CN117183578A - Positioning detection method and printing method for telescopic printing substrate - Google Patents
Positioning detection method and printing method for telescopic printing substrate Download PDFInfo
- Publication number
- CN117183578A CN117183578A CN202311454393.9A CN202311454393A CN117183578A CN 117183578 A CN117183578 A CN 117183578A CN 202311454393 A CN202311454393 A CN 202311454393A CN 117183578 A CN117183578 A CN 117183578A
- Authority
- CN
- China
- Prior art keywords
- printing
- positioning
- roller
- image
- printing substrate
- 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
- 238000007639 printing Methods 0.000 title claims abstract description 272
- 239000000758 substrate Substances 0.000 title claims abstract description 125
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 title claims abstract description 30
- 239000000853 adhesive Substances 0.000 claims abstract description 40
- 230000001070 adhesive effect Effects 0.000 claims abstract description 40
- 239000000463 material Substances 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 22
- 238000013527 convolutional neural network Methods 0.000 claims description 23
- 238000012546 transfer Methods 0.000 claims description 23
- 238000001125 extrusion Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 11
- 238000003825 pressing Methods 0.000 claims description 10
- 230000002787 reinforcement Effects 0.000 claims description 10
- 238000003708 edge detection Methods 0.000 claims description 9
- 238000011176 pooling Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 230000000274 adsorptive effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims 1
- 239000010410 layer Substances 0.000 description 17
- 230000000694 effects Effects 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000000903 blocking effect Effects 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000011248 coating agent Substances 0.000 description 2
- 238000000576 coating method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000001179 sorption measurement Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000007665 sagging Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 239000002344 surface layer Substances 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Abstract
The invention relates to the technical field of detection, in particular to a positioning detection method and a printing method of a telescopic printing substrate, wherein the detection method comprises the following steps: printing positioning color blocks on the printing substrate in groups at two sides of a set time interval; the printing material is covered at the convex printing position, a supporting roller is arranged at the other side of the printing substrate opposite to the roller, the supporting roller supports the printing substrate, and the supporting roller and the roller jointly extrude the printing substrate; selecting at least one datum point at the edge of the positioning color block, and collecting the distance between the datum point and the edge of the printing substrate; and continuously detecting the tension and the position of the printing substrate in the unreeling conveying process according to the continuously generated distances. The invention can effectively judge whether the positioning of the printing substrate is accurate or not in time so as to facilitate an operator to quickly take corresponding measures to effectively adjust the position and/or the tension of the printing substrate in time, thereby finally ensuring the accuracy of the adhesive after printing.
Description
Technical Field
The invention relates to the technical field of detection, in particular to a positioning detection method and a printing method of a telescopic printing substrate.
Background
In the current printing field, the printed substrate requires a local printing adhesive in a specific production scenario for subsequent positional bonding of the desired structure. In the production and processing process, for the printing substrate reaching the printing position of the adhesive in an unreeling mode, the position and the printing area of the adhesive in the front path and the feeding position of the structure required in the back path need to establish accurate relevance, otherwise, the situation that the adhesive is not firmly bonded or the adhesive is printed beyond the range of the required structure is easily generated.
In the working process, the feeding of the required structure is often realized at a set position, the accuracy is easy to control, so that the main factors influencing the relevance are often from the deviation of the position and the printing area of the adhesive, and the main factors generating the problems include:
the printing substrate often has a certain telescopic amount, and the variation of the telescopic amount caused by tension deviation in the unreeling and conveying process can cause the width difference of the printing substrate and the position difference of the width in the direction perpendicular to the conveying direction; or, during the feeding of the coiled material of the printing substrate, the direction of the coiled material may deviate from the originally set direction, that is, the placing direction of the material may be skewed.
Based on the above problems, how to accurately position and detect the stretchable printing substrate becomes a problem to be solved at present.
Disclosure of Invention
The invention provides a positioning detection method and a printing method of a telescopic printing substrate, thereby effectively solving the problems pointed out in the background technology.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a method of positioning a stretchable print substrate comprising:
printing positioning color blocks on the two sides of the printing substrate in groups according to set time intervals by rotating rollers with protruding printing positions on the two sides of the printing substrate in the conveying direction, wherein one group of the positioning color blocks corresponds to a set number of adhesive printing positions, or one group of the positioning color blocks corresponds to one adhesive printing position; the printing device comprises a roller, a printing substrate, a printing material pressing device and a printing material pressing device, wherein the printing material is covered on the convex printing position, a supporting roller is arranged on the other side of the printing substrate opposite to the roller, supports the printing substrate, and presses the printing substrate together with the roller;
selecting at least one datum point at the edge of the positioning color block, and collecting the distance between the datum point and the edge of the printing substrate;
and continuously detecting the tension and the position of the printing substrate in the unreeling conveying process according to the continuously generated distances.
Further, the roller with the convex printing position and the printing plate roller for printing the adhesive are coaxially and integrally arranged, the printing plate roller is also provided with the convex printing position and is equal to the convex height of the roller, the convex printing positions on the roller and the printing plate roller are arranged in a staggered manner, and the convex printing position is always attached to the printing substrate in the rotating process, and the printing substrate is extruded together with the supporting roller.
Further, the method further comprises the following steps: and collecting the printing area of the positioning color block, and continuously detecting the extrusion force born by the printing substrate according to each continuously generated printing area.
Further, the method further comprises continuously detecting the extrusion force applied to the printing substrate according to each continuously generated distance.
Further, continuously detecting the position of the printing roller axis according to each continuously generated printing area is also included.
Further, an ink transfer roller is provided for transferring the printing material to the protruding printing position of the roller, and the ink transfer roller sucks the printing material through the adsorptive material and passively rotates under the friction force of the roller in a manner of being attached to and extruded by the roller.
Further, a scraper is also arranged, and the edge of the scraper is controlled to be parallel to the axis of the ink transfer roller and is attached to the surface of the ink transfer roller.
Further, obtaining images of each group of the positioning color blocks by means of image acquisition, and constructing an image detection model of the positioning color blocks, wherein the image detection model comprises the following steps:
acquiring image data of the printed positioning color blocks, and marking the edges of the positioning color blocks on two sides after printing;
preprocessing the marked image and dividing the marked image into a training set and a testing set;
constructing a convolutional neural network model for positioning color block edge image detection, and training and verifying the convolutional neural network model by using the training set and the testing set;
and optimizing the convolutional neural network model according to the verification result, and detecting the positioning color block image by using the optimized convolutional neural network model.
Further, the method further comprises the step of processing the positioning color block image, and comprises the following steps:
selecting a sharpening kernel, sliding the sharpening kernel on an edge image of a positioning color block, and performing convolution operation, wherein the convolution neural network model comprises:
the convolution layer extracts features related to edges from the positioning color lump image on the printing substrate to obtain a feature image; a pooling layer for reducing the space size of the feature map and simultaneously retaining key information; the full-connection layer is used for mapping the features extracted by the pooling layer to final output; the output layer outputs an edge detection result of the positioning color block;
each neuron of the fully connected layer is provided with a weight set connected with a characteristic value mapped by each characteristic in the pooling layer, and the convolutional neural network model further comprises an reinforcement learning algorithm which is used for enabling the convolutional neural network model to adjust the weight set through interaction with an image acquisition environment and printing materials;
adjusting the weight of the sharpening kernel, and repeatedly performing the convolution operation on the edge image;
comparing the edge images sharpened by different weights, and selecting a sharpened image with optimal definition;
and superposing the sharpened image and the original positioning color block image to obtain a final sharpened image.
A printing method of a stretchable printing substrate adopts the positioning detection method of the stretchable printing substrate to detect the positioning of the printing substrate, and alarms when the detection result is abnormal.
By the technical scheme of the invention, the following technical effects can be realized:
the invention can effectively judge whether the positioning of the printing substrate is accurate or not in time so as to facilitate an operator to quickly take corresponding measures to effectively adjust the position and/or the tension of the printing substrate in time, thereby finally ensuring the accuracy of the adhesive after printing.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings may be obtained according to the drawings without inventive effort to those skilled in the art.
FIG. 1 is a schematic view of a printing apparatus for a flexible print substrate;
FIG. 2 is a schematic view of a set of positioning color patches on both sides of a printing substrate in a conveying direction corresponding to one adhesive printing position, wherein the printing shapes of the positioning color patches and the adhesive are rectangular;
FIG. 3 is a schematic view showing a set of positioning color patches on both sides of a printing substrate in the direction of conveyance corresponding to three adhesive printing positions juxtaposed in the web direction, and the printing shapes of the positioning color patches and the adhesive being likewise rectangular;
FIG. 4 is a schematic view showing two sets of positioning color patches on both sides of a printing substrate in the conveying direction corresponding to one adhesive printing position, wherein the printing shape of the positioning color patches is circular, and the printing shape of the adhesive printing is rectangular;
FIG. 5 is an enlarged view of a portion of FIG. 1 at A;
FIG. 6 is a flow chart of constructing an image detection model of a locating color patch;
FIG. 7 is a flow chart of processing a positioning patch image;
reference numerals: 1. printing a substrate; 2. a raised printing position; 3. a roller; 4. positioning the color block; 5. an adhesive printing position; 6. a support roller; 7. a plate roller; 8. an ink transfer roller; 9. an ink tank; 10. a scraper; 11. a blocking structure.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Example 1
A method of positioning a stretchable print substrate comprising:
on both sides of the conveying direction of the printing substrate 1, the rollers 3 with the convex printing positions 2 rotate, and the positioning color blocks 4 are printed on both sides of the printing substrate 1 in groups according to set time intervals, wherein one group of the positioning color blocks 4 corresponds to a set number of adhesive printing positions 5, or one group of the positioning color blocks 4 corresponds to one adhesive printing position 5; wherein the printing material is covered by the convex printing position 2, the other side of the printing substrate 1 relative to the roller 3 is provided with a supporting roller 6, the supporting roller 6 supports the printing substrate 1 and extrudes the printing substrate 1 together with the roller 3; selecting at least one datum point at the edge of the positioning color block 4, and collecting the distance between the datum point and the edge of the printing substrate 1; the tension and position of the printing substrate 1 during unreeling and conveying are continuously detected from the distances continuously generated, and the shortest distance is preferable here.
As shown in fig. 1, the invention can timely and effectively judge whether the positioning of the printing substrate 1 is accurate, so that an operator can quickly take corresponding measures to timely and effectively adjust the position and/or the tension of the printing substrate 1, and finally, the accuracy of the adhesive after printing is ensured.
In practice, the positioning color block 4 is preferably an ink color block, and of course, other materials that can realize accurate identification of the subsequent edge and selection of the reference point can be used for printing on the surface of the printing substrate 1 to obtain the positioning color block 4, and the subsequent edge of the printing substrate 1 can be discarded by cutting, so that the connection between the required structure and the printing substrate 1 and the subsequent use are not affected.
The number correspondence relationship between each set of positioning color patches 4 and the adhesive printing positions 5 can be specifically set according to the sizes and shapes of the positioning color patches 4 and the adhesive printing areas; as shown in fig. 2, a set of positioning color patches 4 on both sides of the conveying direction of the printing substrate 1 are shown to correspond to one adhesive printing position 5, and the printing shapes of the positioning color patches 4 and the adhesive are rectangular; as shown in fig. 3, a set of positioning color patches 4 on both sides of the transport direction of the printing substrate 1 is shown in a manner corresponding to three adhesive printing positions 5 juxtaposed in the web direction, but of course the selection of the number of three and the selection of the direction along the web is merely an example, and the specific number and direction can be adjusted as desired, in which case the printing shape of the positioning color patches 4 and the adhesive is likewise rectangular; as shown in fig. 4, two sets of positioning color patches 4 on both sides of the conveying direction of the printing substrate 1 are shown in a manner corresponding to one adhesive printing position 5, and the printing shape of the positioning color patches 4 is circular, and the printing shape of the adhesive printing is rectangular; the arrows in the above figures indicate the conveying direction of the printing substrate 1, and the various number correspondences and printing shapes mentioned in the above description are all within the scope of the present invention.
In addition, between the corresponding positioning color block 4 and the adhesive printing position 5, printing of the positioning color block 4 needs to be prioritized over printing of the adhesive, because the printing of the adhesive can be performed only after the acquired distance meets the requirement, and a qualified product can be obtained.
In the process of locating the reference points on the edge of the color patch 4, a specific set or one reference point is preferable, wherein the case of selecting a set of reference points is often a set of points on an edge line parallel to the edge of the printing substrate 1, and the case of selecting one reference point is often a closest point to the edge, a corner point of a pattern, etc., and the above various selection manners are merely specific examples and are not limiting the scope of the present invention.
During the transport of the printing substrate 1, when the tension of the printing substrate 1 is unstable, the following occurs in particular:
when the tension of the printing substrate 1 is too high, the width of the printing substrate 1 is reduced to a certain extent due to the increased expansion and contraction amount, and the raised printing position 2 of the roller 3 is determined, so that the distance from the positioning color block 4 to the edge of the printing substrate 1 is reduced, namely H1 and H2 in figures 2-4 are reduced; in this case, the tension of the printing substrate 1 can be determined to be too high according to the detection results of the decrease in both H1 and H2, so that the tension of the printing substrate 1 needs to be adjusted and controlled to ensure the positioning accuracy of the printing substrate 1. In the implementation process, the initial values of H1 and H2 can be equal or unequal, and are all within the protection scope of the invention. In the case of too little tension, the identification can be performed manually, because sagging and the like of the stretchable printing substrate 1 are easily identified in this manner.
In addition, when the difference between H1 and H2 is not equal to the initial set value, it is also possible to recognize whether the print substrate 1 is skewed, and when one side increases and the other side decreases, it is possible to describe that the print substrate 1 is skewed once toward the increase in distance.
In the present invention, in order to save power, as a preferred embodiment, the roller 3 having the protruding printing position 2 and the plate roller 7 for printing the adhesive are coaxially and integrally provided, the plate roller 7 also has the protruding printing position 2 and has the same protruding height as the roller 3, the protruding printing positions 2 on the roller 3 and the plate roller 7 are arranged in a staggered manner, and the protruding printing position 2 is always attached to the printing substrate 1 during rotation, and the printing substrate 1 is pressed together with the supporting roller 6.
In this way, the integrated roller 3 and plate roller 7 can be made to rotate without power to effect printing, in particular by the friction between the raised printing position 2 and the printing substrate 1; in the implementation process, the staggered arrangement of the protruding printing positions 2 on the roller 3 and the printing plate roller 7 can ensure that the printing of the positioning color block 4 is superior to the printing of the adhesive on one hand, and the integrated structure of the roller 3 and the printing plate roller 7 can be ensured to always receive friction force to ensure the continuity of rotation on the other hand.
In practice, the above-mentioned pressing of the printing substrate 1 will also have an influence on its positioning, and preferably, the above-mentioned embodiment further includes: the printing area of the positioning color lump 4 is collected, and the extrusion force applied to the printing substrate 1 is continuously detected according to each continuously generated printing area.
In the preferred embodiment, taking the printing area S as an example, when the extrusion force applied to the printing substrate 1 is too large, the extrusion force comes from the result that the roller 3 and the plate roller 7 press the printing substrate 1 against the supporting roller 6, and when the extrusion force is too large, the rotation speed of the integral structure of the roller 3 and the plate roller 7 tends to not catch up with the speed of the stably conveyed printing substrate 1, so that the printing area S increases due to the slowing of the printing process, that is, the area of the positioning color lump 4 in fig. 2-4 increases; in this case, the extrusion force applied to the printing substrate 1 can be determined to be too large according to the detection result of the increase of the printing area S, so that the adjustment and control of the extrusion force applied to the printing substrate 1 are required to ensure the positioning accuracy of the printing substrate 1, and the adjustment of the extrusion force can be realized by increasing the heights of the integrated roller 3 and printing plate roller 7. Conversely, the integral structure of the roller 3 and the plate roller 7 may generate slipping phenomenon with the printing substrate 1, so that the printing area S is reduced, and the extrusion force adjustment can be realized by reducing the height of the integrated roller 3 and plate roller 7. In the implementation process, the initial values of the printing areas S at two sides can be equal or unequal, and the two printing areas S are all within the protection scope of the invention.
Also for the purpose of determining the positioning of the printing substrate 1 by detecting the applied pressing force, it further includes continuously detecting the applied pressing force of the printing substrate 1 according to the continuously generated distances.
In the present preferred embodiment, the distance obtained in the above-described embodiment can be further utilized; specifically, when the pressing force applied to the printing substrate 1 is too large, the pressed position will be flatter in the printing process, and the ductility of the printing substrate 1 in the width direction will be increased, so that the distances between the two sides are increased, i.e. H1 and H2 in fig. 2 to 4 are increased. Therefore, based on this detection method, the detection result of the increase in the pressing force can be accurately obtained, and thus the possible positional deviation of the printing substrate 1 can be judged.
In addition, in order to solve the problem, when the axis direction of the printing roller 7 and the axis direction of the roller 3 deviate, the uneven stress of the printing substrate 1 is also caused, and in order to solve the problem, as an optimization mode for further utilizing the printing area S, the method further comprises continuously detecting the axis position of the printing roller according to each continuously generated printing area, and further judging the positioning of the printing substrate 1 according to the axis position; specifically, when the pressure applied to the printing substrate 1 on one side is large and the pressure applied to the other side is small at the same time due to the axis deviation, the printing area S on one side tends to be increased and the printing area S on the other side is decreased, and based on this, the deviation of the printing roller axis can be accurately obtained, and thus the positioning deviation which may occur to the printing substrate 1 can be judged.
In the above-described process, the positioning of the color patch 4 printing effect is extremely critical, since the printing structure determines the accuracy of the detection as a basis for the subsequent steps, and as a preferred embodiment of the above-described embodiment, an ink transfer roller 8 for transferring the printing material to the raised printing position 2 of the roller 3 is provided, the ink transfer roller 8 sucking the printing material by the adsorptive material and being passively rotated by the frictional force of the roller 3 by being pressed against the roller 3.
As shown in fig. 5, a specific embodiment is shown, taking black ink as an example, the ink tank 9 is required to supply black ink, and of course, the ink transfer roller 8 can be provided with a plurality of sequentially jointed ink transfer rollers with parallel axes, and finally, one ink transfer roller 8 is required to be jointed with the roller wheel 3. Whereas for the ink transfer roller 8, the portion that sucks up the printing material is preferably a surface layer structure, and the inner layer is a rigid structure, so that the shape stability of the structure is ensured. In the preferred scheme, printing materials are sucked through the ink transfer roller 8 and then transferred to the roller 3, so that the ink layer of the roller 3 protruding out of the printing position 2 is more uniform, and the effect and the edge definition of the positioning color block 4 after printing are improved. In the implementation process, the mode can be adopted for coating the adhesive on the printing plate roller 7, and the corresponding arrangement of the glue transfer roller can also ensure the printing effect of the adhesive.
In order to further improve the coating uniformity of the printing material on the roller 3, it is further preferable that a doctor blade 10 is further provided, and the edge of the doctor blade 10 is controlled to be parallel to the axis of the ink transfer roller 8 and to be attached to the surface of the ink transfer roller 8. By means of the arrangement of the doctor blade 10, the printing material adsorbed on the ink transfer roller 8 can be uniform, the redundant printing material falls into the ink tank 9 due to the blocking of the doctor blade 10, and of course, the edge of the doctor blade 10 can possibly cause proper pressure on the surface of the ink transfer roller 8 in the use process, so that the adsorption amount of the printing material is limited, and in such a case, the fixed end of the doctor blade 10 can be provided with a structure with an adjustable mounting position, so that the pressure is convenient to adjust.
In the above embodiment, it is preferable to provide the entire doctor blade 10 and restrict the adsorption material simultaneously by the ink transfer rollers 8 on both sides and the intermediate ink transfer roller, but in this case, it is necessary to provide the blocking structure 11 in the doctor blade 10 so that an affected area where the printing material such as ink and the adhesive are not mixed is provided.
As a preference of the above embodiment, in order to accurately obtain the graphic edge of the positioning color block 4 and thus accurately identify the printing area and the reference point, the present invention provides an optimized way to obtain the images of each group of positioning color blocks 4 by means of image acquisition, and construct an image detection model of the positioning color block 4, as shown in fig. 6, including:
s1: acquiring image data of the printed positioning color blocks 4, and marking edges of the positioning color blocks 4 on two sides after printing;
s2: preprocessing the marked image and dividing the marked image into a training set and a testing set;
s3: constructing a convolutional neural network model for detecting the edge image of the positioning color block 4, and training and verifying the convolutional neural network model by using a training set and a testing set;
s4: and optimizing the convolutional neural network model according to the verification result, and detecting the positioning color block 4 image by using the optimized convolutional neural network model.
In the preferred scheme, the convolutional neural network model is constructed, so that the image of the positioning color block 4 can be automatically detected, manual intervention is not needed, the production efficiency can be improved, and the labor cost is reduced; the convolutional neural network is a powerful tool for deep learning, can learn and extract image features, so that high-precision edge detection is realized, and the model can be continuously optimized in the training and verification stage to provide a more accurate edge detection result; this approach is relatively stable and is not susceptible to light, angle or other environmental changes.
In summary, the convolutional neural network is utilized in the preferred scheme to provide an automatic, high-precision and high-robustness method for detecting the printed positioning color block 4 image, and the production efficiency and the quality control level are expected to be improved.
In the present invention, the detection of the edge in the image is particularly critical for the positioning detection, the accurate edge detection can obtain the accurate area calculation result and the accurate reference point selection result, but in the actual implementation process, the printing material is inevitably stained or flowing in a tiny range, which inevitably affects the accurate edge detection, so as to solve the problem, and as a preferred embodiment, the method further includes processing the positioning color block 4 image, as shown in fig. 7, including:
a1, selecting a sharpening kernel, sliding the sharpening kernel on the edge image of the positioning color block 4, and executing convolution operation, wherein the convolution neural network model comprises:
the convolution layer extracts the characteristic related to the edge from the positioning color block 4 image on the printing substrate 1 to obtain a characteristic image; the pooling layer reduces the space size of the feature map and simultaneously reserves key information; the full-connection layer is used for mapping the features extracted by the pooling layer to final output; the output layer outputs the edge detection result of the positioning color block 4;
each neuron of the full-connection layer is provided with a weight set connected with a characteristic value mapped by each characteristic in the pooling layer, and the convolutional neural network model further comprises an reinforcement learning algorithm which is used for enabling the convolutional neural network model to adjust the weight set through interaction with an image acquisition environment and printing materials;
a2, adjusting the weight of the sharpening kernel, and repeatedly carrying out convolution operation on the edge image;
a3, comparing the edge images sharpened by different weights, and selecting a sharpened image with optimal definition;
and A4, superposing the sharpened image and the original positioning color block 4 image to obtain a final sharpened image.
In the above-mentioned optimization scheme, can carry out accurate discernment to the edge image of location colour lump 4, overlap sharp image with original image to realize more accurate edge image discernment and location, this process has combined the context information of original image and the edge feature of sharp image, makes the edge and the detail of image more clear and accurate, thereby has improved the accuracy to location colour lump 4 on the printing substrate 1.
In the implementation process, edges and details in the image are enhanced, so that the edges and the details are clearer and easier to identify; the choice of a suitable sharpening kernel is often dependent on the nature and goal of the image, where Sobel and Prewitt kernels are often used to enhance edges, where the sharpening kernel will perform convolution operations on the digital image of the edge of the location patch 4 after edge detection using a convolutional neural network model, where the design of the other convolution kernels is different, where the design of the sharpening kernel is intended to enhance edges and details in the image, often containing positive and negative values to emphasize differences in pixel values, meaning that the sharpening kernel will amplify edges and details by weighting surrounding pixels.
The sharpening kernel is a matrix in which each element represents the weight of the pixel at the corresponding position in the convolution operation, the positive weight represents strengthening the pixel, the negative weight represents weakening the pixel, the purpose of adjusting the weight of the sharpening kernel is to control the intensity of the sharpening operation, and a larger difference between the positive weight and the negative weight generally results in a stronger sharpening effect, so that the edge is more obvious, but noise and artifacts can be possibly introduced, and the intensity of image enhancement can be balanced by adjusting the weights so as to meet the application requirement.
For weight adjustment of the sharpening kernel, the center element value of the sharpening kernel is typically set to a positive number to emphasize the contribution of the center pixel, the surrounding elements may be set to a negative number or zero to reduce the enhancement to noise, and the absolute value of the weights should be added to zero to ensure that no brightness offset is introduced; in practice, the sharpening kernel may be fine-tuned based on the newly generated image, for example by adding some weight to enhance edges in a particular direction, since sharpening typically enhances noise, the superimposed image may reduce the effect of noise by maintaining a smooth portion of the original image.
The introduction of the reinforcement learning algorithm in the step A1 can enable the model to better adapt to different environments and changes of printing materials, and in the implementation process, a proper reinforcement learning algorithm such as a Deep Reinforcement Learning (DRL) method can be selected firstly; then creating a simulation environment to simulate interactions with the image acquisition environment and the printed material, which may be specifically a virtual environment, simulating the generation and variation of the printed material and the acquired image; a reward function is defined so as to reward or punish decisions of the reinforcement learning model according to changes in model performance and the degree of adaptation to the simulation environment; training the reinforcement learning model by using the simulation environment and the reward function, so that the reinforcement learning model can gradually optimize the weight set in the full-connection layer to adapt to images under different conditions; finally, the trained reinforcement learning model is integrated into the convolutional neural network to automatically adjust the weight set in the step A1.
Through the process, the adaptability, the automatic weight adjustment, the performance optimization and the improvement of generalization capability of the model are facilitated, the labor cost is reduced, and higher efficiency and accuracy can be provided in the edge detection task of the positioning color block 4 on the printing substrate 1.
Example two
A printing method of a stretchable printing substrate, which adopts the positioning detection method of the stretchable printing substrate according to the first embodiment to detect the positioning of the printing substrate, and alarms when the detection result is abnormal.
In this embodiment, as described in the above embodiment, the types of the alarm may specifically include:
when H1 is continuously reduced and H2 is continuously reduced, alarming that the tension of the printing substrate 1 is too high is carried out;
when H1 is continuously increased, H2 is continuously reduced, or H1 is continuously reduced and H2 is continuously increased, performing skew alarm on the printing substrate 1;
when H1 is continuously increased and H2 is continuously increased, alarming that the extrusion force applied to the printing substrate 1 is excessive is carried out;
when the printing areas S on the two sides are continuously increased, alarming that the extrusion force applied to the printing substrate 1 is overlarge is carried out;
when the printing areas S on the two sides are continuously reduced, alarming that the extrusion force born by the printing substrate 1 is too small is carried out;
when the printing area S on one side continuously decreases and the printing area S on the other side continuously increases, the axis skew alarm is carried out.
The foregoing has outlined and described the basic principles, features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. A method for positioning and detecting a stretchable print substrate, comprising:
printing positioning color blocks on the two sides of the printing substrate in groups according to set time intervals by rotating rollers with protruding printing positions on the two sides of the printing substrate in the conveying direction, wherein one group of the positioning color blocks corresponds to a set number of adhesive printing positions, or one group of the positioning color blocks corresponds to one adhesive printing position; the printing device comprises a roller, a printing substrate, a printing material pressing device and a printing material pressing device, wherein the printing material is covered on the convex printing position, a supporting roller is arranged on the other side of the printing substrate opposite to the roller, supports the printing substrate, and presses the printing substrate together with the roller;
selecting at least one datum point at the edge of the positioning color block, and collecting the distance between the datum point and the edge of the printing substrate;
and continuously detecting the tension and the position of the printing substrate in the unreeling conveying process according to the continuously generated distances.
2. The positioning detection method of a retractable printing substrate according to claim 1, wherein a roller having the protruding printing position and a printing plate roller for printing an adhesive are coaxially and integrally arranged, the printing plate roller also has the protruding printing position, the protruding printing position on the roller and the printing plate roller are arranged in a staggered manner, and the protruding printing position on the roller and the printing plate roller is always attached to the printing substrate in the rotation process and is pressed together with the supporting roller.
3. The method for detecting the positioning of a stretchable print substrate according to claim 2, further comprising: and collecting the printing area of the positioning color block, and continuously detecting the extrusion force born by the printing substrate according to each continuously generated printing area.
4. A method of positioning a flexible print substrate according to claim 3, further comprising continuously detecting the pressing force applied to the print substrate based on each of the distances continuously generated.
5. A method of positioning a flexible print substrate according to claim 3, further comprising continuously detecting the position of the print roller axis based on each of the print areas that are continuously generated.
6. The method according to claim 2, wherein an ink transfer roller for transferring the printing material to the protruding printing position of the roller is provided, and the ink transfer roller sucks the printing material by the adsorptive material and is passively rotated by friction force of the roller by being pressed against the roller.
7. The method according to claim 6, wherein a doctor blade is further provided, and the doctor blade edge is controlled to be parallel to the axis of the ink transfer roller and to be attached to the surface of the ink transfer roller.
8. The method for detecting the positioning of the scalable printing substrate according to claim 1, wherein the image of each set of the positioning color patches is obtained by means of image acquisition, and the image detection model of the positioning color patches is constructed, comprising:
acquiring image data of the printed positioning color blocks, and marking the edges of the positioning color blocks on two sides after printing;
preprocessing the marked image and dividing the marked image into a training set and a testing set;
constructing a convolutional neural network model for positioning color block edge image detection, and training and verifying the convolutional neural network model by using the training set and the testing set;
and optimizing the convolutional neural network model according to the verification result, and detecting the positioning color block image by using the optimized convolutional neural network model.
9. The method of claim 8, further comprising processing the positioning patch image, comprising:
selecting a sharpening kernel, sliding the sharpening kernel on an edge image of a positioning color block, and performing convolution operation, wherein the convolution neural network model comprises:
the convolution layer extracts features related to edges from the positioning color lump image on the printing substrate to obtain a feature image; a pooling layer for reducing the space size of the feature map and simultaneously retaining key information; the full-connection layer is used for mapping the features extracted by the pooling layer to final output; the output layer outputs an edge detection result of the positioning color block;
each neuron of the fully connected layer is provided with a weight set connected with a characteristic value mapped by each characteristic in the pooling layer, and the convolutional neural network model further comprises an reinforcement learning algorithm which is used for enabling the convolutional neural network model to adjust the weight set through interaction with an image acquisition environment and printing materials;
adjusting the weight of the sharpening kernel, and repeatedly performing the convolution operation on the edge image;
comparing the edge images sharpened by different weights, and selecting a sharpened image with optimal definition;
and superposing the sharpened image and the original positioning color block image to obtain a final sharpened image.
10. A printing method of a stretchable printing substrate, characterized in that the positioning detection method of a stretchable printing substrate according to any one of claims 1 to 9 is used for positioning detection of the printing substrate, and an alarm is given when abnormality occurs in the detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311454393.9A CN117183578B (en) | 2023-11-03 | 2023-11-03 | Positioning detection method and printing method for telescopic printing substrate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311454393.9A CN117183578B (en) | 2023-11-03 | 2023-11-03 | Positioning detection method and printing method for telescopic printing substrate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117183578A true CN117183578A (en) | 2023-12-08 |
| CN117183578B CN117183578B (en) | 2024-01-26 |
Family
ID=89003729
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311454393.9A Active CN117183578B (en) | 2023-11-03 | 2023-11-03 | Positioning detection method and printing method for telescopic printing substrate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117183578B (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1084455A (en) * | 1992-09-19 | 1994-03-30 | 林特克株式会社 | The detection modification method in printing equipment and printing state in this printing equipment and printing position |
| JPH09254364A (en) * | 1996-03-18 | 1997-09-30 | Kawashima Textile Manuf Ltd | Printing apparatus and printing roll |
| US20090020029A1 (en) * | 2007-07-13 | 2009-01-22 | Klaus Reckefuss | Method and apparatus for automatically regulating the registers between imprints in a multi color rotary printing press |
| CN101574876A (en) * | 2009-06-03 | 2009-11-11 | 无锡光群雷射科技有限公司 | Laser holographic column gravure roller location printing paper |
| JP2011161662A (en) * | 2010-02-05 | 2011-08-25 | Lintec Corp | Printing position deviation detector, printer, and printing position deviation detecting method |
| CN102245389A (en) * | 2008-12-18 | 2011-11-16 | 利乐拉瓦尔集团及财务有限公司 | Method of controlling the relative position between printed patterns and non-printed patterns on a web-shaped material and a system used in the method |
| JP2012240786A (en) * | 2011-05-19 | 2012-12-10 | Hitachi Ltd | Web carrying device |
| KR20130094982A (en) * | 2012-02-17 | 2013-08-27 | 한국기계연구원 | Register marking method and its device for printed electronics on roll-to-roll gravure offset printing press |
| TW201519962A (en) * | 2013-11-01 | 2015-06-01 | Jx Nippon Oil & Energy Corp | Application device for forming coating having discontinuous pattern onto strip-shaped film substrate, and method for manufacturing strip-shaped film substrate having uneven pattern |
| US20150239232A1 (en) * | 2014-02-27 | 2015-08-27 | Eastman Kodak Company | System for reducing artifacts using tension control |
| JP2016087891A (en) * | 2014-10-31 | 2016-05-23 | 株式会社リコー | Recording device and method for control of recording device |
| CN105984206A (en) * | 2015-03-19 | 2016-10-05 | 海德堡印刷机械股份公司 | An additional page sensor |
| CN111065519A (en) * | 2017-07-26 | 2020-04-24 | 柯尼格及包尔公开股份有限公司 | Device for coating printed sheets, printing press and method for coating printed sheets |
| CN211493291U (en) * | 2019-12-19 | 2020-09-15 | 增和包装股份有限公司 | Smooth printing machine of cigarette pack printing |
| CN115302963A (en) * | 2022-10-12 | 2022-11-08 | 深圳市彩昇印刷机械有限公司 | Bar code printing control method, system and medium based on machine vision |
| CN115817039A (en) * | 2021-12-06 | 2023-03-21 | 南通吉美装饰材料有限公司 | Intaglio printing method based on ink-flying inhibition |
| DE102022109630A1 (en) * | 2022-04-21 | 2023-10-26 | Koenig & Bauer Ag | Method for operating an inkjet printing machine using at least one artificial neural network |
-
2023
- 2023-11-03 CN CN202311454393.9A patent/CN117183578B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1084455A (en) * | 1992-09-19 | 1994-03-30 | 林特克株式会社 | The detection modification method in printing equipment and printing state in this printing equipment and printing position |
| JPH09254364A (en) * | 1996-03-18 | 1997-09-30 | Kawashima Textile Manuf Ltd | Printing apparatus and printing roll |
| US20090020029A1 (en) * | 2007-07-13 | 2009-01-22 | Klaus Reckefuss | Method and apparatus for automatically regulating the registers between imprints in a multi color rotary printing press |
| CN102245389A (en) * | 2008-12-18 | 2011-11-16 | 利乐拉瓦尔集团及财务有限公司 | Method of controlling the relative position between printed patterns and non-printed patterns on a web-shaped material and a system used in the method |
| CN101574876A (en) * | 2009-06-03 | 2009-11-11 | 无锡光群雷射科技有限公司 | Laser holographic column gravure roller location printing paper |
| JP2011161662A (en) * | 2010-02-05 | 2011-08-25 | Lintec Corp | Printing position deviation detector, printer, and printing position deviation detecting method |
| JP2012240786A (en) * | 2011-05-19 | 2012-12-10 | Hitachi Ltd | Web carrying device |
| KR20130094982A (en) * | 2012-02-17 | 2013-08-27 | 한국기계연구원 | Register marking method and its device for printed electronics on roll-to-roll gravure offset printing press |
| TW201519962A (en) * | 2013-11-01 | 2015-06-01 | Jx Nippon Oil & Energy Corp | Application device for forming coating having discontinuous pattern onto strip-shaped film substrate, and method for manufacturing strip-shaped film substrate having uneven pattern |
| US20150239232A1 (en) * | 2014-02-27 | 2015-08-27 | Eastman Kodak Company | System for reducing artifacts using tension control |
| JP2016087891A (en) * | 2014-10-31 | 2016-05-23 | 株式会社リコー | Recording device and method for control of recording device |
| CN105984206A (en) * | 2015-03-19 | 2016-10-05 | 海德堡印刷机械股份公司 | An additional page sensor |
| CN111065519A (en) * | 2017-07-26 | 2020-04-24 | 柯尼格及包尔公开股份有限公司 | Device for coating printed sheets, printing press and method for coating printed sheets |
| CN211493291U (en) * | 2019-12-19 | 2020-09-15 | 增和包装股份有限公司 | Smooth printing machine of cigarette pack printing |
| CN115817039A (en) * | 2021-12-06 | 2023-03-21 | 南通吉美装饰材料有限公司 | Intaglio printing method based on ink-flying inhibition |
| DE102022109630A1 (en) * | 2022-04-21 | 2023-10-26 | Koenig & Bauer Ag | Method for operating an inkjet printing machine using at least one artificial neural network |
| CN115302963A (en) * | 2022-10-12 | 2022-11-08 | 深圳市彩昇印刷机械有限公司 | Bar code printing control method, system and medium based on machine vision |
Non-Patent Citations (2)
| Title |
|---|
| 徐富新;杨春艳;申冬玲;陈芳;: "基于机器视觉的果蝇复眼坏区甄别系统的图像处理设计", 生物医学工程研究, no. 02 * |
| 蔡叶菁, 文敦伟, 何静, 彭涛, 龙永红: "基于机器视觉的凹版印刷机位置控制及产品检测", 湖南大学学报(自然科学版), no. 03 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117183578B (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11630618B2 (en) | Correcting registration errors in digital printing | |
| CN107966447B (en) | workpiece surface defect detection method based on convolutional neural network | |
| US20170217152A1 (en) | Screen printing apparatus and methods | |
| US10976263B2 (en) | System and method for aligning an inker of a decorator | |
| CN106897996A (en) | chromatography error detection method based on machine vision | |
| CN205705753U (en) | Flat board Concha Arcae case digital printing equipment | |
| CN105751696A (en) | Flat-plate corrugated-container digital printing equipment | |
| CN104680509B (en) | A kind of real-time circular printing image defect detection method | |
| CN109142383B (en) | Character defect detection method based on morphology | |
| US20150243010A1 (en) | Image inspection device, image inspection system, and recording medium storing image inspection program | |
| CN117183578B (en) | Positioning detection method and printing method for telescopic printing substrate | |
| CN115009794B (en) | Full-automatic plate conveying production line and production control system thereof | |
| CN107745587A (en) | A kind of printing method of paper web products first ironing and then printing | |
| JP7337521B2 (en) | Image inspection with local image distortion correction | |
| ATE507044T1 (en) | METHOD FOR SIMULTANEOUSLY CUTTING PAPER AND OTHER GRAPHIC SUPPORTS ON A ROLL ALONG TWO VERTICAL AXES WITH AUTOMATIC ERROR CORRECTION | |
| CN103547453A (en) | System and method for adjusting and monitoring the pressures of printing rollers in a flexographic printing machine with central drum | |
| CN108920850A (en) | A kind of flexo pressure prediction method based on convolutional neural networks | |
| CN104417043A (en) | Multi-stage control and measurement of opaque white | |
| CN115861217A (en) | System and method for detecting defects of circuit board of backlight plate based on vision | |
| CN114677340A (en) | Concrete surface roughness detection method based on image edge | |
| US10434764B1 (en) | YAW measurement by spectral analysis | |
| CN107146272B (en) | Manufacturing method of automobile glass screen printing negative film | |
| CN207114447U (en) | Web feed printing product visual detection equipment | |
| CN113156832A (en) | Method and system for rapidly analyzing performance parameters of ink path system of offset press | |
| US9260263B2 (en) | Substrate aligning unit, substrate processing apparatus having the same, and method of aligning substrate using the same |
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 |