Splicing precision improving method and equipment based on digital photoetching technology

Abstract

The invention discloses a splicing precision improving method and equipment based on a digital photoetching technology, and belongs to the field of digital photoetching. The method comprises the following steps: a calibration step: calibrating an imaging range and an overlapping region according to DMD inclination direction arrangement of each imaging module, and determining a graph cutting initial position; in the distribution step, the strip width distributed by each lens is calculated, the strip width is determined based on the difference value of the position coordinates of the adjacent lenses and the overlapping allowance, and the overlapping allowance comprises at least two times of the width of the triangular exposure area; and a processing step: cutting the original graphic data according to the graphic cutting initial position and the strip width, deleting areas with the width of a triangular exposure area on each of the left side and the right side of the cut graphic data, replacing the areas with blank graphic data, filling the blank graphic data on the right side to the maximum effective exposure width of the DMD, and generating exposure graphic data of each lens. According to the invention, the splicing debugging of the multi-path DMD exposure pattern is simplified, and the splicing precision and the system stability are improved.

Description

Technical Field

[0001] This invention relates to the field of digital lithography technology, and more specifically, to a method and apparatus for improving the stitching accuracy of multi-channel DMD exposure patterns. Background Technology

[0002] Digital lithography is an advanced lithography technology that uses a digital micromirror device (DMD) as a spatial light modulator to expose computer-generated graphic data onto a photosensitive material through a projection optical system. In large-format lithography applications, because the exposure area of ​​a single DMD is limited, multiple imaging modules (each module containing a DMD and its corresponding optical system) are usually used for stitching together to expand the exposure area.

[0003] In existing technologies, the stitching of multi-channel DMD exposure patterns typically employs the following scheme: During physical exposure, the exposure areas allocated to each DMD overlap by a certain width. By setting a fixed binary template for the DMD, the edge pixels of the corresponding overlapping areas are always kept in a non-flipped state (i.e., no light is emitted). The stitching of multi-channel DMD exposure patterns is achieved by adjusting the parameter combination of the stitching area in the template. Simultaneously, the exposure energy of pixels in non-stitching areas is adjusted to minimize any stitching artifacts that may appear at the stitching points.

[0004] However, the aforementioned existing technical solutions have the following technical drawbacks: First, the debugging process is complex and requires numerous attempts. Since the splicing effect depends on the combination and adjustment of multiple parameters in the splicing template, and these parameters influence each other, the debugging process requires repeated trials. In practical applications, when more than four DMDs need to be spliced, the number of parameter combinations increases exponentially, and a single debugging session may require dozens or even hundreds of attempts. This heavily relies on the operator's experience, resulting in low debugging efficiency and hindering industrial production.

[0005] Secondly, the technology suffers from poor stability and low stitching accuracy. Current technology uses a fixed binary template to control the flipping state of the edge micromirrors, a "static" control method. This static template cannot detect actual changes in the lens's positional relationship. When the lens undergoes slight displacement due to factors such as temperature changes, mechanical vibration, or equipment transportation (especially when the displacement is less than one pixel), the micromirrors that were originally set not to flip may no longer be located in the actual physical overlap area, resulting in overlapping or broken patterns at the stitching point. This severely affects the integrity of the exposure pattern and the stitching accuracy. In this case, the template parameters need to be readjusted, further increasing the debugging workload.

[0006] Therefore, how to simplify the stitching and debugging process of multi-channel DMD exposure patterns and achieve adaptive adjustment when the lens position relationship changes, thereby improving stitching accuracy and system stability, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] To address the aforementioned deficiencies in existing technologies, this invention provides a method and apparatus for improving stitching accuracy based on digital lithography technology. The aim is to solve the problems of numerous stitching region parameters leading to difficult pattern alignment and numerous debugging attempts, as well as to improve the instability of stitching parameters and the difficulty in guaranteeing stitching accuracy due to changes in lens position. Technical solution

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A method for improving stitching accuracy based on digital lithography technology, applied to a digital lithography device containing multiple imaging modules, includes the following steps: Calibration steps: According to the DMD tilt direction of each imaging module, calibrate the imaging range and overlapping area of ​​each imaging module, and determine the starting position of the pattern cutting of each lens exposure based on the imaging range and overlapping area. Allocation steps: Based on the imaging range and overlapping area, calculate the strip width allocated to each lens, wherein the strip width is determined based on the difference in position coordinates of adjacent lenses and the overlap margin; Processing steps: According to the starting position of the graphic cutting and the width of the strip, the original graphic data is cut to obtain the cut graphic data. The graphic data of the boundary areas on both sides of the cut graphic data is deleted and replaced with blank graphic data to generate the exposure graphic data corresponding to each lens. Exposure step: The exposure graphic data is converted into a control signal and transmitted to the corresponding imaging module to control its DMD to perform exposure.

[0009] Accordingly, the present invention also provides a splicing accuracy improvement device based on digital lithography technology, comprising: Multiple imaging modules, each including a DMD; A controller is connected to each of the imaging modules. The controller includes a processor and a memory. The memory stores a computer program. When the processor executes the computer program, it implements the steps of the above method.

[0010] Compared with existing technologies, this invention has the following advantages: 1. Significantly reduces the difficulty and number of debugging steps. This invention transforms the adjustment of stitching parameters into dynamic calculation and data processing based on lens position relationships, eliminating the need for repeated combination and debugging of multiple template parameters. Operators only need to perform lens position relationship calibration once, and the system can automatically complete graphic cutting, boundary processing, and signal generation, greatly simplifying the debugging process and improving production efficiency.

[0011] 2. Achieving adaptive stitching and improving system stability. This invention acquires the actual positional relationship of each lens in real time through a calibration process, and dynamically calculates the starting position and strip width of the graphic cutting based on this positional relationship. When the positional relationship of the lenses changes due to external factors, the system can automatically adjust the cutting parameters and processing method simply by recalibrating, without manual intervention. This "dynamic closed-loop" control method completely overcomes the shortcomings of existing "static templates" that cannot adapt to positional changes, significantly improving the robustness of the system and the stability of stitching accuracy.

[0012] 3. Fundamentally eliminates the effects of overlapping exposure and reduces stitching marks. In the processing steps of this invention, the graphic data in the boundary areas on both sides of the cut intermediate graphic data is deleted and replaced with blank graphic data. The technical principle is that when the physical exposure areas of adjacent lenses overlap, data processing ensures that the graphic data corresponding to the overlapping areas of the two lenses are all blank, thus preventing any DMD from exposing the overlapping area during physical exposure. This fundamentally avoids the dose superposition problem caused by two DMDs simultaneously exposing the same area, effectively reducing stitching marks and improving the quality of the exposed pattern.

[0013] 4. Ensure the integrity of the image stitching. This invention ensures precise stitching between adjacent lens exposure patterns even when the lens position changes slightly by setting the strip width to include an overlap allowance (i.e., the difference in position between adjacent lenses plus at least twice the width of the triangular exposure area) and by accurately processing the boundary data. Specifically, when the lens position changes, the recalibrated x... n and x n+1 It will change accordingly, thus dynamically adjusting L. n The boundary processing area ensures that there is neither overlap nor breakage at the junction, thus guaranteeing the integrity and consistency of the final exposure pattern. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0015] Figure 1 This is a schematic diagram showing the positional relationship of multiple imaging module lenses and the starting position of graphic cutting according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the principle of strip width calculation according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a graphic data processing flow according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the final exposure pattern area of ​​each lens according to an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating adaptive adjustment when the lens position relationship changes according to an embodiment of the present invention; Figure 6 This is a flowchart of a method for improving splicing accuracy according to an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0017] Example 1: A method for improving splicing accuracy based on digital lithography technology This embodiment provides a method for improving stitching accuracy based on digital lithography technology. This method can be applied to digital lithography equipment containing multiple imaging modules, each including a DMD and its corresponding projection optical system. The method achieves high-precision and high-stability multi-channel DMD exposure pattern stitching by dynamically calibrating the lens position relationships and adaptively cutting and processing the graphic data based on the calibration results.

[0018] For ease of illustration, this embodiment is described using k=2 (i.e., the overlap margin is 2Δ) as an example, but the present invention is not limited to this, and k can take any integer greater than or equal to 2.

[0019] like Figure 6 As shown, the method of this embodiment includes the following steps: calibration step S101, allocation step S102, processing step S103, and exposure step S104. The following is a detailed description in conjunction with the attached diagram. Figures 1-5 Each step is explained in detail.

[0020] Step S101: Calibrate the lens position relationship of each imaging module, and determine the starting position of the pattern cutting of each lens exposure according to the lens position relationship.

[0021] like Figure 1As shown in the figure, the imaging range, overlapping area, and coordinates of the starting position of pattern cutting for each lens under the tilted arrangement of DMDs are illustrated. This step first requires calibrating the positional relationship of multiple imaging modules in the digital lithography equipment. Based on the tilt direction of the DMDs, the imaging range and overlapping area of ​​each imaging module are shown. Each imaging module forms a continuous exposure area, and there is an overlapping area between adjacent imaging modules, which is the basis for achieving multi-channel stitching. In this embodiment, the maximum effective exposure size of each DMD is assumed to be: length of M pixels, width of N pixels, physical size of a single micromirror of amm, tilt angle of the DMD relative to the scanning direction of θ, and magnification of the projection lens of β. Based on the above parameters, the pattern coverage width D of each lens (i.e., each imaging module) under the tilted scanning method can be calculated using the following formula: D = M × a × β × cosθ + N × a × β × sinθ (Formula 1) The width D represents the physical width that each DMD can cover in the exposure direction. It is the image width corresponding to the data received or transmitted by each DMD and is also the basis for subsequent graphic data allocation.

[0022] By calibrating the lens position relationships, the physical coordinates of each DMD center point can be obtained. In this embodiment, the calibration can be achieved using position measurement methods known in the art, such as obtaining the physical coordinates of each DMD center point through mechanical positioning, optical measurement, or image acquisition.

[0023] like Figure 1 As shown, with the top left corner as the origin, the camera position coordinates of the 0th, 1st, ..., nth shots are (x0, y0), (x1, y1), ..., (x...). n ,y n Based on these calibration results, and combined with the relative positions of the exposure areas of each lens on the DMD, the starting position for pattern cutting of each lens exposure can be determined. The purpose of determining the starting position for pattern cutting is to provide a precise starting point for subsequently cutting out the corresponding pattern areas of each lens from the original pattern data.

[0024] In a preferred embodiment, the starting position of the pattern cutting of the nth lens is at coordinate X in the horizontal direction (exposure scanning direction). n It can be calculated using the following formula: X n =E+D×i+x n (Formula 1-1) Where E is the platform exposure start position set by the device, D is the maximum effective exposure width of the DMD, i is the exposure setting number of strips (single strip i=0), and x nThis represents the horizontal coordinates of the lens position. The vertical coordinates (Y-coordinates) are perpendicular to the exposure scan direction. n The y-coordinate of the lens position obtained from calibration n There exists a fixed geometric offset (determined by the mounting position of the DMD in the optical system), which is the same for all lenses; therefore, the vertical position difference Y between adjacent lenses is... n+1 -Y n =y n+1 -y n This does not affect the splicing calculation.

[0025] It is worth noting that in practical applications, the top-left corner coordinates of each shot after projection can be transformed horizontally as a whole based on the positional relationship of all shots, or a single shot can be transformed vertically. Such transformations do not affect the relative distances between adjacent shots. Therefore, Figure 1 The coordinates of the upper left corner of the exposure area of ​​each lens are defined based on the calibration results of the lens position relationship.

[0026] In a preferred embodiment, the starting position of the graphic cutting is determined based on the calibration coordinates of the center points of each lens, that is, the coordinates of the center points of the lenses obtained from calibration are directly used and then converted by a fixed offset, which can more accurately reflect the actual spatial position of each lens.

[0027] Step S102: Calculate the strip width allocated to each lens based on the lens position relationship.

[0028] like Figure 2 As shown in the figure, the difference in position between adjacent lenses, the width of the triangular exposure area Δ, and the width of the strip L are illustrated. n The geometric relationship (taking k=2 as an example). This step, based on the lens position relationship calibrated in step S101, calculates the required graphic data strip width L for each lens. n The purpose of calculating the strip width is to determine the amount of data that needs to be cut from the original pattern for each lens, ensuring that the exposure patterns of adjacent lenses can be accurately aligned. During tilt scan exposure, due to the tilt setting of the DMD, triangular exposure areas are generated on both sides of the exposure area of ​​each lens. The width Δ of this triangular area is determined by the tilt angle of the DMD. The calculation formula is as follows: Δ = N × a × β × sinθ (Formula 2) Where N is the number of pixels on the wide side of the DMD, a is the size of a single micromirror, β is the magnification of the projection lens, and θ is the tilt angle of the DMD. The derivation of this formula is based on the geometric relationship of DMD tilt scanning: the tilt angle θ of the DMD causes its effective width projected in the scanning direction to produce a component related to sinθ.

[0029] like Figure 2 As shown in Figure 2 , according to the difference in the lens position relationship, at least one more triangular width Δ needs to be overlapped to ensure consistent dosage for exposure. To optimize the subsequent splicing accuracy and reduce exposure problems caused by overlapping triangular exposure, in this embodiment, the strip width L allocated to each lens is defined n It is the coordinate difference between adjacent lens positions in the horizontal direction plus k times the width of the triangular exposure area. The specific calculation formula is as follows: L n = x n+1 - x n + kΔ  (Formula 3) Among them, L n represents the strip width allocated to the nth lens, x n and x n+1 are the coordinates in the horizontal direction of the lens positions of the nth lens and the (n + 1)th lens respectively, Δ is the width of the triangular exposure area, and k is an integer greater than or equal to 2.

[0030] It should be noted that L n must be less than the maximum effective exposure width D of the DMD, that is, L n < D. In other words, it is required that the adjacent lens interval satisfies x n+1 - x n ≤ D - kΔ to ensure that the allocated strip width is within the exposure capacity range of a single DMD.

[0031] Here, kΔ is the "overlap margin" mentioned in the allocation step. It is a design parameter artificially set to ensure consistent dosage for exposure and optimize the splicing accuracy, rather than the width of the final physical overlapping area.

[0032] As a preferred embodiment, k is taken as 2. At this time, the overlap margin is 2Δ, and the strip width calculation formula is L n = x n+1 - x n + 2Δ. This value can effectively optimize the splicing accuracy and reduce exposure problems caused by overlapping exposure while ensuring consistent dosage for exposure. Figure 2 shows the relationship between the strip width taking L0 as an example, Δ, and the difference in adjacent lens positions when k = 2.

[0033] Step S103: Send the graphic cutting start position and strip width to the host computer for graphic processing.

[0034] This step is the core data processing link of the method of the present invention and is executed by the host computer. The host computer is the main control computer in the digital lithography equipment, including a processor and a memory. The original graphic data and processing programs are stored in the memory.

[0035] Such as Figure 3As shown, the starting position X of the graphic cutting of each lens calculated in steps S101 and S102 is... n Strip width L n And Δ is sent to the host computer. Among them, the starting position X of the graphic cutting... n The starting position of the graphic assignment for each shot, strip width L n Used to confirm the graphic with X n Cut L at the starting position n The width of the graph, Δwidth, is used as a processing parameter for the graphic data after each lens is cut.

[0036] The specific graphics processing flow is as follows: First, the host computer cuts the starting position X based on the graphics from each camera path sent by the software. n and strip width L n The original graphic data is sliced. The slicing operation is performed using X... n Starting from position L, extract n Data of a width of 1 pixel is used to obtain a width of L. n Cut graphic data (such as) Figure 3 L n (As shown in the blue vertical stripe area). This cut graphic data is located at the left starting position of the final graphic data with a width of D to be formed subsequently.

[0037] Secondly, the host computer performs boundary region processing on the cut graphic data obtained from the cutting process. The purpose of this processing is to eliminate the influence of overlapping exposures and ensure that each exposure area has exactly one DMD participating in the exposure. Specifically, the width L is... n After the graphic data is cut, the graphic data within the triangular exposure area with a width of Δ on the left and the exposure area with a width of (k-1)Δ on the right are deleted and replaced with blank graphic data. Thus, the total deletion width is kΔ, corresponding to the overlap margin kΔ set in step S102. In actual exposure, blank graphic data represents that the DMD micromirror does not flip and does not emit light.

[0038] After the above deletion and replacement processes, the original width of the cut graphic data is L. n The effective graphic data becomes a width of (L) n Valid graphic data of -kΔ), L n The area with a width of (Δ) on the left and the area with a width of ((k-1)Δ) on the right have become blank.

[0039] This embodiment takes k=2 as an example, therefore Figure 3 This illustrates the complete data transformation process from the original graphic data, cutting out the left and right Δ regions, filling the right side with blank space, to a width of D.

[0040] Finally, to ensure that the width of the data sent to the DMD is consistent with the maximum effective exposure width D of the DMD, and to facilitate hardware synchronization, the host computer will process the data with a width of (L). n The graphic data of -kΔ) is filled with blank graphic data on the right until its width equals the maximum effective exposure width D of the DMD. The filled blank graphic data also does not emit light during exposure.

[0041] After the above processing, the exposure graphic data for each lens is ready. The final exposure graphic data with a width of D has the following structure from left to right: The blank area on the left (Δ) width (composed of the blank areas where the original left (Δ) area was replaced); Middle (L) n The effective graphic region with a width of -kΔ) Right side (DL) n The blank area is a width of +(k-1)Δ) (composed of the blank area on the right that was replaced and the blank area filled on the right).

[0042] For the preferred embodiment where k=2, the above structure corresponds to Figure 3 As shown: Left side Δ blank, middle (L) n -2Δ) effective, right side (DL) n +Δ) blank.

[0043] Step S104: The processed graphic data is rasterized, and the host computer transmits the signal sequence to finally realize the DMD flip exposure.

[0044] The exposure image data processed in step S103 needs to be rasterized to convert it into a control signal sequence that the DMD can recognize. The rasterization process is performed by the host computer: the host computer converts the image data of width D into a binary control signal sequence row by row, according to the micromirror array arrangement of the DMD. Pixel areas corresponding to valid image data generate micromirror flip control signals, while pixel areas corresponding to blank image data generate micromirror non-flip control signals. The host computer transmits the corresponding signal sequence to the DMD controller of the corresponding imaging module through the data transmission interface. The DMD controller controls the flipping state of each micromirror according to the control signal sequence.

[0045] like Figure 4 As shown in the figure, the technical effect is that the exposure areas of adjacent lenses physically overlap only with blank areas, while the effective graphic areas do not overlap. After the above processing steps, the actual exposure area of ​​each imaging module involves only a single DMD. The following is a geometric analysis of this technical effect: For the nth and (n+1)th adjacent shots: In the final exposure pattern data of the nth shot, the width of the blank area on the right is (DL). n +(k-1)Δ); In the final exposure graphic data of the (n+1)th shot, the width of the blank area on the left is (Δ); Since the starting position X of the graphic cutting of the (n+1)th shot n+1 Located exactly on the right edge of the exposure area of ​​the nth lens, the blank area on the right side of the nth lens completely overlaps with the blank area on the left side of the (n+1)th lens in physical position, forming a total width of (DL). n The physical overlap region of +kΔ).

[0046] Therefore, the overlap margin kΔ (design parameter) set in the allocation step is ultimately converted into a physical total width of (DL). n The blank overlapping region (result parameter) of +kΔ) and the two form a logical closed loop through data processing steps.

[0047] According to step S102, L n =x n+1 -x n +kΔ, and D=M×a×β×cosθ+N×a×β×sinθ in step S101, it can be deduced that the physical overlapping area is completely composed of blank graphic data and does not contain any valid graphic information.

[0048] Therefore, during physical exposure, no DMD is effectively exposed in the overlapping area (all micromirrors are in a non-flipped state), which fundamentally avoids the dose superposition problem caused by two DMDs exposing the same area at the same time, and achieves the technical effect of reducing the overlapping area of ​​the stitching and reducing the impact of overlapping exposure on the stitching.

[0049] Furthermore, a significant advantage of the method of this invention lies in its adaptive capability to changes in lens position relationships. Traditional methods use binary bitmaps to control the tilting scan exposure, keeping the triangles on both sides in a constant, unflipped state. When the lens position relationship changes, especially when the change is less than one pixel, the micromirrors controlled by the binary bitmap remain in a non-flipped light-emitting state, making it impossible to accurately control the stitching connection, resulting in overlapping or broken patterns at the stitching points.

[0050] like Figure 5 As shown in the figure, the adaptive adjustment of the graphic end position is achieved through recalibration when the lens position changes. When the lens position relationship changes due to factors such as temperature changes or mechanical vibration, for example, when the position of the yellow dashed line in the figure changes to the position of the red dashed line, this indicates that the graphic end position of the previous strip assignment has changed.

[0051] In the method of the present invention, when the lens position relationship changes, it is only necessary to re-execute the calibration operation in step S101. The host computer can automatically obtain the new lens position relationship, and recalculate the starting position of the pattern cutting and the strip width based on the new position relationship, and then generate exposure pattern data adapted to the new position relationship in step S103.

[0052] like Figure 5 As shown, when the boundary position changes from the yellow dashed line to the red dashed line, this method controls the transmitted signal through graphic processing, causing the flip state of the green micromirror to dynamically adjust accordingly. This achieves adaptive compensation for changes in lens position, ensuring the integrity of the graphic stitching. This data processing-based "soft control" method offers higher flexibility and accuracy compared to the existing "hard control" method based on hard templates. This invention ensures the integrity of the graphic stitching by adaptively adjusting the stitching through lens position relationship calibration.

[0053] Example 2: Equipment for Improving Splicing Accuracy Based on Digital Lithography Technology This embodiment provides a device for improving splicing accuracy based on digital lithography technology, used to implement the method described in Embodiment 1. The device includes: Multiple imaging modules are used, each including a DMD and its corresponding projection optical system. The imaging modules are arranged according to the tilt direction of the DMD to form a continuous exposure area, with overlap between the exposure areas of adjacent imaging modules.

[0054] A controller is connected to each of the imaging modules. The controller includes a processor and a memory, the memory storing a computer program. When the processor executes the computer program, it implements the steps of the method described in Embodiment 1. Specifically, the controller is configured as follows: The positional relationship of the lenses in each imaging module is calibrated, and the starting position of the pattern cutting for exposure of each lens is determined based on the positional relationship of the lenses; Based on the lens position relationship, calculate the strip width allocated to each lens. The strip width is determined based on the difference in position coordinates of adjacent lenses and the overlap margin. The overlap margin includes at least twice the width of the triangular exposure area. Based on the starting position and strip width of the graphic cutting, the original graphic data is cut to obtain the cut graphic data. The graphic data within the area of ​​the width of a triangle exposure area on the left and right sides of the cut graphic data is deleted and replaced with blank graphic data. Then, blank graphic data is filled on the right side until its width is equal to the maximum effective exposure width of the DMD to generate the exposure graphic data corresponding to each lens. The exposure image data is converted into control signals and transmitted to the corresponding imaging module to control its DMD for exposure.

[0055] When the positional relationship of the lenses changes, the controller adaptively adjusts the exposure graphic data corresponding to each lens by re-executing calibration, allocation and processing steps to ensure the integrity of the graphic stitching.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for improving splicing accuracy based on digital lithography technology, characterized in that, Includes the following steps: Calibration steps: According to the DMD tilt direction of each imaging module, calibrate the imaging range and overlapping area of ​​each imaging module, and determine the starting position of the pattern cutting of each lens exposure based on the imaging range and overlapping area. Allocation steps: Based on the imaging range and overlapping area, calculate the strip width allocated to each lens, wherein the strip width is determined based on the difference in position coordinates of adjacent lenses and the overlap margin; Processing steps: According to the starting position of the graphic cutting and the width of the strip, the original graphic data is cut to obtain the cut graphic data. The graphic data of the boundary areas on both sides of the cut graphic data is deleted and replaced with blank graphic data to generate the exposure graphic data corresponding to each lens. Exposure step: The exposure graphic data is converted into a control signal and transmitted to the corresponding imaging module to control its DMD to perform exposure.

2. The method according to claim 1, characterized in that, The overlap margin includes at least twice the width of the triangular exposure area, which is generated by the tilt scanning method of the DMD. The width Δ of the triangular exposure area is calculated by the following formula: Δ=N×a×β×sinθ, where N is the number of pixels on the wide side of the DMD, a is the size of a single micromirror, β is the magnification of the projection lens, and θ is the tilt angle of the DMD.

3. The method according to claim 2, characterized in that, The strip width is calculated using the following formula: L n =x n+1 -x n +kΔ, where L n The strip width assigned to the nth shot, x n and x n+1 Δ represents the horizontal coordinates of the lens positions of the nth and (n+1)th lenses, respectively, Δ represents the width of the triangular exposure area, and k is an integer greater than or equal to 2.

4. The method according to claim 3, characterized in that, The width of the strip satisfies L n < D, where D is the maximum effective exposure width of the DMD, and the difference between the adjacent lens position coordinates satisfies x n+1 -x n ≤ D - kΔ.

5. The method according to claim 2, characterized in that, The processing step of deleting and replacing the graphic data in the boundary areas on both sides of the cut graphic data with blank graphic data specifically includes: With a width of L n The graphic data within the triangular exposure area with a width of Δ on the left and the exposure area with a width of (k-1)Δ on the right after the cut graphic data are deleted and replaced with blank graphic data.

6. The method according to claim 5, characterized in that, The processing steps further include: Width is L n After the regions on the left and right sides of the cut graphic data are replaced with blank graphic data, the resulting width is (L). n The graphic data of -kΔ) is filled with blank graphic data on the right until its width is equal to the maximum effective exposure width D of the DMD.

7. The method according to claim 6, characterized in that, The maximum effective exposure width D of the DMD is calculated by the following formula: D=M×a×β×cosθ+N×a×β×sinθ, where M is the number of pixels on the long side of the DMD, N is the number of pixels on the wide side of the DMD, a is the size of a single micromirror, β is the magnification of the projection lens, and θ is the tilt angle of the DMD.

8. The method according to claim 1, characterized in that, In the calibration step, the starting position of the graphic cutting is determined based on the calibration coordinates of the center points of each lens.

9. The method according to claim 1, characterized in that, When the positional relationship of the lenses changes, the calibration, allocation, and processing steps are re-executed to adaptively adjust the exposure graphic data corresponding to each lens, so as to ensure the integrity of the graphic stitching.

10. A splicing accuracy improvement device based on digital lithography technology, characterized in that, include: Multiple imaging modules, each including a DMD; A controller, connected to each of the imaging modules, the controller including a processor and a memory, the memory storing a computer program, the processor executing the computer program to implement the steps of the method according to any one of claims 1 to 9.

Images