CN101635055A - Contour extraction device and method thereof - Google Patents
Contour extraction device and method thereof Download PDFInfo
- Publication number
- CN101635055A CN101635055A CN200910160926A CN200910160926A CN101635055A CN 101635055 A CN101635055 A CN 101635055A CN 200910160926 A CN200910160926 A CN 200910160926A CN 200910160926 A CN200910160926 A CN 200910160926A CN 101635055 A CN101635055 A CN 101635055A
- Authority
- CN
- China
- Prior art keywords
- fold
- profile
- sheeting
- workpiece
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title abstract description 10
- 238000005520 cutting process Methods 0.000 claims description 40
- 239000000284 extract Substances 0.000 claims description 28
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000037303 wrinkles Effects 0.000 abstract 8
- 238000007519 figuring Methods 0.000 abstract 2
- 239000010985 leather Substances 0.000 description 55
- 238000012552 review Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 9
- 238000003860 storage Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 6
- 230000015654 memory Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000033228 biological regulation Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Treatment And Processing Of Natural Fur Or Leather (AREA)
Abstract
Contour extraction device and method thereof, which extracting contour points according to images of sheet materials photoed by camera, then figuring out outside contour represening shape of the sheet materials and inner side contour located therein. Area between the inner and the outer contour is used as wrinkle, then figuring out ratio of long diameter and short diameter of shape of the wrinkle, and wrinkle with bigger ratio is used as a strong wrinkle and the smaller one is used as a weak wrinkle. The cut workpieces are distributed onto the sheet materials according to grade of the wrinkles. The invention can figure out shape of the wrinkle and can utilize wrinkle part effectively.
Description
Technical field
The present invention relates to from sheetings such as leather, cloth, extract the technology of profile.
Background technology
The applicant is devoted to develop the Scissoring device of sheeting.In patent documentation 1 (WO2004/010807), inquired into respect to sheeting and disposed the technology of answering the cutting workpiece best.In patent documentation 2 (WO2006/051764), inquired into respect to erose sheetings such as leathers how to dispose workpiece as the cutting target.
The inventor has inquired into the fold of sheeting.The example of having represented the fold of leather among Figure 18,102 is strong fold, does to go up mark with belt 104,106,108 is incompetent fold.Produce fold and be since cutting originally for the leather of 3D shape launches on the worktable of two dimension, for example the situation that heave in the somewhere during configuration on worktable with the sheeting of cutting balloon-like or spherical three-dimensional shape is the same.At the position that has fold, also have the character situation different of leather with other parts, also can heave thus owing to fold causes leather, therefore there is the problem that is difficult to correct shape cutting.The inventor has inquired into the shape of obtaining fold, and the technology that the part of fold also can be utilized as workpiece of the degree by estimating fold, has proposed the present invention.
Patent documentation 1:WO2004/010807
Patent documentation 2:WO2006/051764
Summary of the invention
Problem of the present invention is: the shape by obtaining fold and the degree of fold, also can utilize the part that has fold of sheeting.
The problem of appending of the present invention is: the shape that can obtain fold expeditiously.
The problem of appending of the present invention is: can more effectively utilize the part that has fold.
The present invention proposes a kind of device, and it is exported as the outline line of sheeting by extracting point the image that filters the sheeting of taking from video camera, connect the point extracted, it is characterized in that, comprising:
The profile extraction unit, the nearside wheel profile that is used to obtain the outboard wheel profile of the profile of representing sheeting and is positioned at the inside of described outboard wheel profile;
Be used for obtaining the unit of shape of the fold of sheeting according to described outboard wheel profile and described nearside wheel profile;
The crimp shape evaluation unit is used to obtain the amount of major diameter and the ratio of minor axis of the shape of the fold that expression obtains;
Stage unit, the fold that the amount of described expression ratio is bigger are as strong fold, and the fold that the amount of expression ratio is less carries out classification with this to fold as incompetent fold; And
Allocation units are used for and will be assigned to sheeting according to the classification of fold from the workpiece of sheeting cutting.
In addition, the present invention proposes a kind of method, exports as the outline line of sheeting by extracting point the image that filters the sheeting of taking from video camera, connect the point extracted, it is characterized in that,
By profile the unit is proposed, obtain the expression sheeting profile the outboard wheel profile and be positioned at the nearside wheel profile of the inside of described outboard wheel profile,
The shape of the fold of sheeting according to described outboard wheel profile and described nearside wheel profile, is obtained in the unit of the shape by being used to obtain fold,
By the crimp shape evaluation unit, obtain the amount of major diameter and the ratio of minor axis of the shape of the fold obtained of expression,
By stage unit, the fold that the amount of described expression ratio is bigger is as strong fold, and the fold that the amount of expression ratio is less carries out classification as incompetent fold to fold,
By allocation units, will be assigned on the sheeting from the classification of the workpiece of sheeting cutting according to fold.
In addition, in this instructions, also be directly applied for contour extraction method, also be applicable to contour extraction apparatus about the record of contour extraction method about the record of contour extraction apparatus.
In the present invention, the shape of obtaining fold according to the outboard wheel profile and the nearside wheel profile of sheeting.Under the situation that the original outline line of fold and sheeting joins, be fold by outboard wheel profile and nearside wheel profile area surrounded, the outboard wheel profile is the original outline line of sheeting.In addition, isolate under the situation about existing in the inside of sheeting at fold, nearside wheel profile closure, its inboard is a fold.
According to the shape of fold obtain fold degree, be the power of fold.Elongated fold is generally sharp-pointed strong fold, for example the such fold of the fold 102 of Figure 18.Wide cut be the little mild incompetent fold that rises and falls near circular fold, as the fold of the fold 106,108 of Figure 18.According to the major diameter of fold and the ratio of minor axis, or the data that are equal to therewith, ratio of the major diameter of the ratio of the major diameter of fold and the girth of fold or fold and 1/2 power of the minor axis of fold etc. is for example obtained the degree of fold.Major diameter is the inverse of pi in the fold of circle with the ratio of the girth of fold, in elongated fold near 1/2.In the ratio of the major diameter of fold and 1/2 power of the minor axis of fold, elongated fold is as strong fold, and under the similar situation of the shape of fold, big fold is as stronger fold.If carry out above processing fold is carried out classification, then can in the scope corresponding, fold be used for workpiece with classification.Therefore in the present invention, can obtain the shape of fold, and the part of fold can be assigned on the workpiece also and effectively utilizes.
Preferably, in described profile extraction unit, extract outline line laterally from the inside of sheeting, to be in the inside and outside colour difference of outline line outline line below the rated condition as described nearside wheel profile, and the outline line of more lateral that is in the nearside wheel profile is as described outboard wheel profile.
Whether the extraction of profile is for example carried out to inside from the outside of sheeting, for example can be with the outline line of the closure extracted at first as the outboard wheel profile, by existing outline line to extract the nearside wheel profile to its inboard search.But in this case, no matter sheeting inside and outside, for example need be for the whole surface extraction outline line of the worktable that launches sheeting.
Relative therewith, if from the internal extraction outline line of sheeting, then exist following any may: the initial outline line that extracts is the nearside wheel profile, has the outboard wheel profile in its outside, perhaps from beginning to extract the outboard wheel profile at first.Then, in the nearside wheel profile corresponding with fold, the variation of brightness is bigger, but colour differences such as form and aspect are less.Relative therewith, in the outboard wheel profile corresponding, move to the outside from the inside of sheeting, so colour difference is also bigger with the profile of sheeting.Therefore, be that outline line below the rated condition is as the candidate of described nearside wheel profile with inside and outside colour difference.And, extract the outboard wheel profile if search for its outside, then can limit the scope of search outline line, and can before extracting the outboard wheel profile, prediction be nearside wheel profile or outboard wheel profile, therefore can extract the inside and outside contour line efficiently.
More preferably, make that size a plurality of polygons little, that size is big in low density zone produce based on described point in the high zone of the density of point, thus, the space segmentation unit further is set is used for producing these a plurality of polygonal images, and,
In described profile extraction unit, begin to search for described polygonal image from the point of the inside of sheeting towards described polygonal limit, if touch with the edge joint of curtailment specified length, then end to cross the search on this limit and store this limit, if with length be that edge joint more than the specified length touches, then crossing this length is that search is continued on the above limit of specified length.
The point that also can link simply by filtering extraction produces outline line, but under the unsharp situation of outline line, can not guarantee to extract outline line accurately.For example the R1 of Figure 10 is a leather, and R2 is the background of worktable, and L1, L2 are profile, because the thickness of leather R1, the top limit of leather constitutes profile L1, and the limit of bottom surface side constitutes profile L2.Herein, contour identification L1 and profile L2, select profile L1 to visual be simple thing, but utilize computing machine to use image recognition and automatically be not nothing the matter.
Therefore, preferably, from the inside of object, towards the polygonal limit search that generates based on point.Because the peripheral outline point at outline line distributes thick and fast,, if run into, then end to cross the search on this limit, the limit of the not enough specified length of memory length with the limit of curtailment specified length so polygonal size is less.If arrive the limit of length more than specified length, then cross this limit and continue search.Therefore,, then end search, link the initial outline line that the limit that extracts then obtains object if search a plurality of little polygonal limit that on outline line, distributes.This outline line is outboard wheel profile or nearside wheel profile.Carry out above-mentioned processing then, with colour difference be outline line below the rated condition as the candidate of nearside wheel profile, if continue its outside of search, then can extract the outboard wheel profile.
In addition, preferably, the permission of the region allocation fold in described workpiece or workpiece, described allocation units distribute workpiece according to described classification and described permission in fold.
In workpiece, have important workpiece and unessential workpiece in the design, to the permission of form accuracy etc., promptly to the permission of fold also according to workpiece and difference.Therefore,, then can in the scope of permission, fold be used on workpiece, can effectively utilize sheeting if fold is carried out classification and divides the permission of matching fold to workpiece.
Under the situation that has tape edge etc. on the workpiece, to the requirement of the form accuracy of workpiece, quality according to the zone in the workpiece and difference.Therefore, if specify permission, then can more effectively utilize the part of fold to fold according to the zone in the workpiece.
Description of drawings
Fig. 1 is the block diagram that the hardware of the Scissoring device of expression embodiment constitutes.
Fig. 2 uses the figure that extracts the Scissoring device of function associated module table diagrammatic sketch 1 with profile.
Fig. 3 is the figure of the example of the expression profile filtrator that extracts usefulness.
Fig. 4 is the profile of expression among the embodiment and the figure of the extraction algorithm of fold.
Fig. 5 is the figure of the grade additional algorithm of the fold among the expression embodiment.
Fig. 6 is the figure of the grade additional algorithm of the workpiece among the expression embodiment.
Fig. 7 represents that schematically De Laonei cuts apart the figure of (Delaunay division).
Fig. 8 is the figure that schematically represents Wei Nuotu (Voronoi diagram).
Fig. 9 is the figure of the bianry image of expression leather.
Figure 10 is the figure of the image of expression point.
Figure 11 is the figure that the expression De Laonei corresponding with Fig. 9 cut apart.
Figure 12 is the figure that schematically represents the extraction of the fold among the embodiment.
Figure 13 is the figure of relation that schematically represents the grade of the aspect ratio of the fold among the embodiment and fold.
Figure 14 schematically is illustrated in the figure that there is the leather of fold in the inboard.
Figure 15 is the figure that schematically is illustrated in the situation that form accuracy reduces in the workpiece of cutting open fold.
Figure 16 is the figure that schematically represents each the regional classification in the workpiece.
Figure 17 is the figure that schematically represents to search for from the outside of leather the variation of profile and fold.
Figure 18 is the photo that there is the leather of fold in expression.
Label declaration
2 Scissoring devices
4 cutting worktable
6 cutting heads
8 colour TV cameras
10 cutters
12 X-axis drivers
14 Y-axis drivers
15 leathers
16 colored ribbons
18 contour extraction apparatus
20 buses
21 colour picture monitors
22 manually inputs
23 disc drivers
24 network interfaces
26 input and output
28 processors
30 storeies
32 program storages
34 camera review correction units
36 video memories
38 filtrators
40 file memories
42 merge handling part
44 polygon cutting parts
46 in-profile extraction units
48 exterior contour extraction units
50 fold classification portions
52 workpiece memory portions
54 workpiece matching parts
A 56 cuttings data generating unit
60 Laplce's filtrators
61,62 differential filtrators
80,90 leathers
81 profiles
82,91 inboard profiles
83～87 zones
88 folds
89,92 lateral profiles
110 inboard profiles
112,114 workpiece
120 workpiece
130 leathers
132,133 folds
134 lateral profiles
136 inboard profiles
L1, L2 profile
P1～P6 summit
The R1 leather
The R2 background
R3 is dirty
The a major diameter
The b minor axis
Embodiment
Below explanation is in order to implement most preferred embodiment of the present invention.Among the embodiment, outline line only is called profile.
[embodiment]
In Fig. 1～Figure 17, be example with Scissoring device 2, represented contour extraction apparatus 18, and the embodiment of contour extraction method.Among Fig. 1,4 are the cutting worktable, and 6 for the cutting head, utilizes the coloured image of 8 pairs of leathers 15 of colour TV camera to take, and, for example make 10 rotations of circular cutter and the disconnected leather 15 of cutting.Cutting 6 utilizes X-axis driver 12 and Y-axis driver 14 to move in two directions of cutting worktable 4 upper edge XY.Then, colour TV camera 8 utilizes cutting 6 scanning on leather 15, the image segmentation of leather 15 is become polylith and takes.
28 is processor, and 30 is general-purpose storage, and 32 is program storage, the control program of Scissoring devices 2 such as storage profile extraction procedure.Utilize processor 28 or storer 30,32 etc., contour extraction apparatus 18 acts on as shown in Figure 2.Camera review correction unit 34 is proofreaied and correct the distortion of the image that obtains from video camera 8.Raster images such as camera review are stored in the video memory 36, and the image of the leather integral body of 38 pairs of 1 camera reviews of filtrator or synthetic polylith camera review filters, and extract point.The point that file memory 40 storages are extracted, the content of data is the coordinate of point.Merge the point merging that handling part 42 for example will utilize filtrator 38 to extract from each camera review, form the contour point data of 1 block leather.
Exterior contour extraction unit 48 is extracted in the profile of the outside existence of inboard profile.Utilize exterior contour extraction unit 48 to estimate inside and outside inboard profile, colourity, for example form and aspect and chroma, or whether the Cb signal in the picture signal of Y, Cb, Cr and two signals of Cr signal exist the above difference of rated condition.As the difference more than the rated condition, adopt bigger value in the absolute value of the absolute value sum of difference of the absolute value of the difference of Cb signal for example and Cr signal or these differences etc. whether more than the value of regulation.Replaced C b signal and Cr signal also can adopt H signal in the HLS signal and S signal etc.In addition, the colour difference of replaced C b signal and Cr signal etc. also can adopt aberration in rgb space etc., be the distance in the rgb space.And, under the inside and outside colourity of inboard profile is changed to situation below the rated condition, be speculated as the profile that extracted fold, the search lateral profile.Change under the situation about being in more than the rated condition in colourity,, for example no longer further search for laterally as the image that extracted lateral profile.Profile extraction unit 46,48 will for example link as the vector on the limit of the profile of the sheeting that extracts, and the row of vector data are stored as inside and outside contour.
Fold classification portion 50 carries out classification according to the aspect ratio of the shape, particularly fold of fold to the degree of fold and stores.The shape of the workpiece that workpiece storage part 52 storage should be taken out from leather 52 by cutting and to the permission of fold, the permission of fold can be stored with workpiece unit, also can store corresponding to each zone of cutting apart the workpiece gained.Workpiece matching part 54 will utilize the classification of the part of shape that the inside and outside contour of sheeting determines and fold, compare with the shape of the workpiece that should take out and to the permission of fold, produce best cutting pattern.A cutting data generating unit 56 produces the control data of cutting head according to the cutting pattern that is produced by workpiece matching part 54.
The example of having represented filtrator 38 among Fig. 3.60 is Laplce's filtrator, extracts data and on every side the different pixel of pixel, and 61,62 is the differential filtrator, is suitable for extracting the direction of image change.In addition, under the significant situation of the noise of camera review, adopting before filtrator 60～62 handles, also can adopt median filter etc. to handle, removing denoising.The kind of the filtrator that adopts itself is arbitrarily, camera review is implemented filtered, and carries out binaryzation and extracts point.In the extraction of point, the pixel of brightness or colourity sudden change as point, but also can only be extracted the pixel of jump in brightness.
The extraction algorithm of having represented inside and outside contour among Fig. 4.With respect to the fixing leather 15 of cutting 6, then colored ribbon 16 is sticked at the flaw place on leather if necessary.Make cutting 6 to the motion of XY direction then, the portrait of leather 15 for example is divided into tens of and take (S1).Utilize camera review correction unit 34, the distortion of correcting captured image (S2) is extracted point for brightness and colourity respectively.In the extraction of point, utilize for example filter process camera review of Fig. 3, with the image binaryzation after handling, with the pixel of the value more than the setting as point (S3).Obtain point for brightness and colourity respectively, the point of brightness and the point of colourity are all extracted as point, but also can not extract the point of colourity.Then each image of taking is repeated above processing (S4).
The point of gathering 1 block leather, utilization are the image (S6) that the triangle on summit is cut apart sheeting with the point.Image herein is actually the data that only are made of point.In leg-of-mutton generation, near the summit being in constitutes leg-of-mutton limit thus than the preferential combination in the summit in a distant place.In addition, determine to link which summit in the leg-of-mutton mode of avoiding having little interior angle.In other words, note interior angle minimum in leg-of-mutton 3 interior angles, avoid the less triangle of this interior angle, for example avoid having the triangle of the following interior angle of setting.
Above algorithm is oversimplified can be as described below.Be conceived to a summit, the summit of vicinity in the angular range of the regulation around being connected.Repeating this processing in 360 ° of scopes around the summit of being had in mind, then summit and a plurality of summits on every side link.Above processing is repeated on each summit, then image is cut apart by triangle.And vertex of a triangle is a point.
Then specify 1 point (S7) in the leather.This point can be specified by manual input, perhaps also can utilization with background color, be the different color of color of cutting worktable 4, extract the point of big leg-of-mutton inside by computing machine.Search for from the mode that 1 point of appointment is propagated to cross leg-of-mutton limit.Then, if with length be that the long above edge joint of regulation touches, then cross this limit and search for leg-of-mutton all the other 2 limits.As described above, enlarge the zone (S8) corresponding with the inside of sheeting.
If run into leg-of-mutton minor face, then stop to cross the search on this limit.Certainly search (S9) is proceeded on other limits.Continuing to search for the threshold value that still stops search is constant in an embodiment, but also can be variable.Continue search then till the limit of search has not had (S10).In above process, under the situation about stopping search, store this limit running into minor face.Because the data on limit are vector datas, so the starting point on storage limit and the coordinate of terminal point.The limit that connects storage then obtains the closed contour of leather, the inboard profile (S11) of Here it is leather.
Whether inspection changes more than rated condition (S12) in the inside and outside colourity of inboard profile, under the situation of colourity deficient change rated condition, crosses inboard profile, and then extracts lateral profile (S13).This processing is the same with S9～S11.Then, will be by the inboard of the inboard profile under himself closed individually situation of lateral profile and inboard profile institute's area surrounded and inboard profile as fold (S14).
Fig. 7 represents the example according to the leg-of-mutton search of De Laonei.Figure intermediate cam shape summit is a point, under with visual situation, can see having two profiles of L1, L2 and isolated point P1, P2.But it is difficult automatically carrying out this identification.In Fig. 7, begin to search for from the upper left side.Double-deck line among the figure is represented the minor face that stops search, if run into minor face then stop to cross the search on limit.Link the minor face that stops search then, then obtain profile.In order to extract profile exactly, below 2 outbalances: approaching point is interconnected; Avoid having the triangle of little interior angle.And, as the algorithm that when satisfying above-mentioned condition, comes partition space, cut apart in the known virtuous labor according to triangle.
Fig. 8 represents according to the polygonal space segmentation of dimension promise (Voronoi).For example with respect to point P1, produce the perpendicular bisector that is connected formed limit with point on every side, be connected to each other perpendicular bisector and form polygon.What Fig. 8 represented is hexagon, but the polygonal limit of dimension promise number is indefinite.Dimension promise polygon can be described as the image that the set by point is formed, according to apart from the nearest and resulting polygon of split image of which point.And, in the zone that point is gathered thick and fast, the polygonal size decreases of dimension promise, in sparse zone, it is big that the size of point becomes.This situation under the leg-of-mutton situation of De Laonei too.In addition, in order to carry out the search on limit expeditiously, no matter be in dimension promise polygon or in the De Laonei triangle, store the two ends on each limit, and the additional representation search position of whether finishing.
Represented extraction result among Fig. 9～Figure 11 for the profile of leather.Fig. 9 is the binary image of leather integral body, has pasted 4 colored ribbons in the inside of image.Figure 10 represents the image for the point of the part of leather, and R1 is the leather zone, and R2 is the background area, and R3 is dirty zone, because leather has thickness, so seem to exist 2 parallel profile L1, L2.Begin for example only automatically to extract profile L1 thus, this may not be nothing the matter concerning computing machine.
Figure 11 represents the De Laonei of leather integral body is cut apart.By manually, perhaps, begin thus to search for to the leg-of-mutton limit of De Laonei in the color big leg-of-mutton inner starting point of selecting different with background.If run into the limit then check its length, if more than specified length then cross the limit and continue search, if run into the limit of not enough specified length then do not cross the limit, the limit that storage is checked out.When carrying out above processing, owing on the profile of leather, have little De Laonei triangle in a large number, so can extract profile.In addition, on the border of colored ribbon and leather, also there is fine De Laonei triangle in a large number, therefore can extracts colored ribbon.Can extract the profile of the leather of Fig. 9 as described above.
The then processing of the fold of explanation extraction.As shown in Figure 5, obtain the major diameter a and the minor axis b of fold, with their ratio a/b as aspect ratio (S21).In addition, replace a/b, also can adopt the ratio of major diameter a and the girth of fold, or the ratio of major diameter a and 1/2 power of minor axis b, be a/b
1/2Deng.Then, aspect ratio is big more then as strong more fold, and aspect ratio is more little then as weak more fold, adds grade (S22) to fold by this way.With the grade that is suitable for fold and the mode do not exposed, workpiece is distributed in (S23) in the leather then from the lateral profile of leather.Carry out the cutting (S24) of workpiece then.
The expression processing relevant among Figure 12 with fold 88.To leather 80, extract profile laterally from its inboard, obtain profile 81,82.Profile 81,82 produces for example zone 83～87 for linking the short vector gained along these vectors.Here zone 83～87 disperses to produce, but also can produce in the mode that connects them.Zone 83～87 size, bigger than the interval of profile L1, L2 shown in Figure 10, particularly, be for example more than 2 times of interval at the edge of the edge of the upper surface of leather and lower surface.Then, to zone 83～87 inside and outside colour differences of obtaining at profile 81,82.Colourity is Cb and the Cr signal in Y, Cb, the Cr system for example, or the H signal in the HLS system, S signal etc.Then, under colourity variation situation greatly,, change under the little situation as inboard profile in colourity as the profile that does not have other profiles in the outside.In addition, in the stage, the colourity around the point changes also can be obtained each point in the filtration treatment of Fig. 3.
Under the situation of Figure 12, profile 82 is inboard profile, and profile 81 is a lateral profile.Therefore to cross the mode of inboard profile 82,, ask for lateral profile 89 by the method shown in Fig. 7 etc.Then will be by inside and outside profile 81,89 area surrounded as fold 88.Can ask for major diameter a and minor axis b to fold 88, for example can be according to this than the degree of asking for fold.
Figure 13 represents the relation of the grade of aspect ratio and fold, and the grade of the big more then fold of aspect ratio is high more.In addition, as aspect ratio,, can adopt other suitable values except the ratio of major diameter a with minor axis b.
Fold is not limited to the existence of joining of lateral profile with leather.Figure 14 has represented such example.90 is leather, and 91 is inboard profile, and 92 is lateral profile, this to be fold in the inside of leather isolated exist example.
Represented among Figure 15 that 102,108 is fold shown in Figure 180 to the influence of the reduction precision of the workpiece of fold, fold 102 is the fold of thin point, and fold 108 is wide and mild fold.In addition, with the profile of dot-and-dash line 110 expression folds 102,108, be inboard profile.Because fold 102,108 from cutting work anticlise, so if its cutting is opened, it is such then to become workpiece 112,114, is difficult to be cut into correct shape.In addition, if the grade of known fold 102,108 then also can be inferred the height of fold from cutting work anticlise, the therefore decline that can proofread and correct the form accuracy that causes by fold.That is, if the zone of the high fold of cutting grade, then the size along the minor axis direction workpiece of fold becomes big.Size becomes big degree, depends on the grade of fold and the distance that workpiece enters on the minor axis direction of fold.Therefore, if grade and described distance by fold are stored the influence that the size of workpiece is subjected to as empirical rule, then can compensate the decline of the form accuracy that fold causes.
Grade is corresponding with fold is set, and acceptable precision to fold is also evaluated in the zone of workpiece or workpiece in advance, can judge following item: can dispose which workpiece to which fold, maybe can dispose which part of which workpiece.In the workpiece 120 of Figure 16, the zone of the tape edge around existing in and 2 zones of side within it, label I-A represents not allow fully fold, and label I-B represents to allow very mild fold, and label II represents to allow till the fold of for example grade 2 of Figure 13.
Represent the design of workpiece among Fig. 6, can utilize Scissoring device 2 to design, also can utilize other device to design.The shape (S31) of design workpiece is to each regional degree of admission (S32) of specifying fold of whole workpiece or workpiece, the essential quantity (S33) of input workpiece.
Extract profile laterally from the inboard of leather in an embodiment.Also can be in contrast, the inboard profile that extracts from the lateral of leather.Figure 17 has represented such example, and 130 is leather, and 132,133 is fold.If extract profile from the outside of leather 130, then obtain lateral profile 134, if further proceed the extraction of profile, then obtain inboard profile 136.In this case, begin to proceed the extraction of profile, till running into, therefore need in fact the interior outside integral body of leather 130 is carried out the extraction of profile with other profiles from lateral profile 134.Therefore the extraction scope of profile becomes big.
Can obtain following effect in an embodiment.
(1) can check out the fold of leather.Therefore, can learn which of leather partly to have fold, and can ask for area except that the part of de-shirred, and the total area of leather.
(2) because can be to the fold classification, so in permissible range, can in fold, dispose workpiece to workpiece.Therefore improved the utilization ratio of leather.
(3) if extract profile, obtain colourity and change little profile, then cross this profile and further search for profile from the inboard of leather.Therefore can ask for inboard profile and lateral profile simply.The hunting zone can be limited in addition the inboard of leather and fold around.
(4), and, then can dispose workpiece simply to the prior permission of setting fold in the zone in workpiece or the workpiece if fold is implemented classification.
(5) if adopting De Laonei to cut apart or tie up promise cuts apart, then can extract profile exactly.
Claims (5)
1. contour extraction apparatus, it is exported as the outline line of sheeting by extracting point the image that filters the sheeting of taking from video camera, connect the point extracted, it is characterized in that, comprising:
The profile extraction unit, the nearside wheel profile that is used to obtain the outboard wheel profile of the profile of representing sheeting and is positioned at the inside of described outboard wheel profile;
Be used for obtaining the unit of shape of the fold of sheeting according to described outboard wheel profile and described nearside wheel profile;
The crimp shape evaluation unit is used to obtain the amount of major diameter and the ratio of minor axis of the shape of the fold that expression obtains;
Stage unit, the fold that the amount of described expression ratio is bigger are as strong fold, and the fold that the amount of expression ratio is less carries out classification with this to fold as incompetent fold; And
Allocation units are used for and will be assigned to sheeting according to the classification of fold from the workpiece of sheeting cutting.
2. contour extraction apparatus as claimed in claim 1, it is characterized in that, in described profile extraction unit, extract outline line laterally from the inside of sheeting, to be in the inside and outside colour difference of outline line outline line below the rated condition as described nearside wheel profile, and the outline line of more lateral that is in the nearside wheel profile is as described outboard wheel profile.
3. contour extraction apparatus as claimed in claim 2, it is characterized in that, also be provided with the space segmentation unit, size a plurality of polygons little, that size is big in low density zone produce based on described point in the high zone of the density of point by making, and these a plurality of polygonal images of generation, and
In described profile extraction unit, begin to search for described polygonal image from the point of the inside of sheeting towards described polygonal limit, if touch with the edge joint of curtailment specified length, then end to cross the search on this limit and store this limit, if with length be that edge joint more than the specified length touches, then crossing this length is that search is continued on the above limit of specified length.
4. as each described contour extraction apparatus of claim 1 to 3, it is characterized in that, to the permission of the region allocation fold in described workpiece or the workpiece,
Described allocation units distribute workpiece according to described classification and described permission in fold.
5. a contour extraction method is characterized in that, exports as the outline line of sheeting by extracting point the image that filters the sheeting of taking from video camera, connect the point extracted, it is characterized in that,
By profile the unit is proposed, obtain the expression sheeting profile the outboard wheel profile and be positioned at the nearside wheel profile of the inside of described outboard wheel profile,
The shape of the fold of sheeting according to described outboard wheel profile and described nearside wheel profile, is obtained in the unit of the shape by being used to obtain fold,
By the crimp shape evaluation unit, obtain the amount of major diameter and the ratio of minor axis of the shape of the fold that expression obtains,
By stage unit, the fold that the amount of described expression ratio is bigger is as strong fold, and the fold that the amount of expression ratio is less carries out classification with this to fold as incompetent fold,
By allocation units, will be assigned on the sheeting from the classification of the workpiece of sheeting cutting according to fold.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2008192363 | 2008-07-25 | ||
| JP2008192363A JP5197205B2 (en) | 2008-07-25 | 2008-07-25 | Outline extraction apparatus, outline extraction method, and outline extraction program |
| JP2008-192363 | 2008-07-25 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101635055A true CN101635055A (en) | 2010-01-27 |
| CN101635055B CN101635055B (en) | 2013-08-07 |
Family
ID=41594236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910160926 Expired - Fee Related CN101635055B (en) | 2008-07-25 | 2009-07-24 | Contour extraction device and method thereof |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP5197205B2 (en) |
| CN (1) | CN101635055B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102870135A (en) * | 2010-06-29 | 2013-01-09 | 富士胶片株式会社 | Method and device for shape extraction, and dimension measuring device and distance measuring device |
| CN108287506A (en) * | 2018-01-30 | 2018-07-17 | 陈军 | Intelligent domestic device based on image procossing |
| CN110006915A (en) * | 2018-08-30 | 2019-07-12 | 永康市缘匠贸易有限公司 | Glass blocks bar detection mechanism |
| CN111192250A (en) * | 2019-12-30 | 2020-05-22 | 合肥联宝信息技术有限公司 | Data processing method and device, computer storage medium and computer |
| CN112750099A (en) * | 2019-10-16 | 2021-05-04 | 深圳市理邦精密仪器股份有限公司 | Follicle measurement method, ultrasound apparatus, and computer-readable storage medium |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106157369B (en) * | 2016-07-22 | 2019-02-15 | 中国石油集团东方地球物理勘探有限责任公司 | Geological surface triangulation methodology based on sparse structure interpretation data |
| CN112935575B (en) * | 2021-01-12 | 2024-06-07 | 大族激光科技产业集团股份有限公司 | Cutting path optimization method and device and computer readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09131698A (en) * | 1995-09-08 | 1997-05-20 | Gerber Garment Technol Inc | System for pattern arrangement and cutting |
| US20020007228A1 (en) * | 1998-08-17 | 2002-01-17 | Goldman David A. | Automatically generating embroidery designs from a scanned image |
| CN1671310A (en) * | 2002-07-26 | 2005-09-21 | 株式会社岛精机制作所 | Automatic cutting machine teaching device |
| CN101057242A (en) * | 2004-11-10 | 2007-10-17 | 株式会社岛精机制作所 | Sheet material patterning device, method, and program |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3149117B2 (en) * | 1992-11-16 | 2001-03-26 | 株式会社島精機製作所 | Abnormality detection device for cut material in cutting machine |
| JP2005111596A (en) * | 2003-10-07 | 2005-04-28 | Atelier Sooken:Kk | Device for stacking object to be cut |
-
2008
- 2008-07-25 JP JP2008192363A patent/JP5197205B2/en not_active Expired - Fee Related
-
2009
- 2009-07-24 CN CN 200910160926 patent/CN101635055B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09131698A (en) * | 1995-09-08 | 1997-05-20 | Gerber Garment Technol Inc | System for pattern arrangement and cutting |
| US20020007228A1 (en) * | 1998-08-17 | 2002-01-17 | Goldman David A. | Automatically generating embroidery designs from a scanned image |
| CN1671310A (en) * | 2002-07-26 | 2005-09-21 | 株式会社岛精机制作所 | Automatic cutting machine teaching device |
| CN101057242A (en) * | 2004-11-10 | 2007-10-17 | 株式会社岛精机制作所 | Sheet material patterning device, method, and program |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102870135A (en) * | 2010-06-29 | 2013-01-09 | 富士胶片株式会社 | Method and device for shape extraction, and dimension measuring device and distance measuring device |
| CN102870135B (en) * | 2010-06-29 | 2016-02-03 | 富士胶片株式会社 | For the method and apparatus of shape extracting, dimension measuring device and distance-measuring device |
| CN108287506A (en) * | 2018-01-30 | 2018-07-17 | 陈军 | Intelligent domestic device based on image procossing |
| CN108287506B (en) * | 2018-01-30 | 2020-09-11 | 嵊州市万智网络科技有限公司 | Smart home device based on image processing |
| CN110006915A (en) * | 2018-08-30 | 2019-07-12 | 永康市缘匠贸易有限公司 | Glass blocks bar detection mechanism |
| CN112750099A (en) * | 2019-10-16 | 2021-05-04 | 深圳市理邦精密仪器股份有限公司 | Follicle measurement method, ultrasound apparatus, and computer-readable storage medium |
| CN111192250A (en) * | 2019-12-30 | 2020-05-22 | 合肥联宝信息技术有限公司 | Data processing method and device, computer storage medium and computer |
| CN111192250B (en) * | 2019-12-30 | 2022-02-08 | 合肥联宝信息技术有限公司 | Computer B-side frame detection method and device, computer storage medium and computer |
Also Published As
| Publication number | Publication date |
|---|---|
| JP5197205B2 (en) | 2013-05-15 |
| JP2010029958A (en) | 2010-02-12 |
| CN101635055B (en) | 2013-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101635055B (en) | Contour extraction device and method thereof | |
| Yang et al. | A welding quality detection method for arc welding robot based on 3D reconstruction with SFS algorithm | |
| CN107798330B (en) | A method for extracting feature information of weld image | |
| Shen et al. | Multi-view online vision detection based on robot fused deposit modeling 3D printing technology | |
| CN110717872B (en) | Method and system for extracting characteristic points of V-shaped welding seam image under laser-assisted positioning | |
| CN101549468A (en) | Image-based on-line detection and compensation system and method for cutting tools | |
| JP4944007B2 (en) | Outline extraction apparatus, outline extraction method, and outline extraction program | |
| CN109146838A (en) | A kind of aobvious band adhering chromosome dividing method of the G merged based on geometrical characteristic with region | |
| CN107430775A (en) | For the method and industrial vehicle positioned using ceiling light | |
| Chen et al. | The autonomous detection and guiding of start welding position for arc welding robot | |
| KR20140020837A (en) | Method for the pre-processing of a three-dimensional image of the surface of a tyre for use in the inspection of said surface | |
| CN114325755B (en) | A method and system for detecting retaining walls suitable for self-driving vehicles | |
| JP2017111632A (en) | Information processing device | |
| JP2021060720A (en) | Object detection device | |
| CN103226809A (en) | Image haze removal device and image haze removal method | |
| CN113139557A (en) | Feature extraction method based on two-dimensional multivariate empirical mode decomposition | |
| CN113763273B (en) | Face complementing method, electronic device and computer readable storage medium | |
| CN115358965A (en) | Method and device for generating linear weld grinding track with adaptive welding deformation | |
| WO2001071663A1 (en) | Cell lineage extracting method | |
| US11458892B2 (en) | Image generation device and image generation method for generating a composite image | |
| CN112767425A (en) | Parking space detection method and device based on vision | |
| Lee et al. | A study on modified Hough algorithm for image processing in weld seam tracking system | |
| CN104690976B (en) | The column printing method of three-dimensional printing machine and system | |
| JP2002006034A (en) | Method and apparatus for class discrimination of target in radar | |
| CN108647680A (en) | Framing frame detection method and device |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130807 Termination date: 20150724 |
|
| EXPY | Termination of patent right or utility model |