A pattern planning method for TFE inkjet printing and application thereof

Abstract

The present application belongs to the field of inkjet printing and flexible film packaging, and more particularly relates to a pattern planning method for TFE inkjet printing and application thereof, comprising: identifying pixel points in a bitmap for proofing located in a film edge region according to a preset film edge width and actually determined drop spacing in a lateral direction and a longitudinal direction of a substrate, fixing pixel values of the pixel points in the bitmap for proofing located in the film edge region, setting pixel values of other pixel points to non-printing corresponding pixel values, and obtaining an edge bitmap; forming a sparse bitmap with a size same as the substrate and a printing density same as a film edge sparse printing density according to a preset film edge sparse printing density, logically operating the edge bitmap and the sparse bitmap to obtain a new edge bitmap; and replacing pixel values of corresponding pixel points in the bitmap for proofing with pixel values of each pixel point in the new edge bitmap located in the film edge region, to obtain a printing pattern with edge sparsification. The present application can ensure the uniformity of the entire film thickness.

Description

Technical Field

[0001] This invention belongs to the field of inkjet printing and flexible thin film encapsulation, and more specifically, relates to a pattern planning method for TFE inkjet printing and its application. Background Technology

[0002] Organic light-emitting diodes (OLEDs) are a highly promising technology in luminescence science. The use of organic polymers extends the conjugated free electron system to achieve higher efficiency, flexibility, and stability. Compared to other lighting devices in the digital world, OLEDs have a wider range of applications, such as smart or high-quality television displays, computer monitors, smartwatch displays, various street signal lighting devices, residential lighting, and commercial applications. Thin Film Encapsulation (TFE) technology is an encapsulation technique that protects devices by creating a dense thin film on the device surface to block water and oxygen. TFE is formed by depositing and curing an inorganic / organic liquid onto a substrate. Its advantages include preventing edge penetration of moisture and oxygen, eliminating the need for adhesives, and adapting well to flexible substrates.

[0003] There are various processes for preparing TFE, including screen printing, vapor deposition, and spin coating. Inkjet printing is the preferred deposition technology for TFE preparation due to its material savings and ability to be combined with digital control. The two most critical requirements for thin-film encapsulation in inkjet printing are film thickness and film uniformity. If the film is too thin, its ability to isolate water and oxygen will be insufficient, while if it is too thick, it will waste material and increase the thickness and weight of the light-emitting device. The film uniformity will affect the consistency of light emission from the display device; uneven film layers can lead to varying degrees of Mura defects.

[0004] Because thin-film encapsulation involves droplet diffusion and fusion, it is related to the characteristics of the droplets themselves. Even if the droplet spacing is arranged exactly as ideally, the diffusion mechanism of the droplets will cause the film edges to bulge, forming a "dogear" effect. This bulge will cause uneven film thickness in the final cured film, which will not only affect the sealing properties and water and oxygen barrier properties of the film, but also cause various "mura" defects in the final display device. Summary of the Invention

[0005] In view of the shortcomings of the existing technology and the need for improvement, the present invention provides a pattern planning method for TFE inkjet printing and its application, the purpose of which is to avoid the problem of reduced film thickness uniformity caused by edge protrusions due to the diffusion characteristics of droplet ink.

[0006] To achieve the above objectives, according to one aspect of the present invention, a pattern planning method for TFE inkjet printing is provided, comprising:

[0007] Based on the preset film edge width and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the pixels located in the film edge region of the proofing bitmap are identified. The pixel values ​​of the pixels located in the film edge region of the proofing bitmap are fixed, and the pixel values ​​of other pixels are set to the corresponding pixel values ​​that will not be printed, thus obtaining the edge bitmap.

[0008] Based on the preset thin film edge sparse printing density, a sparse bitmap with the same size as the substrate and the same printing density as the thin film edge sparse printing density is formed. Logical operations are performed on the edge bitmap and the sparse bitmap to thin out the printing points in the thin film edge region according to the thin film edge sparse printing density, and a new edge bitmap is obtained.

[0009] The pixel values ​​of each pixel in the edge region of the film in the new edge bitmap are used to replace the pixel values ​​of the corresponding pixels in the proofing bitmap to obtain a sparser printing pattern and complete the pattern planning.

[0010] Furthermore, the bitmap used for prototyping is constructed in the following way:

[0011] Based on the required film coverage size and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the number of dots in the horizontal and vertical directions of the substrate is obtained, which is used as the number of pixels in the bitmap to be constructed for prototyping. The bitmap for prototyping is an image scaled relative to the known original bitmap of the required film.

[0012] Based on the values ​​of each pixel in the original bitmap, the initial pixel values ​​of each pixel in the proofing bitmap are determined, and based on a preset threshold for binarization, the final pixel values ​​of each pixel in the proofing bitmap are obtained, thus completing the construction of the proofing bitmap.

[0013] Furthermore, bilinear interpolation is used to determine the initial pixel values ​​of each pixel in the bitmap used for prototyping based on the values ​​of each pixel in the original bitmap.

[0014] Furthermore, the actually determined droplet spacing in the lateral and longitudinal directions of the substrate is obtained in the following way:

[0015] According to the equation: The scaled droplet spacing in the horizontal and vertical directions of the substrate is obtained as the actual determined droplet spacing in the horizontal and vertical directions of the substrate.

[0016] In the formula, H Target The target film thickness to be achieved in the current pattern planning is indicated by w and h, which represent the width and height dimensions of the substrate, respectively; x Pitch1 and yPitch1 These represent the droplet spacing in the horizontal and vertical directions of the scaled substrate, respectively; H Test This represents the minimum film thickness required to achieve the desired film uniformity under the current droplet material and substrate conditions; v and v1 are the single droplet volumes ejected from a single nozzle before and after scaling, respectively; x Pitch and y Pitch To achieve the desired film uniformity and the smaller film thickness H under the current droplet material and substrate conditions, respectively. Test The required droplet spacing H in the horizontal and vertical directions of the substrate Test x Pitch and y Pitch All of these are known quantities.

[0017] Furthermore, the smaller film thickness H Test As a preset value, a trained mapping model is used, based on the smaller film thickness H. Test And the required film uniformity, predict the corresponding single-droplet volume of a single nozzle and the droplet spacing x in the transverse and longitudinal directions of the substrate. Pitch and y Pitch The volume of a single droplet ejected from a single nozzle is used to guide inkjet printing.

[0018] Furthermore, the method for identifying pixels located in the film edge region of the bitmap used for proofing is specifically as follows:

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026] In the formula, L Edge d represents the preset film edge width. PixelGapY d represents the actual determined droplet spacing along the longitudinal direction of the substrate. PixelGapX Y represents the actual determined droplet spacing in the lateral direction of the substrate. EdgeNum X represents the number of dots along the vertical direction of the substrate. EdgeNum This indicates the number of dots in the horizontal direction of the substrate. This indicates that the i+Yth bit in the bitmap used for prototyping EdgeNum The pixel value in row j-th column, This indicates that the bitmap used for prototyping is in the i-th row and j-th position. EdgeNum Column pixel values, This indicates that the i-th Y-th bit in the bitmap used for prototyping EdgeNum The pixel value in row j-th column, This indicates that the bitmap used for prototyping is in the i-th row and j+X-th position. EdgeNum The column's pixel values, abs(·) represents taking the absolute value, C Up C Down C Left and C Right All are intermediate variables.

[0027] The present invention also provides a TFE inkjet printing method, which uses a printing pattern planned by the patterning planning method described above to perform inkjet printing.

[0028] Furthermore, it also includes:

[0029] Collect the film thickness of m×n regions in the film obtained after the current printing is completed;

[0030] Based on the deviation between the film thickness of each region and the target film thickness, a gray value matrix is ​​constructed to adjust the printing density of each region to achieve the target film thickness. In this gray value matrix, there is a known correspondence between each gray value element, the required printing density, and the corresponding film thickness.

[0031] Based on each gray value element in the gray value matrix and the corresponding relationship, the adjusted printing density of each region is determined. The printing density includes the number of printing dots in the horizontal and vertical directions of the substrate.

[0032] Perform the next print based on the stated print density.

[0033] Furthermore, the grayscale matrix used to adjust the printing density of each region to achieve the target film thickness is as follows:

[0034]

[0035] In the formula, h Thick,11 h Thick,1n h Thick,m1 h Thick,mn H represents the film thickness in regions 11, 1n, m1, and mn, respectively. Target H represents the target film thickness. min H max These represent the film thicknesses corresponding to the minimum to maximum printing densities that meet the film uniformity requirements, obtained experimentally. Their values ​​are defined as 0 and 255, respectively, and are defined as H... min H max The film thickness data is within the range of 0-255; Val GreyThe grayscale value corresponding to the target film thickness is defined as:

[0036] The present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed by a processor, it controls the device on which the storage medium is located to perform a pattern planning method for TFE inkjet printing as described above.

[0037] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects:

[0038] (1) This invention proposes a pattern planning method for TFE inkjet printing, which adjusts the edge printing density based on a bitmap. Specifically, firstly, based on the pixel interval in the existing proofing bitmap and the preset film edge width, pixel-by-pixel judgment is performed to extract pixels located on the film edge, obtaining an edge bitmap. Then, logical operations are performed on this bitmap with a pattern of a preset sparse ratio to obtain a sparse bitmap with sparsified edges. The sparse bitmap is then combined with the existing proofing bitmap to obtain the printing pixel matrix, i.e., the edge-sparsed printing pattern. This processing method effectively avoids the edge protrusion problem caused by the diffusion characteristics of some droplets of ink, improves the uniformity of film thickness, and uses an image-based edge extraction method, which greatly improves efficiency and accuracy compared to directly sampling the printing matrix.

[0039] (2) The present invention introduces a smaller film thickness H corresponding to the desired film uniformity under current droplet material and substrate conditions. Test And the corresponding droplet spacing x in the horizontal and vertical directions of the substrate. Pitch and y Pitch As a known and universal quantity, it can be directly called each time the method of the present invention is executed to determine the actual droplet spacing in the horizontal and vertical directions of the substrate, in combination with the actual required target film thickness and droplet volume, to scale the droplet spacing in the horizontal and vertical directions of the substrate, which is convenient and fast.

[0040] (3) Use bilinear interpolation to process the pixel values ​​of each pixel in the scaled bitmap to ensure that the image is not distorted after conversion, and set a threshold to make the grayscale image better converted into a binary image.

[0041] (4) By employing machine learning, a neural network model with multiple input and output parameters is constructed and trained using limited experimental data to predict the input parameters corresponding to the predetermined output. This facilitates the determination of the parameter process range, enabling the achievement of the required film uniformity and a smaller film thickness H under the current droplet material and substrate conditions. Test The required droplet spacing in the horizontal and vertical directions of the substrate reduces redundant work in manually determining parameters, improving efficiency while also increasing the accuracy of parameter selection.

[0042] (5) Edge pixels are identified by performing a loop logic check on each pixel, and an edge extraction bitmap is generated. This method takes into account the difference in actual droplet diffusion distance in two directions, and can identify edge pixels that conform to the actual situation. At the same time, it comprehensively considers the four directions of up, down, left and right, which can effectively avoid incorrect judgments.

[0043] (6) In the process of inkjet printing, the present invention also proposes to measure the actual printed film thickness in different areas after printing once, and to represent the film thickness with gray values. The gray matrix is ​​reset according to the deviation between the film thickness of each area and the target film thickness, and the printing density of each area is adjusted in the next printing to compensate for the deviation. This can better improve the film thickness uniformity according to the actual printing effect. Attached Figure Description

[0044] Figure 1 A flowchart of a pattern planning method for TFE inkjet printing provided in an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of image edge sparsity ratio synthesis provided in an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of image edge extraction provided in an embodiment of the present invention;

[0047] Figure 4 This is a scaled image pixel bilinear interpolation value map provided in an embodiment of the present invention;

[0048] Figure 5 This is a flowchart of the input and output of the mapping model provided in an embodiment of the present invention;

[0049] Figure 6 This is a schematic diagram of the printhead and pixel substrate model and the inkjet printing process provided in an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of grayscale compensation provided for an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0052] Example 1

[0053] A pattern planning method for TFE inkjet printing, such as Figure 1 As shown, it includes:

[0054] Based on the preset film edge width and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the pixels located in the film edge region of the proofing bitmap are identified. The pixel values ​​of the pixels located in the film edge region of the proofing bitmap are fixed, and the pixel values ​​of other pixels are set to the corresponding pixel values ​​that will not be printed, thus obtaining the edge bitmap.

[0055] Based on the preset thin film edge sparse printing density, a sparse bitmap with the same size as the substrate and the same printing density as the thin film edge sparse printing density is formed. Logical operations are performed on the edge bitmap and the sparse bitmap to thin out the printing points in the thin film edge region according to the thin film edge sparse printing density, and a new edge bitmap is obtained.

[0056] The pixel values ​​of each pixel in the edge region of the film in the new edge bitmap are used to replace the pixel values ​​of the corresponding pixels in the above proofing bitmap to obtain a sparser printing pattern and complete the pattern planning.

[0057] For most printing inks, even if the droplets are printed accurately and evenly spaced according to the planned dot matrix, the characteristics of the droplets themselves will cause aggregated protrusions at the edges of the fused film, a phenomenon known as the "dogear" effect, which severely affects the overall film quality. Therefore, in actual printing, it is necessary to change the edge printing density, and the variables to be adjusted include the edge width and the edge dot density.

[0058] This embodiment proposes a pattern planning method that adjusts the edge printing density based on a bitmap. It employs BMP bitmap planning for printing, more effectively utilizing the binary information on the bitmap and effectively avoiding aggregated protrusions at the film edges. Specifically, based on the pixel intervals and set edge widths in the existing proofing bitmap, edge pixels are extracted pixel by pixel and logically operated (e.g., AND operation) with the sparsed pattern. The sparsed edges are then combined with the original image to obtain the printing pixel matrix, i.e., the edge-sparsed printing pattern. This processing method effectively avoids edge protrusion problems caused by the diffusion characteristics of some droplets of ink, improves the uniformity of film thickness, and uses an image-based edge extraction method, which greatly improves efficiency and accuracy compared to directly sampling points from the printing matrix.

[0059] Among them, regarding the printing points in the edge region of the sparse film, such as Figure 2As shown in the figure, the leftmost part is the extracted edge bitmap. The "percentage sparsity pattern" in the figure is the sparse bitmap mentioned above. The figure shows three sparse bitmaps, reflecting different degrees of sparsity. In the figure, the edge bitmap and the sparse bitmap are ANDed to sparsify the printing points in the edge region of the film according to the sparse printing density of the film edge, resulting in a new edge bitmap (i.e., the "edge operation result" in the figure). The new edge bitmap is then combined with the proofing bitmap, that is, the pixel values ​​of each pixel in the new edge bitmap located in the edge region of the film replace the pixel values ​​of the corresponding pixels in the proofing bitmap, to obtain the final printing pixel matrix, i.e., the edge-sparsed printing pattern. Figure 2 In the process, after extracting the edges, edge sampling is performed according to a set percentage. The specific percentage should be chosen based on the desired printing effect and experience. As shown in the figure, comparing the corresponding percentage sparsification patterns (three percentage examples are given here: 25%, 50%, and 75%), the number of pixels in the width and height of this sparsification pattern is the same as the edge BMP bitmap. In the edge BMP bitmap, white pixels remain unchanged, while black pixels are ANDed with pixels at the same position as the sparsification percentage pattern. If they match, they are set to black; otherwise, they are set to white, resulting in a sparsified edge bitmap. Finally, the printing ratio is combined to obtain the final pixel matrix to be printed.

[0060] More specifically, the sparsification matrix in this figure is obtained through an AND operation. Based on a pre-set fill percentage, the extracted edge matrix and the percentage matrix are ANDed to obtain the sparse matrix.

[0061]

[0062] The above formula is the sparse edge matrix obtained by AND operation between the original image edge and 50% of the printed pattern, resulting in the updated bitmap.

[0063] Since the edge widths of the substrate are fixed in both the horizontal and vertical directions, but the pixel spacing in the two directions is different, the number of pixels extracted is also different. In this embodiment, the extraction method can be implemented in any conceivable manner, but it may be considered a preferred implementation, such as... Figure 3 As shown in this embodiment, the method for identifying pixels located in the film edge region of the bitmap used for proofing is specifically as follows:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] In the formula, L Edge d represents the preset film edge width. PixelGapY d represents the actual determined droplet spacing along the longitudinal direction of the substrate. PixelGapX Y represents the actual determined droplet spacing in the lateral direction of the substrate. EdgeNum X represents the number of dots along the vertical direction of the substrate. EdgeNum This indicates the number of dots in the horizontal direction of the substrate. This indicates that the i+Yth bit in the bitmap used for prototyping EdgeNum The pixel value in row j-th column, This indicates that the bitmap used for prototyping is in the i-th row and j-th position. EdgeNum Column pixel values, This indicates that the i-th Y-th bit in the bitmap used for prototyping EdgeNum The pixel value in row j-th column, This indicates that the bitmap used for prototyping is in the i-th row and j+X-th position. EdgeNum The column's pixel values, abs(·) represents taking the absolute value, C Up C Down C Left and C Right All are intermediate variables.

[0072] It should be noted that film uniformity can be defined according to the process standard as follows:

[0073]

[0074] In the formula, H0 is the film thickness matrix measured by dividing the printed film into regions.

[0075] In this embodiment, the bitmap used for prototyping can be implemented using other methods, which can be considered as a preferred implementation. The aforementioned bitmap for prototyping is constructed in the following way:

[0076] Based on the required film coverage size and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the number of dots in the horizontal and vertical directions of the substrate is obtained, which is used as the number of pixels in the bitmap to be constructed for prototyping. The bitmap for prototyping is an image scaled relative to the known original bitmap of the required film.

[0077] Based on the values ​​of each pixel in the original bitmap, the initial pixel values ​​of each pixel in the proofing bitmap are determined, and based on the preset threshold for binarization, the final pixel values ​​of each pixel in the proofing bitmap are obtained, thus completing the construction of the proofing bitmap.

[0078] This preferred method constructs a bitmap for proofing by scaling the original bitmap and determining the pixel values ​​based on the original bitmap pixel values. For BMP bitmap representation, the advantage of the BMP bitmap image format is its lossless nature; its stored binary information can be easily used to plan the printing dot matrix. White pixels (255) and black pixels (0) correspond to 0 and 1 in inkjet printing, respectively, meaning one drop is printed or no print is made. The screen displays pixels with equal spacing in both width and height. However, due to the diffusion characteristics of the droplets, the diffusion range in the two directions is different during actual printing; that is, the pixels are not necessarily square. Based on the actual determined X and Y spacing of the droplets, the actual number of pixels in the X and Y directions can be obtained according to the film coverage size.

[0079] There are many ways to determine pixel values ​​by combining the pixel values ​​of the original bitmap. Considering that bilinear interpolation is simple and effective, it can be used as a preferred implementation method. In this embodiment, bilinear interpolation is used to determine the initial pixel values ​​of each pixel in the bitmap for prototyping based on the values ​​of each pixel in the original bitmap.

[0080] Regarding bilinear interpolation, such as Figure 4 As shown, assuming the values ​​of function f at four points Q11 = (x1, y1), Q12 = (x1, y2), Q21 = (x2, y1), and Q22 = (x2, y2) are known, to obtain the value of the unknown function f at point P = (x, y), we first interpolate in the X direction, obtaining:

[0081]

[0082]

[0083] Then, interpolation is performed in the Y direction to obtain:

[0084]

[0085] First, we obtain the original bitmap `src`, with h×w pixels in both the height and width directions, and indexes `p×q`, where `p∈[1,h]`, `q∈[1,w]`. The scaled bitmap is then denoted as `dst`, with m×n pixels in both the height and width directions, and indexes `i×j`, where `i∈[1,m]`, `j∈[1,n]`. The process of addressing the original bitmap `src[p][q]` using the index at `dst[i][j]` is as follows:

[0086]

[0087]

[0088] Here, `floor` is the floor function. After obtaining the values ​​of `p` and `q`, the pixel values ​​of the four points `src[p][q]`, `src[p][q+1]`, `src[p+1][q]`, and `src[p+1][q+1]` are substituted into the aforementioned bilinear interpolation formula to obtain the pixel value at `dst[i][j]`. By adding / subtracting 0.5 to the formula, the geometric centers of the two images before and after scaling are aligned, avoiding the problem that some pixels on the image edges are not actually involved in the calculation.

[0089] For inkjet printing, each pixel is categorized as either printed or unprinted. Printing one drop sets the value to 1, otherwise it's set to 0. From the image display perspective, this results in black and white, i.e., a binary image. The aforementioned bilinear interpolation method produces an image where each pixel value ranges from 0 to 255, i.e., a grayscale image. Therefore, the conversion from grayscale to binary involves setting a threshold. Above this threshold, the pixel represents pure black (printed) and the grayscale value is set to 0; otherwise, it represents pure white (unprinted) and the grayscale value is set to 255.

[0090] For example, f(i',j') and f(i,j) are the pixel values ​​at (i,j) before and after the transformation, respectively, and 127 is the threshold value, which is determined based on the actual effect.

[0091] Furthermore, the droplet spacing in the horizontal and vertical directions of the substrate, as actually determined above, can be determined based on actual conditions. This embodiment does not limit this, but it can be considered a preferred implementation. The droplet spacing in the horizontal and vertical directions of the substrate, as actually determined above, can be obtained in the following ways:

[0092] According to the equation: The scaled droplet spacing in the horizontal and vertical directions of the substrate is obtained and used as the actually determined droplet spacing in the horizontal and vertical directions of the substrate; where H Target The target film thickness to be achieved in the current pattern planning is indicated by w and h, which represent the width and height dimensions of the substrate, respectively; x Pitch1 and y Pitch1 These represent the droplet spacing in the horizontal and vertical directions of the scaled substrate, respectively; H Test This represents the minimum film thickness required to achieve the desired film uniformity under the current droplet material and substrate conditions; x Pitch and y Pitch To achieve the desired film uniformity and the smaller film thickness H under the current droplet material and substrate conditions, respectively. Test The required droplet spacing H in the horizontal and vertical directions of the substrate Test x Pitch and y Pitch All of these are known quantities.

[0093] This preferred method proposes a known, universal H Test x Pitch and y Pitch These quantities are treated as known quantities and can be directly called in each execution of the method in this embodiment to scale the droplet spacing in the horizontal and vertical directions of the substrate, which is convenient and quick.

[0094] Regarding H Test x Pitch and y Pitch The acquisition of the aforementioned smaller film thickness H can be considered as a preferred embodiment. Test As a preset value, a trained mapping model is used, based on a smaller membrane thickness H. Test And the required film uniformity, predict the corresponding single-droplet volume of a single nozzle and the droplet spacing x in the transverse and longitudinal directions of the substrate. Pitch and y Pitch The volume of a single droplet ejected from a single nozzle is used to guide inkjet printing.

[0095] In other words, this preferred method involves experimental analysis of ink diffusion characteristics (by testing the diffusion characteristics of the sample ink, including spreading radius, viscosity, contact angle, etc., setting a range of sample parameters, and selecting different combinations of values ​​within each parameter range), controlling variables to conduct various fusion film formation experiments, using instrument measurement results, and applying machine learning algorithms to the experimental data to obtain suitable input parameters. The mapping model is used for mapping, which is convenient, fast, and ensures accuracy.

[0096] In engineering applications, the main adjustable factors affecting the uniformity and thickness of TFE films are the droplet spacing and voltage magnitude in the X and Y directions, with the voltage magnitude primarily affecting the size of a single droplet. Here, the droplet spacing and droplet size in the X and Y directions are the inputs, and film thickness and film uniformity are the outputs. To construct a model of the relationship between the inputs and outputs, and to predict the inputs under the target output, specifically, as... Figure 5 As shown, an MLP (Multilayer Neural Network) can be used, which includes an input layer, an output layer, and a hidden layer. For example, the neural network used includes a three-layer structure, with an output and input layer and a hidden layer. The input layer has three neurons, corresponding to three inputs: the X and Y direction spacing of the droplets (i.e., the droplet spacing in the horizontal and vertical directions of the substrate) and the droplet volume. The output layer outputs the film thickness and film uniformity.

[0097] The neuron activation function is chosen to be the sigmoid function.

[0098]

[0099] The loss function is as follows:

[0100]

[0101] Where f is the activation function, and the loss function is defined as follows:

[0102]

[0103] Where i is the sample size, obtained from the previous experimental step. The three variables are the input parameters, namely the distance between the droplets in the X and Y directions and the size of the droplet volume.

[0104] The optimization algorithm chosen is the steepest descent method (preferred) from the gradient-based algorithms. Unlike gradient descent, the learning rate is not fixed in each iteration, thus ensuring that the iteration rate is neither too large nor too small. The parameters w and b for each iteration are as follows:

[0105]

[0106]

[0107] K represents the number of iterations. The learning rate for the k-th iteration is obtained by substituting the w parameter of the (k+1)-th iteration into the loss function and solving it using any one-dimensional optimization method.

[0108] After collecting the experimental data, it needs to be classified. The specific process is as follows: divide the entire dataset into K parts, take one part for model evaluation each time, and use the remaining (K-1) parts to train the model. The model evaluation uses the coefficient of determination R. 2 Two regression model evaluation metrics, namely, mean square root absolute error, are used to evaluate the predictive performance:

[0109]

[0110]

[0111]

[0112] In the formula, y i and These are the predicted and actual values ​​of the output, respectively. R 2 A score closer to 1 indicates better model predictive performance. RMSE is an indicator of model error.

[0113] For example, such as Figure 5The diagram shows the flowchart for predicting input parameters using a neural network model to control the uniformity of TFE printing film thickness. x1, x2, and x3 are the input parameters, representing the droplet spacing and droplet volume in the X and Y directions, respectively. y1 and y2 are the output parameters, representing the film thickness and film uniformity, respectively. The left-hand module collects experimental data for training the model; the superscript i∈[1,n] indicates the number of data sets. The middle section shows the neural network structure, including an input layer, one hidden layer, and one output layer, with 3, 5, and 2 neurons respectively.

[0114] Under the premise that the film uniformity is within an acceptable range, the trained neural network model is used to select the minimum film thickness as much as possible to broaden the range of process parameters. At the same time, the spacing in the X and Y directions is kept as small as possible. After obtaining the appropriate X and Y spacing and droplet volume, it is necessary to scale the parameters, the target film thickness and the actual droplet volume to obtain the actual X and Y spacing for printing, so as to facilitate image scaling.

[0115] Example 2

[0116] A TFE inkjet printing method uses a printing pattern planned by the patterning planning method described above to perform inkjet printing.

[0117] In actual printing, it is necessary to first perform parametric modeling of the printhead and pixel substrate, and then plan the printhead ignition timing and path to proceed with inkjet printing.

[0118] For example, a method for parametric modeling of inkjet printer nozzles and pixel substrates is presented. There are many types of nozzles, and the physical coordinate arrangement of the nozzles on each nozzle is different. However, due to the built-in real-time processing algorithm within the nozzle, the nozzles on the nozzle can be linearly processed according to the order of their Y-coordinate size. Only the nozzle angle, nozzle spacing, and the coordinates of the first nozzle are needed to set the coordinates of all nozzles. The pixel model is divided into different large areas and different panels; parametric modeling can adapt to substrates with different arrangements.

[0119] The nozzle coordinates are determined as follows:

[0120] X Nozzle,i =X Nozzle,1

[0121] Y Nozzle,i =Y Nozzle,1 +i*d NozzleGapY

[0122]

[0123] X Nozzle,i and Y Nozzle,i The X and Y coordinates of the i-th nozzle are respectively, and d NozzleGapYFor nozzle spacing, d NozzleGapY This represents the number of nozzles. If the nozzle has an angle, the coordinates are determined as follows:

[0124]

[0125] X Nozzle,iR and Y Nozzle,iR Let X and Y be the coordinates of the i-th nozzle after rotation, and θ be the nozzle angle.

[0126] The pixel substrate model is described from two levels: the printing area level and the pixel arrangement level. The pixel coordinates can be represented as:

[0127]

[0128]

[0129]

[0130]

[0131] Among them, X Pixel,pq Y Pixel,pq For the p-th column of the substrate pixel coordinate model (p∈[1,N) Row The qth row (q∈[1,N) Col ]) Pixel X and Y coordinate values, floor is the floor function, N PRow N PCol d represents the number of pixel rows and columns in a single printed area. PAGapX d PAGapY d represents the spacing between the X and Y axes of the printing area. PixelGapX d PixelGapY N represents the pixel spacing in the X and Y directions within a single printed area. Row N Col These represent the number of rows and columns in the substrate pixel coordinate model, respectively. When the pixel substrate has an angle, the pixel coordinates are:

[0132]

[0133] The first of the arrayed pixel substrate model Line 1 The coordinate array after rotating the column pixel coordinate matrix For the first Line 1 Column panel angle, For the first Line 1 The original coordinate matrix of the columns of pixels, with ceil being the floor function. Based on these parameters, any form of pixel arrangement and substrate arrangement can be represented parametrically by the model.

[0134] After being linearized into a single-line nozzle, the nozzle is composed of the following parameters:

[0135]

[0136] The substrate pixel model consists of two layers: a printed area array and a pixel array.

[0137] The abstract model of substrate pixel array parameters can be described using the following parameters:

[0138]

[0139]

[0140] Each parameter as follows Figure 6 As shown. The printing process is carried out in sections, with the lower diagram illustrating the positional relationship between the inkjet printhead and the substrate. Each time, the printhead moves a certain distance in the Y direction, while the substrate moves back and forth once in the X direction. During this movement, the printhead ejects ink according to the ignition control command, depositing it into individual pixel pits on the substrate. Since TFE printing does not have pixel pits, these pixel pits can represent the diffusion range of the droplets.

[0141] Figure 6 In the lower part of the diagram, the colored pixels on the pixel substrate represent printed pixels. Taking a circle as an example, the gray colored area represents the edge of the circle after sparsening (the sparse pattern here should be a black and white pixel pattern, but it is represented by gray for easy and obvious display).

[0142] By parametrically modeling and linearly processing the arrayed printheads in the inkjet printing display system, the system is freed from the limitations of fixed hardware printheads. Similarly, the substrate pixel array is abstracted into a parametric model, consisting of two parts: the printing area layer and the pixel layer. Through this abstract model representation of the printhead and pixel substrate, the system's adaptability is increased, making it suitable for printheads with different arrangements and splicing methods, as well as substrates with different arrangements and shapes.

[0143] In addition, before printing, positioning and testing are performed. The volume values ​​of all nozzles of the printhead and their deviation values ​​from the reference are observed and measured by the top-view camera and the bottom-view camera respectively, and nozzles that do not meet the requirements are filtered out.

[0144] The relevant technical solutions are the same as in Embodiment 1, and will not be repeated here.

[0145] Furthermore, considering that even if droplets are printed in a lattice pattern with ideal spacing and sparse edges, the film thickness will still be uneven due to the diffusion characteristics of the droplets, it is necessary to divide different regions to measure the film thickness according to the actual printing situation, and set the grayscale based on the deviation value to adjust the droplet density of the printed region in the next printing. Therefore, this can be considered a preferred implementation method, such as... Figure 7 As shown, it also includes:

[0146] (1) Collect the film thickness of m×n regions in the film obtained after the current printing is completed; that is, after printing the film according to the pixel matrix after edge sparsification, the film thickness can be measured in m×n regions using a step meter.

[0147] (2) Based on the deviation between the film thickness and the target film thickness in each region, a grayscale value matrix is ​​constructed to adjust the printing density of each region to achieve the target film thickness. In this grayscale value matrix, there is a known correspondence between each grayscale value element, the required printing density, and the corresponding film thickness. That is, based on the deviation between the actual film thickness and the target film thickness, the grayscale value of each region is set, the printing density is adjusted, and the uniformity of the next printing is optimized. This grayscale value matrix is ​​the grayscale value of the target film thickness for the next printing. The spacing between the droplets in the X and Y directions is scaled according to this grayscale value to adjust the printing density and compensate for the deviation.

[0148] (3) Based on each gray value element in the gray value matrix and the above correspondence, determine the adjusted printing density of each region. The printing density includes the number of printing dots in the horizontal and vertical directions of the substrate.

[0149] (4) Based on the print density, perform the next print.

[0150] This preferred method uses negative feedback to regulate the next print, which can be adapted to different nozzles and substrate models, and effectively improves the uniformity of TFE film formation.

[0151] As a further preferred embodiment, the grayscale matrix used to adjust the printing density of each region to achieve the target film thickness is as follows:

[0152]

[0153] In the formula, h Thick,11 h Thick,1n h Thick,m1 h Thick,mn H represents the film thickness in regions 11, 1n, m1, and mn, respectively. Target H represents the target film thickness. min H max These represent the film thicknesses corresponding to the minimum to maximum printing densities that meet the film uniformity requirements, obtained experimentally. Their values ​​are defined as 0 and 255, respectively, and are defined as H...min H max The film thickness data is within the range of 0-255; Val Grey The grayscale value corresponding to the target film thickness is defined as:

[0154] This invention proposes a pattern planning method for TFE inkjet printing. Currently, there is a lack of patterning planning and control schemes adapted to various resolutions for TFE inkjet printing. Firstly, the statistical analysis of ink diffusion characteristics on film uniformity is not fully considered; different droplet spacing and volume need to be adjusted to improve uniformity for different droplets. Secondly, there is a lack of a scheme to control the printing process using information from the image after scaling and edge processing of the bitmap. Finally, factors such as printhead wear and environmental changes during the printing process that cause uneven printing in subsequent cycles do not consider using a negative feedback process to guide the next print. In summary, this invention provides a comprehensive solution for controlling the uniformity of TFE film formation, addressing all the above-mentioned problems.

[0155] Example 3

[0156] A computer-readable storage medium includes a stored computer program, wherein when the computer program is executed by a processor, it controls the device where the storage medium is located to perform a pattern planning method for TFE inkjet printing as described in Embodiment 1 above.

[0157] The relevant technical solutions are the same as in Embodiment 1, and will not be repeated here.

[0158] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A pattern planning method for TFE inkjet printing, characterized in that, include: Based on the preset film edge width and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the pixels located in the film edge region of the proofing bitmap are identified. The pixel values ​​of the pixels located in the film edge region of the proofing bitmap are fixed, and the pixel values ​​of other pixels are set to the corresponding pixel values ​​that will not be printed, thus obtaining the edge bitmap. The proofing bitmap is an image scaled relative to the known original bitmap of the required film. Based on the preset thin film edge sparse printing density, a sparse bitmap with the same size as the substrate and the same printing density as the thin film edge sparse printing density is formed. Logical operations are performed on the edge bitmap and the sparse bitmap to thin out the printing points in the thin film edge region according to the thin film edge sparse printing density, and a new edge bitmap is obtained. The pixel values ​​of each pixel in the edge region of the film in the new edge bitmap are used to replace the pixel values ​​of the corresponding pixels in the proofing bitmap to obtain a printing pattern with sparsed edges, thus completing the pattern planning. The actual determined droplet spacing in the horizontal and vertical directions of the substrate is obtained through the following method: According to the equation: The scaled droplet spacing in the horizontal and vertical directions of the substrate is obtained and used as the actual determined droplet spacing in the horizontal and vertical directions of the substrate. In the formula, This indicates the target film thickness that needs to be achieved in the current pattern planning; , These represent the width and height dimensions of the substrate, respectively. and These refer to the droplet spacing in the horizontal and vertical directions of the scaled substrate, respectively. This indicates the minimum film thickness required to achieve the desired film uniformity under the current droplet material and substrate conditions; and These represent the volume of a single droplet ejected from a single nozzle before and after scaling. and To achieve the desired film uniformity and the smaller film thickness under the current droplet material and substrate conditions, respectively. The required droplet spacing in the horizontal and vertical directions of the substrate. , and All of these are known quantities.

2. The pattern planning method according to claim 1, characterized in that, The bitmap used for prototyping is constructed in the following way: Based on the required film coverage size and the actual determined droplet spacing in the horizontal and vertical directions of the substrate, the number of dots in the horizontal and vertical directions of the substrate is obtained, which is used as the number of pixels in the bitmap to be constructed for prototyping. Based on the values ​​of each pixel in the original bitmap, the initial pixel values ​​of each pixel in the proofing bitmap are determined, and based on a preset threshold for binarization, the final pixel values ​​of each pixel in the proofing bitmap are obtained, thus completing the construction of the proofing bitmap.

3. The pattern planning method according to claim 2, characterized in that, Using bilinear interpolation, the initial pixel values ​​of each pixel in the bitmap used for prototyping are determined based on the values ​​of each pixel in the original bitmap.

4. The pattern planning method according to claim 1, characterized in that, The smaller film thickness As a preset value, a trained mapping model is used, based on the smaller film thickness. And based on the desired film uniformity, predict the corresponding single-droplet volume per nozzle and the droplet spacing in the transverse and longitudinal directions of the substrate. and The volume of a single droplet ejected from a single nozzle is used to guide inkjet printing.

5. The pattern planning method according to claim 1, characterized in that, The method for identifying pixels located in the film edge region of the bitmap used for proofing is specifically as follows: In the formula, This indicates the preset film edge width. This indicates the actual, determined droplet spacing along the longitudinal direction of the substrate. This indicates the actual determined droplet spacing on the lateral side of the substrate. Indicates the number of dots along the longitudinal direction of the substrate. This indicates the number of dots in the horizontal direction of the substrate. The bitmap used for prototyping represents the first... Line 1 Column pixel values, The bitmap used for prototyping represents the first... Line 1 Column pixel values, The bitmap used for prototyping represents the first... Line 1 Column pixel values, The bitmap used for prototyping represents the first... Line 1 Column pixel values, This indicates taking the absolute value. , , and All are intermediate variables; , These represent the total number of rows and columns of pixels in the bitmap used for proofing, respectively.

6. A TFE inkjet printing method, characterized in that, The printing pattern planned by the pattern planning method as described in any one of claims 1 to 5 is used for inkjet printing.

7. A TFE inkjet printing method according to claim 6, characterized in that, Also includes: Collect the film thickness of m×n regions in the film obtained after the current printing is completed; Based on the deviation between the film thickness of each region and the target film thickness, a gray value matrix is ​​constructed to adjust the printing density of each region to achieve the target film thickness. In this gray value matrix, there is a known correspondence between each gray value element, the required printing density, and the corresponding film thickness. Based on each gray value element in the gray value matrix and the corresponding relationship, the adjusted printing density of each region is determined. The printing density includes the number of printing dots in the horizontal and vertical directions of the substrate. Perform the next print based on the stated print density.

8. A TFE inkjet printing method according to claim 7, characterized in that, The grayscale matrix used to adjust the printing density of each region to achieve the target film thickness is as follows: ； In the formula, , , , Representing regions 11 and 1 respectively n , m 1. mn The film thickness on This indicates the target film thickness. , These represent the film thicknesses corresponding to the minimum and maximum printing densities that meet the film uniformity requirements, obtained experimentally. Their values ​​are defined as 0 and 255, respectively, and are defined as being located at... , The film thickness is within the range of 0-255. The grayscale value corresponding to the target film thickness is defined as: .

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed by a processor, it controls the device on which the storage medium is located to perform a pattern planning method for TFE inkjet printing as described in any one of claims 1 to 5.

Images