A single light source horizontal calibration method for DLP 3D printing

Abstract

A single light source horizontal calibration method for DLP 3D printing relates to the technical field of intelligent control, image acquisition and image processing. The method aims to realize high-precision printing of surface exposure single light source, and comprises the following steps: image acquisition: fix the camera at the position of the printing resin tank above the projection. The gray scale images under different deflection angles are collected in turn; image analysis: for the collected images, the contour of the image is extracted, and the barycentric coordinates are calculated. The change of the pixel spacing and the slope under different deflection angles is calculated by using the barycentric coordinates. The curve graph is drawn to obtain the relationship expression. Projection correction: compare the calculated average value with the standard value. If they are inconsistent, it means that deflection has occurred. Then the corresponding deflection formula is calculated to correct the projector. The present application can improve the forming quality of the product, and can improve the printing success rate and material utilization rate.

Description

Technical Field

[0001] The present invention relates to intelligent control, image acquisition and image processing technologies, and specifically to the acquisition and analysis of images at different positions on an exposure surface. By combining the light source distribution characteristics of a digital micromirror device (DMD) with the pixel distribution of a single light source, the spatial position of the pixels in the captured image is analyzed and corrected, thereby achieving horizontal alignment of a single DLP light source. Background Art

[0002] 3D printing is an additive manufacturing technology. Based on a three-dimensional digital model, it creates objects from scratch by adding layers of printed material under the control of computer software. Currently, 3D-printed products are widely used in aerospace, automotive, medical, and other fields. Additive manufacturing technology has garnered widespread attention from governments, research institutions, businesses, and the media, with countries increasingly identifying it as a new growth point for future industrial development.

[0003] There are many types of 3D printing technologies, depending on the printing process and materials. (1) Fused Deposition Modeling (FDM) technology; (2) Laminated Object Manufacturing (LOM) technology; (3) Selective Laser Sintering (SLS) technology; (4) Selective Laser Melting (SLM) technology; (5) Stereolithography 3D printing technology. Among them, Stereolithography 3D printing technology is the earliest developed and most mature. Compared with other printing methods, it has obvious advantages in both molding speed and molding quality.

[0004] There are many types of light-curing 3D printing technologies, including Stereo Lithography Appearance (SLA), PolyJet TM ) and Deep Light Projection (DLP). Comparisons show that the hardware structure of DLP is simpler and the cost is lower. It offers a significant cost-performance advantage over similar light-curing technologies. Therefore, DLP has become the primary research topic for 3D continuous light-curing printing.

[0005] 3D printing technology projects the shape of the printed slice onto an exposure surface as an image. The accuracy of the projected image impacts the quality of the printed object. High-precision DLP light sources require that the exposure surface be perpendicular to the normal of the light source and that the light source be horizontal. During the printing process, if a single projection is tilted, the projector will tilt, causing the image on the exposure surface to become out of focus, which can affect the printing quality of a single projection.

[0006] In existing methods, calibration is performed by manually observing the liquid level, but the calibration accuracy cannot be maintained. This paper proposes a vision-based scheme to estimate the horizontality of the projector and achieve final calibration by analyzing the projected image. Summary of the Invention

[0007] The embodiments of the present invention provide a single light source level calibration method for DLP 3D printing, which is used to improve the accuracy of printed objects.

[0008] To solve the above technical problems, the embodiments of the present invention adopt the following technical solutions:

[0009] A single light source level calibration method for DLP 3D printing includes the following steps:

[0010] Step 100: manually place the projector and focus the light source so that its focal plane is exactly on the upper surface of the resin tank. Place an industrial camera above the resin tank and adjust the camera position so that its focal plane coincides with the focal plane of the projection light source.

[0011] Step 200: Control the projector to deflect to three different angles of X, Y, and Z. The projector projects a test image. The camera receives images at different deflection angles and stores the images at equal intervals on the PC.

[0012] Step 300: Analyze the captured color image, analyze the pixel changes under different deflection angles, fit the curve, and obtain the relationship expression.

[0013] Step 400: Analyze the pixel parameter values ​​of different images to estimate the direction and angle of projector deflection. Determine whether deflection has occurred. If so, move the projector position for horizontal correction to keep the light source in a horizontal position.

[0014] The step 100 of determining the position of the light source and the camera includes:

[0015] The printing system consists of three units: the projection unit, the resin tank unit, and the camera unit. The projection unit includes a projection stand and a projector. The stand is adjustable, allowing for translation, raising, and lowering by turning knobs. Based on the build platform settings, determine the distance from the projection light source to the resin tank (340mm in this experiment). Focus the projector to ensure the plane of the resin tank is in focus.

[0016] The camera unit contains a camera and a motion axis capable of moving along the X, Y, and Z axes. The camera uses CMOS imaging and lacks a focus function, so we need to move the lens directly to the focal plane where the resin tank is located. Due to the small size of the built-in chip in the CMOS camera, it can only capture an image of an area approximately 5mm x 5mm. Because the projector's light intensity is too strong, the camera is overexposed, so we need to reduce the light intensity. We can use the following methods: project a low-grayscale slice image or reduce the projector brightness when projecting the image, and then add a filter to the camera. Fine-tune the camera until you get the clearest image possible.

[0017] Controlling the projector to deflect at three different angles, X, Y, and Z, projecting a test image, using a camera to receive images at different deflection angles, and storing the images at equal intervals on a PC, step 200 includes:

[0018] After controlling the projector's movement, adjust the focus until the light source is well-focused and clearly projected on the focal plane. Project a test pattern to capture a single-light source image. Select an area within this image to define as the camera capture area. Set the interval time to capture a certain number of images at equal intervals.

[0019] The captured color image is analyzed. Step 300 of analyzing pixel changes at different deflection angles and obtaining a relational expression of the changes includes:

[0020] The acquired color image is binarized to remove noise and preserve the original image. For each contour in the image, its centroid coordinates are calculated. Using these centroid coordinates, the change in pixel spacing and slope at different deflection angles is calculated. The average value for the entire image is calculated. Based on the obtained data, a graph is plotted showing the change in pixel size at different deflection angles, and a formula is used to express the relationship between pixel size change and angle.

[0021] By analyzing the parameter values ​​of the pixels of different images, the direction and angle of the projector deflection can be estimated. Determining whether deflection has occurred, and if so, moving the projector position to perform leveling correction so that the light source is in a horizontal position, step 400 includes:

[0022] Calculate the average pixel spacing and slope of the captured image and compare the calculated averages with the reference values. Any discrepancies indicate deflection. Substituting these values ​​into the fitted curve, the projector's deflection direction and angle can be estimated. Calibrate the projector based on these values. After each fine-tuning step, recapture the image, perform image analysis, and perform projection calibration. Repeat this process until the calculated values ​​match the reference values. This completes single-light source level calibration.

[0023] The single light source level calibration method based on surface exposure according to the embodiment of the present invention has the following advantages:

[0024] 1) Improved molding quality. By adjusting the level of the single light source so that the normal of the projected light source is perpendicular to the printing platform, the entire layer of the projected slice image is in focus, improving the printing quality of the object.

[0025] 2) Improve printing success rates and material utilization. Leveling before printing ensures that the resin is completely cured on the print plate during UV exposure. This prevents the resin from being partially cured on the plate and partially in the resin tank due to inconsistent focus. This significantly improves product success rates and prevents contamination of the remaining resin by materials in the tank.

[0026] 3) Universal applicability to multiple light sources: The method proposed in this invention can be used for image acquisition, analysis, and correction in all surface-exposure 3D printing methods, regardless of the type of light source used, to achieve precise alignment of a single light source, thereby improving print quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic flow chart of the present invention.

[0028] Figure 2 It is a detailed picture of the test image.

[0029] Figure 3 It is to establish the projector coordinate system and the schematic diagram of different deflection angles

[0030] Figure 4 It is the image analysis process.

[0031] Figure 5 It is the relationship curve between the pixel size and angle change in the XYZ direction. DETAILED DESCRIPTION

[0032] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other in any manner.

[0033] Surface exposure 3D printing using DLP as the light source is called DLP 3D printing technology. This printing method can be divided into upper exposure printing and lower exposure printing according to the position of the light source. In this invention, lower exposure is used as an example to illustrate. DLP technology was first developed by Texas Instruments (TI). It mainly projects an image through a projector to form a two-dimensional pattern in the resin tank solution. After irradiation, a polymerization reaction occurs to solidify the layer. The mobile platform controls the layer-by-layer solidification to eventually form a three-dimensional entity. Figure 1 The figure shows an actual printing environment. The hardware system consists of two main components: a camera unit and a projection platform unit. The camera moves to the projector's focal plane, maintaining its position during image capture. The projection platform unit includes the projector and a three-axis motion system. The projector serves as the light source, and the three-axis system allows for positional correction of the projector.

[0034] To improve the precision and quality of printed products, existing research has primarily focused on processing slice images. Research on the impact of projector position changes on pixels, based on projector principles, is largely absent. The method in this embodiment maintains the accuracy of the projected image of the printed slice by correcting the projection position, ensuring that the normal of the light source is perfectly perpendicular to the printing liquid surface in the resin tank.

[0035] The core of DLP projection technology is the digital micromirror device (DMD) chip, which features millions of closely packed micromirrors. Each micromirror represents a pixel. Ignoring the gaps between the micromirrors, we can think of the projected image as a matrix of pixels. The camera uses CMOS imaging and lacks a focus function. Therefore, we need to directly move the lens to the focal plane where the resin tank is located to control the camera to capture the image.

[0036] The embodiment of the present invention proposes a single light source level calibration method for DLP 3D printing, the flow chart is as follows: Figure 1 As shown. Includes:

[0037] Step 100: Fix the position of the printing light source and focus the light source so that its focal plane is exactly on the upper surface of the resin tank. Place an industrial camera above the resin tank and adjust the camera position so that its focal plane coincides with the focal plane of the projection light source.

[0038] Optionally, step 100 includes the following sub-steps:

[0039] Sub-step 110, manually place the projector and focus the projector. Place the projection light source under the printing platform, between 300-400mm away from the resin tank. Turn on the projector, set the projector brightness to 20, and present an image on the resin tank with a projection area that meets the printing requirements; when the projected image does not flicker, it means that the light of the projector remains stable. The first layer in the actual printing process, that is, the bottom of the resin tank, is used as the focusing plane. Place a piece of white paper on the plane and judge whether it is focused by the clarity of the projected pattern on the white paper. If it is not focused, turn the focus ring to a clear position to complete the initial setting. Whether it is focused is determined by observing whether the edge of the projected pattern is clear.

[0040] Sub-step 120: Move the camera so that its focal plane coincides with the focal plane of the projection light source. The camera is fixed to a three-axis mechanical system directly above the projection light source, and different functions are achieved by moving along different axes. The three-axis mechanical system described in sub-step 120 includes three axes: X, Y, and Z, which control the left-right and forward-backward movement of the camera on the XY plane to reach the specified image capture area; it also enables movement along the Z axis to fine-tune the camera up and down to align its focal plane with the focal plane of the projection light source.

[0041] Step 200: Control the projector to deflect to three different angles of X, Y, and Z. The projector projects a test image. The camera receives images at different deflection angles and stores the images at equal intervals on the PC.

[0042] Optionally, step 200 includes the following sub-steps:

[0043] Sub-step 210: Determine the test image. In order to clearly observe the distribution of pixels, the experiment uses a checkerboard pattern. The standard resolution of the projector is 1024x768, and the test image is also set to the same size. Image details are as follows: Figure 2 As shown, each grid represents a pixel, 1 represents a grayscale value R=255, G=B=0, and 0 represents a grayscale value R=G=B=0.

[0044] Sub-step 220: Control the projector to deflect at different angles. After the projector is placed on the projection stand, it may deflect at various angles due to human error. In order to study the impact of angle deflection in different directions on pixel points, we establish a coordinate system with the projector as the origin and obtain its deflection images in the X, Y, and Z directions, as shown in the following example: Figure 3 Deflect the projector at a frequency of 1°. If the focus change causes the image to be unclear, turn the focus ring to adjust the focus of the projector.

[0045] Sub-step 230, capture a single light source image. Project the test image into the resin tank through a projector, and wait for the projected image to be clearly displayed on the focusing surface. Select the center position on the focusing interface as the image capture area, mark the focusing surface, remove the resin tank, move the camera to the image capture area, and then fine-tune it up and down to make the image clear. Set the interval time, and continue deflecting after acquiring images at equal intervals. Store images between deflections of ±10°. Repeat the above steps to record the deflection images in three directions.

[0046] Step 300: Analyze the captured color image, analyzing pixel changes at different deflection angles to obtain a relational expression of the changes.

[0047] Optionally, step 300 includes the following sub-steps:

[0048] Sub-step 310: Perform image preprocessing. The captured color image is thresholded and binarized. Gaussian filtering and image opening are performed on the image to remove noise and preserve the original image. For each contour in the image, calculate its centroid coordinates and store them in a point array.

[0049] Sub-step 320: Perform horizontality analysis on the image deflected in the XY direction. The distance between the centroid coordinates of the two contours in each row and column of the image is calculated to represent the distance between the two bright pixels. Taking any contour point as the center, calculate the distance between the centroid coordinates of each row / column in the image and record it as the average distance of the row / column. Figure 4 As shown. Traverse all rows and columns within the error range on the image. If there are M*N (rows*columns), we can obtain M average row spacings and N average column spacings. Then, average these values ​​to obtain the average spacing value M1 between all rows and the average spacing value N1 between all columns. Repeat the above steps to obtain the average pixel row spacing value (R-avg) after deflection at different deflection angles in the X direction and the average pixel column spacing value (C-avg) after deflection at different deflection angles in the Y direction.

[0050] To avoid the effects of different focal lengths during printing, the pixel size at the origin is set to standard values, including the standard row spacing (SV-R) and the standard column spacing (SV-C). We analyze the relationship between the difference and angle between the average row / column spacing and the standard row / column spacing of the image at different deflection angles, and record the average pixel spacing and the standard row / column spacing as: R-avg-D, C-avg-D respectively. We can obtain a set of discrete points. Curve fitting is performed to obtain the pixel relationship curve, and the X-axis curve is as follows: Figure 5 As shown in (a), the Y-axis curve is Figure 5 (b) shown.

[0051] Sub-step 330: Analyze the horizontality of the image deflected in the Z direction. The slope between the centroid coordinates of two adjacent contours of the image is used to represent the slope of two bright pixels. With any contour point as the center, crop an image with a size of 110*220 and calculate the slope of two pixels in the image. Figure 4 As shown. Traverse all the contour points within the error on the image. If there are N contour points, we can get the average slope of N cropped 110*220 images. Then we calculate the average value of the whole image and finally get a value representing the slope of the two bright pixels at the deflection angle. We can get a set of discrete points. Perform curve fitting to get the pixel relationship curve. The Z-axis curve is as follows Figure 5 (c) shown.

[0052] Step 400: Analyze the pixel parameter values ​​of different images to estimate the direction and angle of projector deflection. Determine whether deflection has occurred. If so, move the projector position for horizontal correction to keep the light source in a horizontal position.

[0053] Optionally, step 400 includes the following sub-steps:

[0054] Sub-step 410: Before the actual printing process begins, after the projector is manually placed, the average pixel row spacing, column spacing, and average slope between pixels of the contour in this state should be calculated.

[0055] Sub-step 420: Calculate the difference between the calculated average pixel spacing and the standard value. If the difference is not 0, it means that the XY axis is not in the horizontal position. Substitute Figure 5 Using curves (a) or (b), you can determine the deflection direction and angle along that axis and calibrate the projector accordingly. After calibration, recapture the image, calculate the average pixel pitch, and compare it until it matches the standard. Calibration is complete.

[0056] Sub-step 430: In the horizontal state, the standard inter-pixel slope = 0. Compare the obtained average inter-pixel slope with the standard slope. If they are inconsistent, it means that the Z direction has deflected. Substitute the average slope value into Figure 5 Using the middle curve (c), you can determine the deflection direction and angle along that axis and calibrate the projector accordingly. After calibration, recapture the image, calculate the slope, and compare it until it matches the standard value. Calibration is complete.

Claims

1. A single light source level calibration method for DLP 3D printing, characterized in that: The following steps are involved:

100. Fix the position of the printing light source and focus the light source so that its focal plane is exactly on the upper surface of the resin tank. Place an industrial camera above the resin tank and adjust the camera position so that its focal plane coincides with the focal plane of the projection light source; 200. Control the projector to deflect to three different angles of X, Y, and Z. The projector projects a test image. The camera receives images at different deflection angles and stores the images at equal intervals on the PC.

300. Analyze the captured color image; analyze pixel changes at different deflection angles to obtain a relational expression of the changes; 400. For different images, analyze the parameter values ​​of their pixels and estimate the direction and angle of projector deflection; Determine whether deflection occurs. If so, move the projector to perform level correction so that the light source is in a horizontal position. Step 100 includes: Manually position and focus the projector. Place the projection light source below the build platform, 300-400mm from the resin tank. Use CMOS imaging to move the lens directly to the focal plane of the resin tank. Step 200 includes: Control the projector to deflect to different angles. After placing the projector on the projection stand, in order to study the impact of different angles in different directions on the pixels, a coordinate system is established with the projector as the origin, and its deflection images in the X, Y, and Z directions are obtained respectively. The projector is moved 1° at a time. If the focal length changes and the image is not clear, turn the focus ring to adjust the projector. Project the test image into the resin tank using a projector until the projected image is clearly displayed on the focusing surface. Select the center position on the focusing interface as the image capture area, mark the focusing surface, remove the resin tank, move the camera to the image capture area, and then fine-tune the camera up and down to make the image clear. Set the interval time, capture images at equal intervals, and then continue deflection. Store images within the deflection range of ±10°. Repeat the above steps to record deflection images in three directions. Step 300 includes: Perform image preprocessing; the collected color image is binarized; the image is Gaussian filtered and image opening is performed to remove the influence of noise and retain the original image; for each contour on the image, its center of gravity coordinates are calculated and stored in a point array; Perform horizontality analysis on an image deflected in the XY direction; calculate the distance between the centroid coordinates of two contours in each row and column of the image to represent the distance between two bright pixels; with any contour point as the center, calculate the distance between the centroid coordinates of each row / column in the image, and record it as the average spacing of the row / column; traverse all rows / columns within the error range on the image; if there are M rows * N columns, obtain M average row spacings and N average column spacings, and then calculate the average value of the overall spacing to finally obtain the average spacing value M1 between all rows and the average spacing value N1 between all columns; repeat the above steps to obtain the pixel spacing value after deflection at different deflection angles in the XY direction; The pixel size of the origin position is set to a standard value, including a standard row spacing and a standard column spacing; the relationship between the difference and the angle between the average row / column spacing and the standard row / column spacing of the image at different deflection angles is analyzed to obtain a fitting curve; Perform horizontality analysis on the image deflected in the Z direction; calculate the slope between the centroid coordinates of two adjacent contours in the image to represent the slope of the two bright pixels; crop an image with a size of 110*220 with any contour point as the center, and calculate the slope of the two pixels in the image; traverse all contour points within the error range on the image. If there are N contour points, the average slope of the N cropped 110*220 images can be obtained, and then the average value is calculated for the entire image to finally obtain a value representing the slope of the two bright pixels at the deflection angle, and obtain a fitting curve with the angle; Step 400 includes: Before the actual printing process begins, after placing the projector, the average pixel spacing of the outline in this state and the average slope between pixels should be calculated; Based on the calculated average pixel spacing, find the difference between it and the standard value. If the difference is not 0, it means that the XY axis direction is not in the horizontal position. Substitute it into the fitting to obtain the deflection angle in that direction, and calibrate the projector according to the angle. After calibration, capture the image again, calculate the average pixel spacing, and compare it until it is consistent with the standard value. The calibration is complete. In the horizontal state, the standard inter-pixel slope = 0; compare the obtained average inter-pixel slope with the standard slope. If they are inconsistent, it means that deflection has occurred in the Z direction; substitute the average slope value into the fitting curve to obtain the deflection angle in that direction, and calibrate the projector according to the angle; after calibration, capture the image again, calculate the slope value, and compare it until it is consistent with the standard value, and the calibration is complete.

Images