Drawing Reading text method and system based on clustering
Technical field
The present invention relates to text composition field, and in particular to the drawing Reading text method and system based on clustering.
Background technology
As international cooperation is increasingly strengthened, either Chinese companies or foreign corporation, in international project and international scientific research Project verification and development during, be required for being exchanged with other side by drawing file.Due to originals such as multiple modification, manual typesettings Cause, it should be complete a word that drawing, which often occurs, be split as multiple text boxes be put into manually it is adjacent to each other Position.And due to the characteristic (text box is write successively according to the write time hereof) of drawing file itself, institute backward Might not be adjacent on content is preserved with the text box for seeming adjacent on drawing, some are possibly even very remote every obtaining, this During so that extracting drawing text, continuous content, which is resolved to, is separated by very remote place, and this just goes forward side by side to extracting drawing text Row translation causes huge puzzlement.
The content of the invention
The technical problems to be solved by the invention be extract drawing text when, continuous content be resolved to be separated by it is very remote Place, cause extract drawing text and carry out translation inconvenience, it is therefore intended that provide the drawing Reading text based on clustering Method and system, solve the above problems.
The present invention is achieved through the following technical solutions:
Drawing Reading text method based on clustering, comprises the following steps:S1:By the text box on drawing according to it Angle carries out angle classification;S2:Extract the coordinate characteristic value of the text box of same angular type;S3:To same angular type Text box carries out clustering so that the similar text box of coordinate characteristic value is gathered in same class, and according to cluster result pair Text box is ranked up;S4:Text box after sequence is subjected to word output according to the angular type of text frame.
Due to reasons such as multiple modification, manual typesettings, it should be complete a word that drawing, which often occurs, be split It is divided into multiple text boxes and has been put into position adjacent to each other manually.And due to the characteristic of drawing file itself, (text box is in file In write backward successively according to the write time), so seeming adjacent text box on drawing on content is preserved and differing Fixed adjacent, some are possibly even very remote every obtaining, and when this to extract drawing text, continuous content, which is resolved to, to be separated by very Remote place, this just causes huge puzzlement to extracting drawing text and translate.
When the present invention is applied, angle classification first is carried out according to its angle to the text box on drawing, text box is divided into many Individual different angular type, as often occurred in cad drawings:0 °, 90 °, 180 ° and 270 °;Then same angular type is extracted The coordinate characteristic value of text box, this coordinate characteristic value is the coordinate value for identifying the exclusive position of each text box;Subsequently, it is right The text box of same angular type carries out clustering so that the similar text box of coordinate characteristic value is gathered in same class, and Text box is ranked up according to cluster result;Finally, the text box after sequence is carried out according to the angular type of text frame Word is exported.By taking cad drawings as an example:Angle such as text box is 0 °, then the order of word output is from left to right;Such as text box Angle be 90 °, then the order of word output is from top to bottom;Angle such as text box is 180 °, then the order of word output For from right to left;Angle such as text box is 270 °, then the order of word output is from top to bottom.The present invention is by by text box Cluster so that the close text box of content is gathered same class, then output character, it is to avoid continuous content is resolved to It is separated by very remote place so that extract drawing text and translated more convenient.
Further, the coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, the upper right corner in the text box upper left corner Coordinate value, the coordinate value of the coordinate value in the lower right corner or central point.
When the present invention is applied, coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, the upper right corner in the text box upper left corner Coordinate value, the coordinate value of the coordinate value in the lower right corner or central point, because coordinate characteristic value is that to identify each text box exclusive The coordinate value of position, above-mentioned five kinds of coordinate values can express the exclusive position of text frame, effectively increase cluster of the present invention Accuracy.
Further, the cluster uses optics algorithms;The optics algorithms are by reading in orderly text box group The coordinate characteristic value of text box determines the relative distance between text box;The similar standard of the coordinate characteristic value be it is relative away from From less than or equal to threshold value.
When the present invention is applied, due to the text box on drawing have rule but it is irregular the characteristics of, inventor pass through wound The property made work is found, because text box is irregular, so it is difficult to accurately determine clustering parameter, and when using optics algorithms, gather When reasonable change occurs for class parameter, cluster result change is little.And in the optics algorithms that the present invention is applied, have by reading The coordinate characteristic value of sequence text box group Chinese version frame determines the relative distance between text box, is then made with this relative distance The data required for optics algorithms, this mode can effectively determine whether text box is assembled, and the text that will do not assemble Frame is assigned in different classes.In optics algorithms of the present invention, the similar standard of coordinate characteristic value is that relative distance is less than threshold value, The threshold value is density value in present invention application, it is only necessary to provide rational density value, you can reasonably cluster text box, have Effect reduction cluster error.
Further, step S3 also includes following sub-step:After the completion of cluster, according to the coordinate characteristic value of text box to every The shape that one class Chinese version frame is constituted is judged;If what text box was constituted is shaped as preset shape, to such execution S4;If the shape that text box is constituted is not preset shape, clustering parameter is adjusted to such and S3 is performed.
When the present invention is applied, by taking civil construction cad drawings as an example:The characteristics of civil construction cad drawing Chinese version frames is, such as The angle of fruit text box is 0 °, then the horizontal coordinate of the first row text box is same or similar, meanwhile, the longitudinal direction of first row text box Coordinate is same or similar, so preset shape is same or similar while first row text using the horizontal coordinate of the first row text box The same or analogous shape of longitudinal coordinate of frame.And in clustering, it sometimes appear that the shape that the text box after cluster is constituted Shape is T-shape or circle, and at this moment, it is not preset shape to assert the shape, and is clustered to such adjustment clustering parameter and again. The present invention effectively raises the accuracy of cluster by the judgement of the shape constituted to text box.
Drawing Reading text system based on clustering, including for the text box on drawing to be carried out according to its angle The sort module of angle classification;Extraction module for the coordinate characteristic value of the text box that extracts same angular type;For right The text box of same angular type carries out clustering so that the similar text box of coordinate characteristic value is gathered in same class, and The cluster module being ranked up according to cluster result to text box;For by angle of the text box after sequence according to text frame Type carries out the output module of word output.
When the present invention is applied, sort module first carries out angle classification to the text box on drawing according to its angle, by text Frame is divided into multiple different angular types, as often occurred in cad drawings:0 °, 90 °, 180 ° and 270 °;Then extraction module is carried The coordinate characteristic value of the text box of same angular type is taken, this coordinate characteristic value is to identify the exclusive position of each text box Coordinate value;Subsequently, cluster module carries out clustering to the text box of same angular type so that coordinate characteristic value is similar Text box is gathered in same class, and text box is ranked up according to cluster result;Finally, output module is by the text after sequence This frame carries out word output according to the angular type of text frame.By taking cad drawings as an example:Angle such as text box is 0 °, then literary The order of word output is from left to right;Angle such as text box is 90 °, then the order of word output is from top to bottom;Such as text The angle of frame is 180 °, then the order of word output is from right to left;Angle such as text box is 270 °, then word output is suitable Sequence is from top to bottom.The present invention is by the way that text box is clustered so that the close text box of content is gathered same class, then defeated Go out word, it is to avoid continuous content, which is resolved to, is separated by very remote place so that extract drawing text and simultaneously translated more Plus conveniently.
Further, the coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, the upper right corner in the text box upper left corner Coordinate value, the coordinate value of the coordinate value in the lower right corner or central point.
When the present invention is applied, coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, the upper right corner in the text box upper left corner Coordinate value, the coordinate value of the coordinate value in the lower right corner or central point, because coordinate characteristic value is that to identify each text box exclusive The coordinate value of position, above-mentioned five kinds of coordinate values can express the exclusive position of text frame, effectively increase cluster of the present invention Accuracy.
Further, the cluster uses optics algorithms;The optics algorithms are by reading in orderly text box group The coordinate characteristic value of text box determines the relative distance between text box;The similar standard of the coordinate characteristic value be it is relative away from From less than or equal to threshold value.
When the present invention is applied, due to the text box on drawing have rule but it is irregular the characteristics of, inventor pass through wound The property made work is found, because text box is irregular, so it is difficult to accurately determine clustering parameter, and when using optics algorithms, gather When reasonable change occurs for class parameter, cluster result change is little.And in the optics algorithms that the present invention is applied, have by reading The coordinate characteristic value of sequence text box group Chinese version frame determines the relative distance between text box, is then made with this relative distance The data required for optics algorithms, this mode can effectively determine whether text box is assembled, and the text that will do not assemble Frame is assigned in different classes.In optics algorithms of the present invention, the similar standard of coordinate characteristic value is that relative distance is less than threshold value, The threshold value is density value in present invention application, it is only necessary to provide rational density value, you can reasonably cluster text box, have Effect reduction cluster error.
Further, cluster module is additionally operable to after the completion of cluster, according to the coordinate characteristic value of text box to each class The shape that Chinese version frame is constituted is judged;If what text box was constituted is shaped as preset shape, such is sent to output Module;If the shape that text box is constituted is not preset shape, cluster to such adjustment clustering parameter and again.
When the present invention is applied, by taking civil construction cad drawings as an example:The characteristics of civil construction cad drawing Chinese version frames is, such as The angle of fruit text box is 0 °, then the horizontal coordinate of the first row text box is same or similar, meanwhile, the longitudinal direction of first row text box Coordinate is same or similar, so preset shape is same or similar while first row text using the horizontal coordinate of the first row text box The same or analogous shape of longitudinal coordinate of frame.And in clustering, it sometimes appear that the shape that the text box after cluster is constituted Shape is T-shape or circle, and at this moment, it is not preset shape, and to being different from preset shape in such text box to assert the shape Part cluster again.The present invention effectively raises the accuracy of cluster by the judgement of the shape constituted to text box.
The present invention compared with prior art, has the following advantages and advantages:
1st, the drawing Reading text method of the invention based on clustering, by the way that text box is clustered so that content is close Text box gathered same class, then output character, it is to avoid continuous content, which is resolved to, is separated by very remote place, More facilitate so that extracting drawing text and being translated;
2nd, the drawing Reading text method of the invention based on clustering, the coordinate characteristic value of use can express this article The exclusive position of this frame, effectively increases the accuracy that the present invention is clustered;
3rd, the drawing Reading text method of the invention based on clustering, unwise to clustering parameter using optics algorithms Sense, can effectively reduce cluster error;
4th, the drawing Reading text method of the invention based on clustering, by the judgement of the shape constituted to text box, Effectively raise the accuracy of cluster;
5th, the drawing Reading text system of the invention based on clustering, by the way that text box is clustered so that content is close Text box gathered same class, then output character, it is to avoid continuous content, which is resolved to, is separated by very remote place, More facilitate so that extracting drawing text and being translated;
6th, the drawing Reading text system of the invention based on clustering, the coordinate characteristic value of use can express this article The exclusive position of this frame, effectively increases the accuracy that the present invention is clustered;
7th, the drawing Reading text system of the invention based on clustering, unwise to clustering parameter using optics algorithms Sense, can effectively reduce cluster error;
8th, the drawing Reading text system of the invention based on clustering, by the judgement of the shape constituted to text box, Effectively raise the accuracy of cluster.
Brief description of the drawings
Accompanying drawing described herein is used for providing further understanding the embodiment of the present invention, constitutes one of the application Point, do not constitute the restriction to the embodiment of the present invention.In the accompanying drawings:
Fig. 1 is the inventive method step schematic diagram;
Fig. 2 is present system structural representation;
Fig. 3 is the schematic diagram of embodiment 5.
Embodiment
For the object, technical solutions and advantages of the present invention are more clearly understood, with reference to embodiment and accompanying drawing, to this Invention is described in further detail, and exemplary embodiment and its explanation of the invention is only used for explaining the present invention, does not make For limitation of the invention.
Embodiment 1
As shown in figure 1, the drawing Reading text method of the invention based on clustering, comprises the following steps:S1:By drawing On text box according to its angle carry out angle classification;S2:Extract the coordinate characteristic value of the text box of same angular type;S3: Clustering is carried out to the text box of same angular type so that the similar text box of coordinate characteristic value is gathered in same class, And text box is ranked up according to cluster result;S4:Text box after sequence is carried out according to the angular type of text frame Word is exported.
When the present embodiment is implemented, angle classification first is carried out according to its angle to the text box on drawing, text box is divided into Multiple different angular types, as often occurred in cad drawings:0 °, 90 °, 180 ° and 270 °;Then same angular type is extracted Text box coordinate characteristic value, this coordinate characteristic value is the coordinate value for identifying each text box exclusive position;Subsequently, Clustering is carried out to the text box of same angular type so that the similar text box of coordinate characteristic value is gathered in same class, And text box is ranked up according to cluster result;Finally, the text box after sequence is entered according to the angular type of text frame Word of composing a piece of writing is exported.By taking cad drawings as an example:Angle such as text box is 0 °, then the order of word output is from left to right;Such as text The angle of frame is 90 °, then the order of word output is from top to bottom;Angle such as text box is 180 °, then word output is suitable Sequence is from right to left;Angle such as text box is 270 °, then the order of word output is from top to bottom.The present invention is by by text Frame is clustered so that the close text box of content is gathered same class, then output character, it is to avoid continuous content is resolved To being separated by very remote place so that extract drawing text and simultaneously translated and more facilitate.
Embodiment 2
The present embodiment is on the basis of embodiment 1, and the coordinate characteristic value uses the coordinate value in the text box upper left corner, lower-left The coordinate value at angle, the coordinate value in the upper right corner, the coordinate value of the coordinate value in the lower right corner or central point.
When the present embodiment is implemented, coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, upper right in the text box upper left corner The coordinate value of the coordinate value at angle, the coordinate value in the lower right corner or central point, due to coordinate characteristic value be identify each text box it is only There is the coordinate value of position, above-mentioned five kinds of coordinate values can express the exclusive position of text frame, effectively increase the present invention poly- The accuracy of class.
Embodiment 3
The present embodiment is on the basis of embodiment 1, and the cluster uses optics algorithms;The optics algorithms are by reading The coordinate characteristic value of orderly text box group Chinese version frame is taken to determine the relative distance between text box;The coordinate characteristic value phase As standard be relative distance be less than or equal to threshold value.
The present embodiment implement when, due to the text box on drawing have rule but it is irregular the characteristics of, inventor passes through Creative work is found, because text box is irregular, so it is difficult to accurately determine clustering parameter, and when using optics algorithms, When reasonable change occurs for clustering parameter, cluster result change is little.And in the optics algorithms that the present invention is applied, by reading The coordinate characteristic value of orderly text box group Chinese version frame determines the relative distance between text box, then with this relative distance As data required for optics algorithms, this mode can effectively determine whether text box is assembled, and the text that will do not assemble This frame is assigned in different classes.In optics algorithms of the present invention, the similar standard of coordinate characteristic value is that relative distance is less than threshold Value, the threshold value is density value in present invention application, it is only necessary to provide rational density value, you can reasonably gather text box Class, effectively reduction cluster error.
Embodiment 4
The present embodiment is on the basis of embodiment 1, and step S3 also includes following sub-step:After the completion of cluster, according to text The shape that the coordinate characteristic value of frame is constituted to each class Chinese version frame judges;If what text box was constituted is shaped as presetting Shape, then to such execution S4;If the shape that text box is constituted is not preset shape, clustering parameter is adjusted to such and held Row S3.
When the present embodiment is implemented, by taking civil construction cad drawings as an example:The characteristics of civil construction cad drawing Chinese version frames is, If the angle of text box is 0 °, the horizontal coordinate of the first row text box is same or similar, meanwhile, first row text box it is vertical It is same or similar to coordinate, so preset shape is same or similar while first row is literary using the horizontal coordinate of the first row text box The same or analogous shape of longitudinal coordinate of this frame.And in clustering, it sometimes appear that what the text box after cluster was constituted T-shape or circle are shaped as, at this moment, it is not preset shape to assert the shape, and such adjustment clustering parameter is clustered again. The present invention effectively raises the accuracy of cluster by the judgement of the shape constituted to text box.
Embodiment 5
As shown in figure 3, the present embodiment is on the basis of embodiment 1 to 4, the drawing word in Fig. 3 is handled.
When the present embodiment is implemented,
It is 0 ° that the text box on drawing first is carried out into text boxes all on angle classification, drawing according to its angle;Then The coordinate characteristic value of the text box of same angular type is extracted, coordinate characteristic value is defined as the text box upper left corner and sat by the present embodiment Scale value.
Text box extraction effect is as follows:
3637:One:[coordinate is:X=1447657412.032166 y=926543984.4671117]
3638:1 [coordinate is:X=1447657587.564081 y=926543984.4671117]
3639:, this engineering ± [coordinate is:X=1447657618.814081 y=926543984.4671117]
3640:0.000 [coordinate is:X=1447658040.20888 y=926543984.4671117]
3641:[coordinate is above wall:X=1447658233.95888 y=926543984.4671117]
3642:200 [coordinate is:X=1447658585.02271 y=926543984.4671117]
3643:Thick concrete brick, [coordinate is:X=1447658716.27271 y=926543984.4671117]
3644:Two:[coordinate is engineering way:X=1447661758.77883 y=926544012.7807627]
3645:Strength grade is [coordinate is:X=1447657709.707235 y=926543795.2022269]
3646:[coordinate is MU10:X=1447658148.537022 y=926543795.2022269]
3647:, use [coordinate for:X=1447658329.787022 y=926543795.2022269]
3648:[coordinate is M5:X=1447658593.084894 y=926543795.2022269]
3649:Mixed mortar is built by laying bricks or stones, and [coordinate is for thickness of wall body and column size:X=1447658693.084895 y= 926543795.2022269]
3650:[coordinate is exterior wall:X=1447661936.167565 y=926543783.6524901]
3651:1 [coordinate is:X=1447662111.69948 y=926543783.6524901]
3652::[coordinate is:X=1447662142.94948 y=926543783.6524901]
3653:1:1 [coordinate is:X=1447662373.119285 y=926543783.6524901]
3654:[coordinate is white cement wiping seam:X=1447662454.369285 y=926543783.6524901]
3655:Detailed plane and profile.± [coordinate is:X=1447657703.398538 y= 926543593.179681]
3656:0.000 [coordinate is:X=1447658475.857167 y=926543593.179681]
3657:Following wall in detail apply by knot.[coordinate is:X=1447658669.607167 y=926543593.179681]
3658:[coordinate is 8-10:X=1447662373.119285 y=926543599.0564725]
3659:Thick light gray facing tile, concrete interface agent, strong adherence force are brushed one times on brick sticking veneer with patch with painting [coordinate is:X=1447662535.619285 y=926543599.0564725]
3660:2 [coordinate is:X=1447657568.554384 y=926543382.6952552]
3661:, enclosure wall is every [coordinate is:X=1447657612.304384 y=926543382.6952552]
3662:40 [coordinate is:X=1447658051.134172 y=926543382.6952552]
3663:Rice sets an expansion joint, and [coordinate is slit width:X=1447658144.884172 y= 926543382.6952552]
3664:[coordinate is 30mm.:X=1447658934.777789 y=926543382.6952552]
3665:3 [coordinate is:X=1447657558.485019 y=926543184.2143952]
3666:, wall damp-proof course:[coordinate is under terrace:X=1447657602.235019 y= 926543184.2143952]
3667:- 0.060 [coordinate is:X=1447658567.660551 y=926543184.2143952]
3668:Place does that [coordinate is:X=1447658805.160551 y=926543184.2143952]
3669:20 [coordinate is:X=1447658980.692466 y=926543184.2143952]
3670:It is thick that [coordinate is:X=1447659068.192466 y=926543184.2143952]
3671:1:2 [coordinate is:X=1447659155.958423 y=926543184.2143952]
3672:In cement mortar plus equivalent to [coordinate is:X=1447659249.708423 y= 926543184.2143952]
3673:[coordinate is cement weight:X=1447657722.28492 y=926542961.0158562]
3674:5% [coordinate is:X=1447658073.34875 y=926542961.0158562]
3675:Polymer mortar (this absolute altitude be reinforced concrete construction when do not do).[coordinate is:X= 1447658185.84875 y=926542961.0158562]
3676:4 [coordinate is:X=1447657558.485019 y=926542757.3695289]
3677:, cross between water hole two one, hole inwall smears that [coordinate is:X=1447657608.485019 y= 926542757.3695289]
3678:20 [coordinate is:X=1447658749.442466 y=926542757.3695289]
3679:It is thick that [coordinate is:X=1447658836.942466 y=926542757.3695289]
3680:1:2 [coordinate is:X=1447658924.708423 y=926542757.3695289]
3681:Cement mortar adds that [coordinate is:X=1447659018.458423 y=926542757.3695289]
3682:5% [coordinate is:X=1447659457.288211 y=926542757.3695289]
3683:Waterproofing agent.[coordinate is:X=1447659569.788211 y=926542757.3695289]
3684:5 [coordinate is:X=1447657558.485019 y=926542575.1381977]
3685:, iron artistic fence by specialized factory's fabrication and installation, its color and fancy have Party A to make by oneself.[coordinate is:X= 1447657602.235019 y=926542575.1381977]
3686:6 [coordinate is:X=1447657558.485019 y=926542375.0185961]
3687:, have not clear part in work progress on drawing, should contact and resolve through consultation with designing unit in time.[coordinate For:X=1447657602.235019 y=926542375.0185961]
3688:It all should be performed in construction by Current Building ' installing engineering acceptance specification and relevant standard, specification, regulation.[sit It is designated as:X=1447657725.294654 y=926542188.788369]
3690:6 [coordinate is:X=1447662373.119285 y=926543404.5512771]
3691:It is thick that [coordinate is:X=1447662416.869285 y=926543404.5512771]
3692:1:2.5 [coordinate is:X=1447662504.635242 y=926543404.5512771]
3693:[coordinate is cement mortar (mixing building adhesive):X=1447662660.885242 y= 926543404.5512771]
3694:12 [coordinate is:X=1447662357.412471 y=926543231.9392951]
3695:It is thick that [coordinate is:X=1447662432.412471 y=926543231.9392951]
3696:1 [coordinate is:X=1447662520.178428 y=926543231.9392951]
3697::[coordinate is:X=1447662551.428428 y=926543231.9392951]
3698:3 [coordinate is:X=1447662639.194386 y=926543231.9392951]
3699:[coordinate is for cement mortar bottoming brooming:X=1447662682.944385 y= 926543231.9392951]
3700:[coordinate is base course wall:X=1447662391.443578 y=926543056.9064493]
Foremost is the numbering of text box, and centre is text box content, is finally text box top left co-ordinate.By above-mentioned interior Appearance is visible, after drawing text box is read, and the text box that No. 3644 grade should largely belong to right side has been inserted into left side text In this frame.Then clustering is carried out to text box so that the similar text box of coordinate characteristic value is gathered in same class, and root Text box clooating sequence is adjusted according to cluster result;Clustering algorithm uses optics algorithms, and the density value of optics algorithms takes drawing 1 the percent of size.
The later effect of cluster is as follows:
Class 1:
3637:One:[coordinate is:X=1447657412.032166 y=926543984.4671117]
3638:1 [coordinate is:X=1447657587.564081 y=926543984.4671117]
3639:, this engineering ± [coordinate is:X=1447657618.814081 y=926543984.4671117]
3640:0.000 [coordinate is:X=1447658040.20888 y=926543984.4671117]
3641:[coordinate is above wall:X=1447658233.95888 y=926543984.4671117]
3642:200 [coordinate is:X=1447658585.02271 y=926543984.4671117]
3643:Thick concrete brick, [coordinate is:X=1447658716.27271 y=926543984.4671117]
3645:Strength grade is [coordinate is:X=1447657709.707235 y=926543795.2022269]
3646:[coordinate is MU10:X=1447658148.537022 y=926543795.2022269]
3647:, use [coordinate for:X=1447658329.787022 y=926543795.2022269]
3648:[coordinate is M5:X=1447658593.084894 y=926543795.2022269]
3649:Mixed mortar is built by laying bricks or stones, and [coordinate is for thickness of wall body and column size:X=1447658693.084895 y= 926543795.2022269]
3655:Detailed plane and profile.± [coordinate is:X=1447657703.398538 y= 926543593.179681]
3656:0.000 [coordinate is:X=1447658475.857167 y=926543593.179681]
3657:Following wall in detail apply by knot.[coordinate is:X=1447658669.607167 y=926543593.179681]
3660:2 [coordinate is:X=1447657568.554384 y=926543382.6952552]
3661:, enclosure wall is every [coordinate is:X=1447657612.304384 y=926543382.6952552]
3662:40 [coordinate is:X=1447658051.134172 y=926543382.6952552]
3663:Rice sets an expansion joint, and [coordinate is slit width:X=1447658144.884172 y= 926543382.6952552]
3664:[coordinate is 30mm.:X=1447658934.777789 y=926543382.6952552]
3665:3 [coordinate is:X=1447657558.485019 y=926543184.2143952]
3666:, wall damp-proof course:[coordinate is under terrace:X=1447657602.235019 y= 926543184.2143952]
3667:- 0.060 [coordinate is:X=1447658567.660551 y=926543184.2143952]
3668:Place does that [coordinate is:X=1447658805.160551 y=926543184.2143952]
3669:20 [coordinate is:X=1447658980.692466 y=926543184.2143952]
3670:It is thick that [coordinate is:X=1447659068.192466 y=926543184.2143952]
3671:1:2 [coordinate is:X=1447659155.958423 y=926543184.2143952]
3672:In cement mortar plus equivalent to [coordinate is:X=1447659249.708423 y= 926543184.2143952]
3673:[coordinate is cement weight:X=1447657722.28492 y=926542961.0158562]
3674:5% [coordinate is:X=1447658073.34875 y=926542961.0158562]
3675:Polymer mortar (this absolute altitude be reinforced concrete construction when do not do).[coordinate is:X= 1447658185.84875 y=926542961.0158562]
3676:4 [coordinate is:X=1447657558.485019 y=926542757.3695289]
3677:, cross between water hole two one, hole inwall smears that [coordinate is:X=1447657608.485019 y= 926542757.3695289]
3678:20 [coordinate is:X=1447658749.442466 y=926542757.3695289]
3679:It is thick that [coordinate is:X=1447658836.942466 y=926542757.3695289]
3680:1:2 [coordinate is:X=1447658924.708423 y=926542757.3695289]
3681:Cement mortar adds that [coordinate is:X=1447659018.458423 y=926542757.3695289]
3682:5% [coordinate is:X=1447659457.288211 y=926542757.3695289]
3683:Waterproofing agent.[coordinate is:X=1447659569.788211 y=926542757.3695289]
3684:5 [coordinate is:X=1447657558.485019 y=926542575.1381977]
3685:, iron artistic fence by specialized factory's fabrication and installation, its color and fancy have Party A to make by oneself.[coordinate is:X= 1447657602.235019 y=926542575.1381977]
3686:6 [coordinate is:X=1447657558.485019 y=926542375.0185961]
3687:, have not clear part in work progress on drawing, should contact and resolve through consultation with designing unit in time.[coordinate For:X=1447657602.235019 y=926542375.0185961]
3688:It all should be performed in construction by Current Building ' installing engineering acceptance specification and relevant standard, specification, regulation.[sit It is designated as:X=1447657725.294654 y=926542188.788369]
Class 2:
3644:Two:[coordinate is engineering way:X=1447661758.77883 y=926544012.7807627]
3650:[coordinate is exterior wall:X=1447661936.167565 y=926543783.6524901]
Class 3:
3651:1 [coordinate is:X=1447662111.69948 y=926543783.6524901]
3652::[coordinate is:X=1447662142.94948 y=926543783.6524901]
3653:1:1 [coordinate is:X=1447662373.119285 y=926543783.6524901]
3654:[coordinate is white cement wiping seam:X=1447662454.369285 y=926543783.6524901]
3658:[coordinate is 8-10:X=1447662373.119285 y=926543599.0564725]
3659:Thick light gray facing tile, concrete interface agent, strong adherence force are brushed one times on brick sticking veneer with patch with painting [coordinate is:X=1447662535.619285 y=926543599.0564725]
3690:6 [coordinate is:X=1447662373.119285 y=926543404.5512771]
3691:It is thick that [coordinate is:X=1447662416.869285 y=926543404.5512771]
3692:1:2.5 [coordinate is:X=1447662504.635242 y=926543404.5512771]
3693:[coordinate is cement mortar (mixing building adhesive):X=1447662660.885242 y= 926543404.5512771]
3694:12 [coordinate is:X=1447662357.412471 y=926543231.9392951]
3695:It is thick that [coordinate is:X=1447662432.412471 y=926543231.9392951]
3696:1 [coordinate is:X=1447662520.178428 y=926543231.9392951]
3697::[coordinate is:X=1447662551.428428 y=926543231.9392951]
3698:3 [coordinate is:X=1447662639.194386 y=926543231.9392951]
3699:[coordinate is for cement mortar bottoming brooming:X=1447662682.944385 y= 926543231.9392951]
3700:[coordinate is base course wall:X=1447662391.443578 y=926543056.9064493]
It can see from the effect after cluster, the text box for belonging to right side has been gathered in class 2 and class 3, and belong to left side Text box gathered in class 1.
Finally, the text box after adjustment order is subjected to word output according to the angular type of text frame.
Content after output is as follows:
One:1st, the thick concrete brick of more than this engineering ± 0.000 wall 200, strength grade is MU10, using M5 mixed mortars Build by laying bricks or stones, thickness of wall body and the detailed plane of column size and profile.Less than ± 0.000 wall in detail apply by knot.2nd, enclosure wall is set every 40 meters One expansion joint, slit width 30mm3, wall damp-proof course:20 thickness 1 are done at -0.060 under terrace:In 2 cement mortar plus equivalent to cement The polymer mortar (not done when this absolute altitude is reinforced concrete construction) of weight 5%.4th, one is crossed between water hole two, and hole inwall is smeared 20 thickness 1:2 cement mortar add 5% waterproofing agent.5th, iron artistic fence has Party A certainly by specialized factory's fabrication and installation, its color and fancy It is fixed.6th, there is not clear part in work progress on drawing, should contact and resolve through consultation with designing unit in time.All should be by existing in construction Building installation engineering acceptance specification and relevant standard, specification, regulation are performed.Two:Engineering way exterior wall 1:1:1 white cement wiping seam 8- 10 thick light grey facing tiles, concrete interface agent, the strong thickness 1 of adherence force 6 are brushed one times on brick sticking veneer with patch with painting:2.5 water The thickness 1 of cement mortar (mixing building adhesive) 12:3 cement mortar bottoming brooming base course walls
From the foregoing, it will be observed that the content after output meets reading logic, it is to avoid continuous content, which is resolved to, to be separated by Very remote place so that extract drawing text and translated more convenient.
Embodiment 6
As shown in Fig. 2 the drawing Reading text system based on clustering of the invention, including for by the text on drawing Frame carries out the sort module of angle classification according to its angle;Coordinate characteristic value for the text box that extracts same angular type Extraction module;Clustering is carried out for the text box to same angular type so that the similar text box quilt of coordinate characteristic value Gather in same class, and the cluster module being ranked up according to cluster result to text box;For by the text box root after sequence The output module of word output is carried out according to the angular type of text frame.
When the present embodiment is implemented, sort module, extraction module and output module are preferably ARM7 processors, and cluster module is excellent Elect Cortex-A7 processors as.Sort module first carries out angle classification to the text box on drawing according to its angle, by text box Multiple different angular types are divided into, as often occurred in cad drawings:0 °, 90 °, 180 ° and 270 °;Then extraction module is extracted The coordinate characteristic value of the text box of same angular type, this coordinate characteristic value is the seat for identifying the exclusive position of each text box Scale value;Subsequently, cluster module carries out clustering to the text box of same angular type so that the similar text of coordinate characteristic value This frame is gathered in same class, and text box is ranked up according to cluster result;Finally, output module is by the text after sequence Frame carries out word output according to the angular type of text frame.By taking cad drawings as an example:Angle such as text box is 0 °, then word The order of output is from left to right;Angle such as text box is 90 °, then the order of word output is from top to bottom;Such as text box Angle be 180 °, then the order of word output is from right to left;Angle such as text box is 270 °, then the order of word output For from top to bottom.The present invention is by the way that text box is clustered so that the close text box of content is gathered same class, is then exported Word, it is to avoid continuous content, which is resolved to, is separated by very remote place so that extract drawing text and simultaneously translated more It is convenient.
Embodiment 7
The present embodiment is on the basis of embodiment 6, and the coordinate characteristic value uses the coordinate value in the text box upper left corner, lower-left The coordinate value at angle, the coordinate value in the upper right corner, the coordinate value of the coordinate value in the lower right corner or central point.
When the present embodiment is implemented, coordinate characteristic value uses the coordinate value, the coordinate value in the lower left corner, upper right in the text box upper left corner The coordinate value of the coordinate value at angle, the coordinate value in the lower right corner or central point, due to coordinate characteristic value be identify each text box it is only There is the coordinate value of position, above-mentioned five kinds of coordinate values can express the exclusive position of text frame, effectively increase the present invention poly- The accuracy of class.
Embodiment 8
The present embodiment is on the basis of embodiment 6, and the cluster uses optics algorithms;The optics algorithms are by reading The coordinate characteristic value of orderly text box group Chinese version frame is taken to determine the relative distance between text box;The coordinate characteristic value phase As standard be relative distance be less than or equal to threshold value.
The present embodiment implement when, due to the text box on drawing have rule but it is irregular the characteristics of, inventor passes through Creative work is found, because text box is irregular, so it is difficult to accurately determine clustering parameter, and when using optics algorithms, When reasonable change occurs for clustering parameter, cluster result change is little.And in the optics algorithms that the present invention is applied, by reading The coordinate characteristic value of orderly text box group Chinese version frame determines the relative distance between text box, then with this relative distance As data required for optics algorithms, this mode can effectively determine whether text box is assembled, and the text that will do not assemble This frame is assigned in different classes.In optics algorithms of the present invention, the similar standard of coordinate characteristic value is that relative distance is less than threshold Value, the threshold value is density value in present invention application, it is only necessary to provide rational density value, you can reasonably gather text box Class, effectively reduction cluster error.
Embodiment 9
The present embodiment is on the basis of embodiment 6, and cluster module is additionally operable to after the completion of cluster, according to the coordinate of text box The shape that characteristic value is constituted to each class Chinese version frame judges;If what text box was constituted is shaped as preset shape, Such is sent to output module;If text box constitute shape be preset shape, adjustment clustering parameter to such again Secondary cluster.
When the present embodiment is implemented, by taking civil construction cad drawings as an example:The characteristics of civil construction cad drawing Chinese version frames is, If the angle of text box is 0 °, the horizontal coordinate of the first row text box is same or similar, meanwhile, first row text box it is vertical It is same or similar to coordinate, so preset shape is same or similar while first row is literary using the horizontal coordinate of the first row text box The same or analogous shape of longitudinal coordinate of this frame.And in clustering, it sometimes appear that what the text box after cluster was constituted T-shape or circle are shaped as, at this moment, it is not preset shape, and adjust clustering parameter such is clustered again to assert the shape. The present invention effectively raises the accuracy of cluster by the judgement of the shape constituted to text box.
Above-described embodiment, has been carried out further to the purpose of the present invention, technical scheme and beneficial effect Describe in detail, should be understood that the embodiment that the foregoing is only the present invention, be not intended to limit the present invention Protection domain, within the spirit and principles of the invention, any modification, equivalent substitution and improvements done etc. all should be included Within protection scope of the present invention.

Claims (8)

1. the drawing Reading text method based on clustering, it is characterised in that comprise the following steps:
S1:Text box on drawing is subjected to angle classification according to its angle;
S2:Extract the coordinate characteristic value of the text box of same angular type;
S3:Clustering is carried out to the text box of same angular type so that the similar text box of coordinate characteristic value is gathered same In one class, and text box is ranked up according to cluster result;
S4:Text box after sequence is subjected to word output according to the angular type of text frame.
2. the drawing Reading text method according to claim 1 based on clustering, it is characterised in that the coordinate is special Value indicative using the coordinate value in the text box upper left corner, the coordinate value in the lower left corner, the coordinate value in the upper right corner, the coordinate value in the lower right corner or in The coordinate value of heart point.
3. the drawing Reading text method according to claim 1 based on clustering, it is characterised in that the cluster is adopted Use optics algorithms;The optics algorithms determine text by reading the coordinate characteristic value of orderly text box group Chinese version frame Relative distance between frame;The similar standard of the coordinate characteristic value is that relative distance is less than or equal to threshold value.
4. the drawing Reading text method according to claim 1 based on clustering, it is characterised in that step S3 is also wrapped Include following sub-step:
After the completion of cluster, judged according to the shape that the coordinate characteristic value of text box is constituted to each class Chinese version frame;
If what text box was constituted is shaped as preset shape, to such execution S4;
If the shape that text box is constituted is not preset shape, clustering parameter is adjusted to such and S3 is performed.
5. the drawing Reading text system based on clustering of the method using claim 1, it is characterised in that including:
Sort module for the text box on drawing to be carried out to angle classification according to its angle;
Extraction module for the coordinate characteristic value of the text box that extracts same angular type;
Clustering is carried out for the text box to same angular type so that the similar text box of coordinate characteristic value is gathered same In one class, and the cluster module being ranked up according to cluster result to text box;
Output module for the text box after sequence to be carried out to word output according to the angular type of text frame.
6. the drawing Reading text system according to claim 5 based on clustering, it is characterised in that the coordinate is special Value indicative using the coordinate value in the text box upper left corner, the coordinate value in the lower left corner, the coordinate value in the upper right corner, the coordinate value in the lower right corner or in The coordinate value of heart point.
7. the drawing Reading text system according to claim 5 based on clustering, it is characterised in that the cluster is adopted Use optics algorithms;The optics algorithms determine text by reading the coordinate characteristic value of orderly text box group Chinese version frame Relative distance between frame;The similar standard of the coordinate characteristic value is that relative distance is less than or equal to threshold value.
8. the drawing Reading text system according to claim 5 based on clustering, it is characterised in that
Cluster module is additionally operable to after the completion of cluster, and each class Chinese version frame is constituted according to the coordinate characteristic value of text box Shape is judged;
If what text box was constituted is shaped as preset shape, such is sent to output module;
If the shape that text box is constituted is not preset shape, cluster to such adjustment clustering parameter and again.
CN201710187205.9A 2017-03-27 2017-03-27 Drawing text reading method and system based on cluster analysis Active CN107092864B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710187205.9A CN107092864B (en) 2017-03-27 2017-03-27 Drawing text reading method and system based on cluster analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710187205.9A CN107092864B (en) 2017-03-27 2017-03-27 Drawing text reading method and system based on cluster analysis

Publications (2)

Publication Number Publication Date
CN107092864A true CN107092864A (en) 2017-08-25
CN107092864B CN107092864B (en) 2020-02-21

Family

ID=59649050

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710187205.9A Active CN107092864B (en) 2017-03-27 2017-03-27 Drawing text reading method and system based on cluster analysis

Country Status (1)

Country Link
CN (1) CN107092864B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111611936A (en) * 2020-05-22 2020-09-01 青矩技术股份有限公司 Automatic identification system for similar vector diagrams in CAD drawings
CN114998885A (en) * 2022-06-23 2022-09-02 小米汽车科技有限公司 Page data processing method and device, vehicle and storage medium
CN115492311A (en) * 2022-09-19 2022-12-20 同济大学建筑设计研究院(集团)有限公司 Assembly integral type concrete precast column component deepening drawing parametric design method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101408937A (en) * 2008-11-07 2009-04-15 东莞市微模式软件有限公司 Method and apparatus for locating character row
CN102592108A (en) * 2011-01-04 2012-07-18 中国移动通信集团设计院有限公司 Picture label information extracting method and device
CN102693334A (en) * 2012-03-01 2012-09-26 成都鹏业软件股份有限公司 Dynamic construction member identifying method based on CAD (computer-aided design) electronic drawing
CN103279628A (en) * 2013-06-19 2013-09-04 福建龙净环保股份有限公司 Method for manufacturing parameterized engineering drawings
CN104899357A (en) * 2015-05-12 2015-09-09 中山大学 Topology data extraction method based on AutoCAD plane intersection engineering drawing
CN105224706A (en) * 2014-06-30 2016-01-06 上海神机软件有限公司 Based on engineering drawing management system and method, row's modular system and method for workspace

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101408937A (en) * 2008-11-07 2009-04-15 东莞市微模式软件有限公司 Method and apparatus for locating character row
CN102592108A (en) * 2011-01-04 2012-07-18 中国移动通信集团设计院有限公司 Picture label information extracting method and device
CN102693334A (en) * 2012-03-01 2012-09-26 成都鹏业软件股份有限公司 Dynamic construction member identifying method based on CAD (computer-aided design) electronic drawing
CN103279628A (en) * 2013-06-19 2013-09-04 福建龙净环保股份有限公司 Method for manufacturing parameterized engineering drawings
CN105224706A (en) * 2014-06-30 2016-01-06 上海神机软件有限公司 Based on engineering drawing management system and method, row's modular system and method for workspace
CN104899357A (en) * 2015-05-12 2015-09-09 中山大学 Topology data extraction method based on AutoCAD plane intersection engineering drawing

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111611936A (en) * 2020-05-22 2020-09-01 青矩技术股份有限公司 Automatic identification system for similar vector diagrams in CAD drawings
CN111611936B (en) * 2020-05-22 2021-05-14 青矩技术股份有限公司 Automatic identification system for similar vector diagrams in CAD drawings
CN114998885A (en) * 2022-06-23 2022-09-02 小米汽车科技有限公司 Page data processing method and device, vehicle and storage medium
CN115492311A (en) * 2022-09-19 2022-12-20 同济大学建筑设计研究院(集团)有限公司 Assembly integral type concrete precast column component deepening drawing parametric design method
CN115492311B (en) * 2022-09-19 2024-03-19 同济大学建筑设计研究院(集团)有限公司 Parameterized design method for deepening diagram of assembled integral concrete precast column member

Also Published As

Publication number Publication date
CN107092864B (en) 2020-02-21

Similar Documents

Publication Publication Date Title
CN107092864A (en) Drawing Reading text method and system based on clustering
CN104392482B (en) Substation project project three-dimensional calculation amount model generation system and method
CN103942388B (en) Large-scale building information model and three-dimensional digital city integration method
CN105781064A (en) Floor tile laying construction method based on building information modeling (BIM)
CN104850709B (en) Building decoration engineering construction three-dimensional simulation method
CN104695486B (en) Construction method of basement roof polyvinyl chloride root thorn resistance waterproof roll
CN110765520A (en) Parameterized BIM modeling method
CN111177824A (en) Aluminum alloy formwork construction method based on BIM technology
CN106934155A (en) A kind of cable-truss structure looks for shape method
CN111046459A (en) Building model image design method based on 3D software
CN106340066A (en) Building-parameterization three-dimensional modeling method
Hu et al. Batch modeling of 3d city based on esri cityengine
Januszkiewicz et al. Glass as a component of curvilinear architecture in 21st century
CN104963476B (en) Widened decoration aluminum plate for house outer wall
CN105421581A (en) Three-dimension component library of building
Januszkiewicz et al. Parametric architecture in the urban space
KR20240060382A (en) BIM modeling method of tile roof based on architectural principles of traditional wooden buildings
CN106199745A (en) A kind of recognition methods of the pore throat of connection
CN105464263B (en) A kind of large-scale ornamental strip attachment structure of curtain wall
EP3365223A1 (en) Method and system for determining the manufacturing dimensions for a connecting element
CN102324105A (en) The heliogreenhouse three-dimensional modeling method
CN208422247U (en) A kind of spliced architectural design model
CN207032768U (en) A kind of small space house type ground fast spreading structure
Du RETRACTED ARTICLE: Research on roof waterproof construction of agricultural construction engineering based on big data technology
CN105089304B (en) GIS dust-proof sheds

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
PE01 Entry into force of the registration of the contract for pledge of patent right
PE01 Entry into force of the registration of the contract for pledge of patent right

Denomination of invention: Drawing text reading method and system based on cluster analysis

Effective date of registration: 20210927

Granted publication date: 20200221

Pledgee: Bank of Chengdu science and technology branch of Limited by Share Ltd.

Pledgor: CHENGDU UE INFORMATION TECHNOLOGY Co.,Ltd.

Registration number: Y2021510000236

PC01 Cancellation of the registration of the contract for pledge of patent right
PC01 Cancellation of the registration of the contract for pledge of patent right

Date of cancellation: 20230508

Granted publication date: 20200221

Pledgee: Bank of Chengdu science and technology branch of Limited by Share Ltd.

Pledgor: CHENGDU UE INFORMATION TECHNOLOGY CO.,LTD.

Registration number: Y2021510000236