CN102968774B - Computer-assisted mosaicing method for color patterns - Google Patents
Computer-assisted mosaicing method for color patterns Download PDFInfo
- Publication number
- CN102968774B CN102968774B CN201210434381.5A CN201210434381A CN102968774B CN 102968774 B CN102968774 B CN 102968774B CN 201210434381 A CN201210434381 A CN 201210434381A CN 102968774 B CN102968774 B CN 102968774B
- Authority
- CN
- China
- Prior art keywords
- seamed edge
- mosaicing
- splicing
- drawing mould
- color
- 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.)
- Active
Links
Abstract
The invention discloses a computer-assisted mosaicing method for color patterns and belongs to the technical field of computer image processing. The method comprises the steps of scanning, coding, pairing, simulation mosaicing and real-object mosaicing, wherein n scanned mosaicing modules and edges of mosaicing modules are numbered, points are sampled at equal intervals at edges, color codes of sampled points are extracted, data are established, color codes of edges having the same point sampling number are matched to judge whether edges are paired, and real mosaicing is conducted. According to the method, color information of edge characteristic points and edge lengths serve as mosaicing evidences, and initial screening is conducted with edge information length matching serving as the first condition for mosaicing, so that the time for mosaicing computation is saved.
Description
Technical field
The present invention relates to computer image processing field, relate to a kind of area of computer aided joining method of multicolour pattern more specifically.
Background technology
Production and life in, people need some fragments mosaicings to be restored to complete figure, such as sheet metal, restored by the splicing of the planar materials such as the important paper document of shredded paper process and bank note because of maloperation.But splicing Comparision is loaded down with trivial details, its labor capacity increases along with the increase exponentially of number of tiles.Artificial splicing is wasted time and energy.And in fact, most times of splicing spend in compare, try in splicing and the repetitive work such as adjustment.And due to the composition artificially participated in more, often occur because the splicing mistake that causes of the reason such as human factor (fatigue of such as people).
Fragments mosaicing problem has several feature.First, in most of the cases cannot predict the overall picture of last splicing result, and mainly be positioned at chip edges as the important information of splicing foundation; Secondly, when losing without fragment, during splicing, fragment is neither overlapping, also there is not gap; Again, for physical object, during splicing, generally modal geometric warping distortion can not be carried out to fragment module.For the common fragment module with multicolour pattern and straight line seamed edge feature, splicing reality of work can be completed by area of computer aided.Utilize computerized information acquisition and processing technology, the splicing strategy of optimization can be obtained in conjunction with stitching algorithm, thus avoid the comparison of repetition, examination splicing and the tedious work such as adjustment, greatly save the splicing time, improve splicing efficiency and accuracy.
Patent (application number 200910073179.2) proposes a kind of digital jointing method of three-dimensional body fragment of the height map based on fragment section, this method adopts fragment profile height coupling as splicing foundation, therefore three-dimensional body fragment is applicable to, but for there is fragment module the method for multicolour pattern and straight line seamed edge feature and inapplicable.
Summary of the invention
The object of this invention is to provide a kind of area of computer aided joining method for splicing the module with multicolour pattern and straight line seamed edge.
Technical scheme of the present invention is: comprise the following steps
1) scan; Colored for n block Nesting Drawing Mould is scanned under the same conditions, to colored Nesting Drawing Mould corresponding with the coloured image of scanning from 1 to n number consecutively;
2) encode; To all seamed edges in the coloured image of colored Nesting Drawing Mould and scanning from 1 to m
inumber consecutively, wherein i=1,2,3 ..., n; Then get a little at each seamed edge by d at equal intervals, to get a rule be first is the mid point of each seamed edge, then symmetrically successively gets a little to first both sides, and extracting the color coding of each point, is this
bar seamed edge is encoded, and building database, described database comprises each point color coding on seamed edge numbering, the colored Nesting Drawing Mould numbering in seamed edge place, splicing successful matching mark, seamed edge;
3) match; First divide into groups according to getting to count out to all seamed edges, the identical seamed edge of counting out of getting divides at same group; Then respectively splicing pairing is carried out to the seamed edge in each group from more to less according to getting to count out; Described splicing pairing is whether the color coding of taken point relative on judge two seamed edge is consistent, and marks the seamed edge of successful matching;
4) picture mosaic is simulated; Carrying out simulation picture mosaic according to matching mark in step 3) and check, as do not reached requirement, after adjustment d value, returning step 2);
5) picture mosaic in kind; The result of step 4) splices colored Nesting Drawing Mould;
Further, described step 2) in be that centered by taken point, R value delimit a region for radius, gets average, obtain average color coding to the color value of all picture elements in this region to color-coded extracting method.
The invention has the beneficial effects as follows:
1) this method is except adopting the colouring information of Edge Feature Points as except splicing foundation, simultaneously using edge length as splicing foundation, the chip edges information of different length is obtained according to edge length, screen for the first time as the most important condition of splicing using marginal information length coupling, thus save the time of splicing calculating.
2) splicing efficiency is drastically increased.Adopt the method, splicing has been come by the computing machine of pre-set stitching algorithm, eliminates the process that artificial comparison, examination splicing and adjustment etc. are wasted time and energy.People is broken away from, and loaded down with trivial details splicing is worked, and greatly reduces the splicing mistake caused because of reasons such as human factors (fatigue of such as people), improves the order of accuarcy of splicing.
Accompanying drawing explanation
Table 1 seamed edge gets a number order schematic diagram
Table 2 splices the seamed edge database table before calculating
Table 3 splices the seamed edge database table after calculating
Fig. 1 operational flowchart;
Fig. 2 seamed edge gets a number order schematic diagram;
Fig. 3 Nesting Drawing Mould (wherein the numbering of I, II, III 3 modules is respectively 1,2,3);
Fig. 4 module has spliced figure;
Scheme Chinese and English letter representation different colours, B: blue; G: green; K: black; R: red; Y: yellow.
Embodiment
Its operational flowchart is as Fig. 1.Specific implementation process can be divided into 5 large steps:
1, scan.Suppose to need the colored Nesting Drawing Mould of splicing to have n block, every block Nesting Drawing Mould has m
ibar seamed edge (i=1,2,3 ..., n).First this n block Nesting Drawing Mould is scanned successively under same scan condition, be converted into the coloured image of digitized format, and identical sequence number is marked successively to these colored Nesting Drawing Mould and coloured image simultaneously in material object and computing machine.Pay special attention to the seamed edge edge color of Nesting Drawing Mould and the accuracy of length information collection.
2, encode.Every bar seamed edge of the n block Nesting Drawing Mould gathered (is total to
bar seamed edge) number consecutively from 1, then carry out equi-spaced apart (such as spacing distance d is taken as 1 mm) and get a little, first point is the mid point of every bar seamed edge, and all the other are got a little to both sides successively.For this
bar seamed edge is encoded, and building database.Database comprises following information: the numbering i of this seamed edge, the numbering j of this seamed edge place Nesting Drawing Mould, the splicing successful matching whether initial value of mark flag(are all set to 0), the RGB color coding [a of each point successively on seamed edge, b, c] (be first point with seamed edge mid point), and store in a database and become a string color coding sequence.
Table 1 coded data storage format
Its storage order is with reference to shown in table 1, and first memory location saves the number information of seamed edge; Second memory location saves the number information of this seamed edge place Nesting Drawing Mould; 3rd memory location saves splicing successful matching whether flag information; 4th and subsequent memory location save the RGB color-coded information of each point on seamed edge.Fig. 2 illustrates the number order of each point on seamed edge, and first point is positioned at seamed edge mid point, and second point is positioned at first some lower left side, 3rd point is positioned at first some upper right side, 4th point is positioned at first some lower left side, and the 5th point is positioned at first some upper right side, the like.Further, adjacent distance between two points is equal.The numbering i of seamed edge is marked at by the Nesting Drawing Mould that scans simultaneously, so that according to computer picture-arrangement result splicing electronic objects units.
In order to improve efficiency and the accuracy of splicing pairing, the method of improvement can also be adopted, namely after on seamed edge, the position of each point determines, centered by each point, R value delimit a region (being a half-circle area when seamed edge edge is straight line) for radius, and get average to after the color coding of all picture elements in this region, obtain average color coding [A, B, C].Wherein
,
,
.Here p is the number of all picture elements in this region, a
i, b
i, c
ibe respectively the RGB color coding of each pixel in region.
3, match.First divide into groups according to getting to count out to all seamed edges, the identical seamed edge of counting out of getting divides at same group.Then carry out splicing pairing to the seamed edge in each group respectively from more to less according to getting to count out, namely first many groups of counting out are got in splicing, and few group of counting out is got in rear splicing.For a seamed edge in any a group, if the flag information in database is 0, then successively with same group in flag information be 0 seamed edge compare, more whether there is same color coded sequence, thus to judge whether and this seamed edge matches.In each comparison procedure, if judge not mate, then the order of the color coding sequence of the wherein seamed edge compared is readjusted, namely keep mid point color coding still the 1st position, 2nd, the 3rd color coding order is exchanged, 3rd, the 4th color coding order is exchanged, and the 5th, the 6th color coding order is exchanged ...Avoid with this because the order of color coding sequence of the both sides of mid point is reversed record and situation about can not splice smoothly on seamed edge.Splice the flag memory location in a database of the seamed edge successfully and be labeled as the seamed edge numbering of matching with it, otherwise be still set to 0.Adopt same procedure to splicing the seamed edge in other group successively.
4, figure is modeled to.According to the mark shown by pairing result, adopt computing machine to carry out simulation picture mosaic, check actual picture mosaic effect, judge whether picture mosaic effect is satisfied with.Dissatisfied then get back to step 2, need, carry out the operation cycle of a new round, namely adjust d value and R value, re-start next round and encode, match, be spliced into figure step; Satisfied then stop, complete and be spliced into figure work.
5, picture mosaic in kind.According to simulating the result of picture mosaic to carrying out colored Nesting Drawing Mould splicing.
Embodiment 1
The area of computer aided joining method of multicolour pattern of the present invention, for splicing the module with multicolour pattern and straight line seamed edge.Below in conjunction with embodiment, illustrate its operation steps:
1, scan.As in Fig. 3, always have the colored Nesting Drawing Mould that 3 pieces need splicing.Be numbered it, in figure, the numbering of I, II, III 3 modules is respectively 1,2,3.Nesting Drawing Mould shape is all isosceles right triangle, wherein the right angle length of side 13.8 cm of module 1, hypotenuse long 27.6 cm, the right angle length of side 13.8 cm of module 3 of module 2.Be numbered the seamed edge of each Nesting Drawing Mould successively, concrete numbering as shown in Figure 3.These 3 pieces of Nesting Drawing Mould are scanned successively under same scan condition, is converted into the color graphics of digitized format.Pay special attention to the edge color of Nesting Drawing Mould and the accuracy of length information collection.
2, encode.Carry out number consecutively from 1 to every bar seamed edge (totally 9 seamed edges) of the 3 pieces of Nesting Drawing Mould gathered, then carry out equi-spaced apart and get a little, spacing distance is d=2 cm.First point is the mid point of every bar seamed edge, and all the other get a little to both sides successively, obtains the color coding sequence of these 9 seamed edges, and building database, as shown in table 2.In addition, the numbering 1 ~ 9 of seamed edge is marked at by the Nesting Drawing Mould that scans simultaneously, so that according to computer picture-arrangement result splicing electronic objects units.
Table 2 splices the seamed edge database table before calculating
In order to improve efficiency and the accuracy of splicing pairing, adopt the method improved, namely after on seamed edge, the position of each point determines, centered by each point, R=1 mm value delimit a region (being a half-circle area when seamed edge edge is straight line) for radius, and get average to after the color coding of all picture elements in this region, obtain average color coding [A, B, C].Wherein
,
,
.Here p is the number of all picture elements in this region, a
i, b
i, c
ibe respectively the RGB color coding of each pixel in region.
3, match.First divide into groups according to getting to count out to all seamed edges, the identical seamed edge of counting out of getting divides at same group.Like this, 9 seamed edges can be divided into 3 groups, first group: seamed edge 5; Second group: seamed edge 3,4,6,7; 3rd group: seamed edge 1,2,8,9.Then respectively splicing pairing is carried out to the seamed edge in each group from more to less according to getting to count out.The seamed edge first matched in first group, pairing result: the seamed edge not having successful matching; Match second group again, pairing result: seamed edge 3 and 4 can match, and seamed edge 6 and 7 seamed edge also can successful matching after changing the order of color coding sequence wherein.Flag information in the seamed edge of change successful matching; Finally the seamed edge in the 3rd group is matched, pairing result: do not have the seamed edge that can match.Final pairing result is as shown in table 3.
Table 3 splices the seamed edge database table after calculating
4, figure is modeled to.Number according to the seamed edge of computer picture-arrangement result (table 3) and Nesting Drawing Mould, each Nesting Drawing Mould of simulation splicing, namely seamed edge 3 and 4 splices, and seamed edge 6 and 7 splices, and formation image as shown in Figure 4, this picture mosaic meets the requirements on inspection.
5, picture mosaic in kind.According to the result of simulation picture mosaic, finally carry out picture mosaic in kind.
Claims (5)
1. the area of computer aided joining method of multicolour pattern, is characterized in that comprising the following steps:
1) scan; Colored for n block Nesting Drawing Mould is scanned under the same conditions, to colored Nesting Drawing Mould corresponding with the coloured image of scanning from 1 to n number consecutively;
2) encode; To all seamed edges in the coloured image of colored Nesting Drawing Mould and scanning from 1 to
number consecutively, supposes that total seamed edge number of i-th Nesting Drawing Mould is m
i, wherein i=1,2,3 ..., n; Then get a little at each seamed edge of the coloured image of scanning by d at equal intervals, to get a rule be first is the mid point of each seamed edge, then symmetrically successively gets a little to first both sides, extract the color coding of each point, is the coloured image of this scanning
bar seamed edge is encoded, and building database;
3) match; First divide into groups according to getting to count out to all seamed edges, the identical seamed edge of counting out of getting divides at same group; Then respectively splicing pairing is carried out to the seamed edge in each group from more to less according to getting to count out; Described splicing pairing is whether the color coding of taken point relative on judge two seamed edge is consistent, and marks the seamed edge of successful matching;
4) picture mosaic is simulated; According to step 3) in pairing mark carry out simulation picture mosaic and check, as do not reached requirement, after adjusting d value, return step 2);
5) picture mosaic in kind; Step 4) result splice colored Nesting Drawing Mould.
2. the area of computer aided joining method of multicolour pattern according to claim 1, it is characterized in that, described step 2) in color-coded extracting method be, centered by taken point, R value delimit a region for radius, average is got to the color value of all picture elements in this region, obtains average color coding.
3. the area of computer aided joining method of multicolour pattern according to claim 1, it is characterized in that, described step 2) database comprise each point color coding on seamed edge numbering, the colored Nesting Drawing Mould numbering in seamed edge place, splicing successful matching mark, seamed edge.
4. the area of computer aided joining method of multicolour pattern according to claim 2, is characterized in that, in described step 4) simulation picture mosaic; According to step 3) in pairing mark carry out simulation picture mosaic and check, as do not reached requirement, after adjusting d value and R value, return step 2).
5. the area of computer aided joining method of multicolour pattern according to claim 3, is characterized in that, described step 3) mark be labeled on the splicing successful matching mark of database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210434381.5A CN102968774B (en) | 2012-11-05 | 2012-11-05 | Computer-assisted mosaicing method for color patterns |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210434381.5A CN102968774B (en) | 2012-11-05 | 2012-11-05 | Computer-assisted mosaicing method for color patterns |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102968774A CN102968774A (en) | 2013-03-13 |
| CN102968774B true CN102968774B (en) | 2015-06-24 |
Family
ID=47798901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210434381.5A Active CN102968774B (en) | 2012-11-05 | 2012-11-05 | Computer-assisted mosaicing method for color patterns |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102968774B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103700081B (en) * | 2013-12-17 | 2016-08-17 | 河海大学 | A kind of shredder crushes the restoration methods of English document |
| CN104537368B (en) * | 2015-01-07 | 2018-10-09 | 北京工业大学 | One kind being crushed file reset analysis method for English printing word duplex printing |
| CN108628920A (en) * | 2017-11-13 | 2018-10-09 | 淄博职业学院 | A kind of Art Design internet auxiliary puzzle system and design method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008099269A3 (en) * | 2007-02-14 | 2009-06-11 | Photint Venture Group Inc | Banana codec |
| CN101499166A (en) * | 2009-03-16 | 2009-08-05 | 北京中星微电子有限公司 | Image splicing method and apparatus |
| CN102467738A (en) * | 2010-11-08 | 2012-05-23 | 北京大学 | Image splicing method and system |
-
2012
- 2012-11-05 CN CN201210434381.5A patent/CN102968774B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008099269A3 (en) * | 2007-02-14 | 2009-06-11 | Photint Venture Group Inc | Banana codec |
| CN101499166A (en) * | 2009-03-16 | 2009-08-05 | 北京中星微电子有限公司 | Image splicing method and apparatus |
| CN102467738A (en) * | 2010-11-08 | 2012-05-23 | 北京大学 | Image splicing method and system |
Non-Patent Citations (3)
| Title |
|---|
| a faster image registration and stitching algorithm;Chaminda Namal Senarathne等;《2011 6th International Conference on Industrial and Information Systems》;20110816;66-69 * |
| 计算机自动拼图中颜色匹配算法的研究与应用;姬照中;《计算机应用与软件》;20090731;第26卷(第7期);68-71 * |
| 边缘颜色匹配算法在计算机拼图中的应用;支玲玲;《计算机工程与设计》;20101231;3705-3707,3711 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102968774A (en) | 2013-03-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105547151B (en) | Point Cloud Data from Three Dimension Laser Scanning acquires the method and system with processing | |
| CN109447994A (en) | In conjunction with the remote sensing image segmentation method of complete residual error and Fusion Features | |
| CN108428229A (en) | It is a kind of that apparent and geometric properties lung's Texture Recognitions are extracted based on deep neural network | |
| CN104077447B (en) | Urban three-dimensional space vector modeling method based on paper plane data | |
| CN109711413A (en) | Image, semantic dividing method based on deep learning | |
| CN104680531B (en) | A kind of connection amount statistical information extracting method and VLSI structure | |
| CN107590772A (en) | A kind of cultural relic fragments method for automatically split-jointing based on adaptive neighborhood matching | |
| CN105069746A (en) | Video real-time human face substitution method and system based on partial affine and color transfer technology | |
| CN106204503B (en) | Based on the image repair algorithm for improving confidence level renewal function and matching criterior | |
| CN106127771A (en) | Tunnel orthography system and method is obtained based on laser radar LIDAR cloud data | |
| CN105975655B (en) | A kind of imitative Tang and Song Dynasty ancient architecture abnormity roofing tile parametric modeling method based on BIM | |
| CN102968774B (en) | Computer-assisted mosaicing method for color patterns | |
| CN109711099A (en) | A kind of BIM automatic modeling system based on the study of image recognition machine | |
| CN101840582B (en) | Boundary digitizing method of cadastral plot | |
| CN106501090A (en) | Crack characterizing method for hydraulic fracturing simulation experiment | |
| CN108460833A (en) | A kind of information platform building traditional architecture digital protection and reparation based on BIM | |
| CN106355607A (en) | Wide-baseline color image template matching method | |
| CN103996180A (en) | Paper-shredder broken-document restoration method based on English character characteristics | |
| CN105069395A (en) | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology | |
| CN103295019B (en) | A kind of Chinese fragment self-adaptive recovery method based on probability-statistics | |
| CN106339985B (en) | A method of selection is inlayed line and is inlayed to aviation image from vector house data | |
| CN103236056B (en) | Based on the image partition method of template matches | |
| CN105046632A (en) | Space-time efficient binary image binary logic operation method | |
| CN104978466A (en) | Three-dimensional definite-number bar arrangement method used for multiple guide wires | |
| CN113159052B (en) | Method for identifying failure mode of flexural reinforced concrete simply supported beam based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201119 Address after: 224000 building 202, building 57, shuanglongxing village, Hede Town, Sheyang County, Yancheng City, Jiangsu Province Patentee after: Chen Guoquan Address before: Gehu Lake Road Wujin District 213164 Jiangsu city of Changzhou province No. 1 Patentee before: CHANGZHOU University |
|
| TR01 | Transfer of patent right |