Machine Vision-Based Method and System for Baseboard Height Calibration in 3D Printing Equipment

Abstract

This invention relates to the field of additive manufacturing, specifically to a method and system for calibrating the substrate height of a 3D printing device based on machine vision. The method includes: S1, initializing the system and parameters; S2, acquiring a substrate image; S3, determining the substrate height; S4, coarsely adjusting the substrate height; S5, acquiring a powder-spreading image; S6, calculating indicators based on the powder-spreading image; S7, determining whether to perform substrate height adjustment; S8, determining whether to perform fine-tuning of the substrate height; S9, fine-tuning the substrate height; and S10, loop control and termination conditions. The system includes a metal 3D printing device, an industrial camera, a vision reference calibration block, and a control system. This invention monitors the substrate height and powder-spreading state using machine vision, and achieves precise control of the first layer of powder-spreading state using a two-stage precision adjustment height calibration mechanism.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing, and more specifically to a method and system for calibrating the height of a substrate for 3D printing equipment based on machine vision. Background Technology

[0002] In powder bed metal 3D printing, the substrate height and the initial powder layer's placement directly affect the dimensional accuracy of the formed part, the quality of interlayer bonding, and the printing success rate. In existing technologies, the substrate typically needs to be leveled and its height calibrated before printing to ensure that the substrate's upper surface is aligned with the machined surface of the equipment.

[0003] In existing technologies, there are various technical solutions for leveling substrates, but substrate height calibration still mainly relies on manual operation: the operator repeatedly raises and lowers the forming piston or substrate platform, and judges whether the substrate has reached the processing height using a ruler, feeler gauge, or visual observation. Then, the first layer of powder is spread, and the flatness and continuity of the powder spread are observed manually to confirm whether the first layer processing conditions are met. This method currently has the following shortcomings: 1. The adjustment process relies on human experience, and different operators have inconsistent judgment standards, resulting in poor repeatability and consistency. 2. The operation steps are cumbersome, requiring multiple platform raising and lowering, manual measurement and confirmation, long debugging time, and low equipment utilization. 3. The quality of the first layer of powder spreading cannot be quantitatively evaluated; it can usually only be judged visually, making it difficult to accurately identify local height deviations or powder spreading abnormalities. 4. In applications with large powder bed dimensions or high levels of equipment automation, the above manual method is no longer sufficient to meet the production requirements of high stability and high consistency.

[0004] Therefore, there is an urgent need for a method and system that can automatically calibrate the height of the substrate of a metal 3D printing equipment to improve adjustment efficiency and the reliability of the first layer forming. Summary of the Invention

[0005] To address the shortcomings of the existing technology, this invention provides a method and system for calibrating the height of a 3D printing equipment substrate based on machine vision. This method achieves automatic calibration of the height of a metal 3D printing equipment substrate by objectively judging the powder spreading state of the first layer using machine vision.

[0006] Specifically, on the one hand, the present invention provides a method for calibrating the height of a substrate for a 3D printing device based on machine vision, which includes the following steps:

[0007] S1, initialize the system and parameters;

[0008] S2, Obtain substrate image: The substrate image includes the substrate and a visual reference calibration block;

[0009] S3, Determine substrate height: Obtain substrate height based on visual reference calibration block. ;

[0010] S4, Coarsely adjust substrate height: Adjust the substrate height Height relative to the preset machining reference surface The height deviation was obtained by comparison. As the substrate height adjustment amount, the substrate height is adjusted to the preset processing reference surface height;

[0011] S5, Obtain powder spreading image: Obtain the powder spreading image after performing one powder spreading operation;

[0012] S6, Calculate indicators based on the powder spreading image: Indicators include powder coverage indicators. Powder uniformity index Comprehensive powder spreading quality evaluation indicators and the localized accumulation index pile;

[0013] S7, Determine whether to perform substrate height adjustment: Based on comprehensive powder coating quality evaluation indicators. Determine whether the current powder spreading status meets the first layer processing conditions. If it does, proceed to S10; otherwise, execute S8.

[0014] S8, Determine whether to perform fine-tuning of substrate height: When If the current powder spreading state is determined to be in the critical range of the first layer processing, execute S9; otherwise, execute S10.

[0015] S9, fine-tuning the substrate height;

[0016] S9 specifically includes S91-S93:

[0017] S91, Step Length Calculation: Based on the comprehensive powder spreading quality evaluation index With the first layer processing judgment threshold The deviation between them is used to calculate the substrate height adjustment step size. Specifically:

[0018] ;

[0019] in, The number of times the substrate height is finely adjusted. The nominal layer thickness for a single application of powder. The threshold for determining whether the powder spreading state is in the critical range of the first layer processing; This is the threshold for determining the first layer of processing.

[0020] S92, Determine the adjustment direction: Determine the adjustment direction of the substrate height based on the comprehensive powder coating quality evaluation index;

[0021] S93, Adjust substrate height: Adjust the step size according to the substrate height. Adjust the orientation and height of the substrate, and record the number of adjustments;

[0022] S10, Loop control and termination condition;

[0023] Determine whether the preset conditions for ending substrate height adjustment are met. If they are met, end the adjustment. If not, determine whether the substrate height has been fine-tuned. If it has, execute S5. If it has not, execute S2.

[0024] Preferably, the initialization parameters in S1 include at least:

[0025] Initialize the automatic calibration parameters for substrate height: , , ;in The number of times the substrate height is fine-tuned; This indicates the number of times the comprehensive powder coating quality evaluation index continuously meets the processing conditions of the first layer; This indicates whether the substrate height has been fine-tuned; 0 indicates that it has not been fine-tuned yet.

[0026] Preferably, S2 is as follows:

[0027] An industrial camera positioned above the side of the powder spreading vehicle is used to capture an initial substrate image containing the substrate and a visual reference calibration block. The industrial camera is set at an angle so that the initial substrate image it captures includes the substrate and the visual reference calibration block. Distortion correction and perspective transformation processing are performed on the captured initial substrate image to obtain the substrate image.

[0028] Preferably, S3 is as follows:

[0029] The distance between the upper surface of the visual reference calibration block and the working surface of the scraper is a known fixed value. Based on the proportional relationship between the size of the visual reference calibration block and the substrate, and based on the correspondence between the pixel position of the visual reference calibration block in the substrate image and its actual height, the substrate height is calculated according to the mapping relationship. .

[0030] Preferably, S5 is as follows:

[0031] After performing a powder spreading operation, control the industrial camera to capture the initial powder spreading image, which includes the substrate and the visual reference calibration block;

[0032] Based on the position of the visual reference calibration block in the initial powder spreading image, the powder spreading area is located, and the powder spreading area is determined by a preset area mapping relationship or a method based on image boundary recognition. The powder spreading area is then cropped from the initial powder spreading image to obtain a cropped powder spreading image.

[0033] The cropped powder-spreading image is preprocessed, including distortion correction, perspective transformation, grayscale conversion, filtering and denoising, and brightness normalization to obtain the powder-spreading image.

[0034] Preferably, S6 is as follows:

[0035] S61, Extract feature parameters of the powder spreading image;

[0036] Based on the powder-spreading image, gray-level distribution features are extracted. The gray-level distribution features include at least the pixel gray-level value, the gray-level threshold range corresponding to the powder area, the average value and standard deviation of the pixel gray-level value.

[0037] S62, obtain the powder coverage index;

[0038] Powder coverage index Calculated based on the pixel proportion of the powder region in the powder-spreading image:

[0039] ;

[0040] in, This refers to the powder coverage index. This represents the total number of pixels in the powder-coated image. For pixels grayscale value; This represents the grayscale threshold range corresponding to the powder region. For indicator functions, when belong The value is 1 if it is true, and 0 otherwise.

[0041] S63, obtain the powder spreading uniformity index;

[0042] Powder uniformity index The following calculations were made based on the statistical characteristics of the grayscale distribution in the powder-coated area:

[0043] ;

[0044] in, For the uniformity of powder spreading, This is the average grayscale value of all pixels in the powder-coated image. This represents the standard deviation of the grayscale values ​​of all pixels in the powder-coated image.

[0045] S64, Obtain comprehensive powder spreading quality evaluation indicators ;

[0046] ;

[0047] in, , These are the weighting coefficients, and This is used to balance the requirements of coverage integrity and uniformity;

[0048] S65, obtain local accumulation index;

[0049] The local accumulation index is represented by "pile". Initially, pile=0 indicates that there is no local accumulation. Local accumulation detection is performed using a detection window. In the powder-spreading image, the detection window traverses the pixels in the powder-spreading image with a step size of one pixel at a time. The average gray value of all pixels in the detection window is compared with the preset local accumulation threshold. If it exceeds the preset local accumulation threshold, it indicates that there is local accumulation at the location of the detection window. In this case, pile is set to pile=1 and the detection is exited. Otherwise, the detection window is moved to the next area and the detection continues until the entire powder-spreading image has been detected.

[0050] Preferably, S92 is as follows:

[0051] First determine the powder coverage index If the coverage rate is below a preset threshold, the substrate is adjusted downwards and step S93 is executed; if it is above the threshold, the powder uniformity index is further evaluated. If the substrate is below the preset uniformity threshold and the local stacking index pile=1, then the substrate is adjusted upwards; otherwise, the substrate is adjusted downwards.

[0052] Preferably, S10, the loop control and termination condition, is as follows:

[0053] Determine continuity Do all the comprehensive powder spreading quality evaluation indicators meet the requirements? Once the desired height is achieved, the substrate height adjustment process ends. It is a positive integer;

[0054] Alternatively, it can determine whether the preset maximum number of adjustments has been reached; if so, the substrate height adjustment will end.

[0055] Or judgment If yes, continue to determine whether the substrate height has been fine-tuned before. If yes, it means that there is an abnormal fluctuation and the substrate height adjustment ends. If no, it means that the substrate height still needs to be coarsely adjusted and then execute S2.

[0056] If none of the above three conditions are met, then execute S5 to continue fine-tuning the substrate height.

[0057] The present invention also provides a calibration system for the above-mentioned machine vision-based 3D printing equipment substrate height calibration method, comprising: a metal 3D printing equipment, an industrial camera, a visual reference calibration block, and a control system;

[0058] The industrial camera and vision reference calibration block are both fixedly mounted on the powder spreading carriage of the metal 3D printing equipment, allowing them to move synchronously during the powder spreading process. The control system is connected to the industrial camera and the cylinder lifting mechanism of the metal 3D printing equipment, and is used to adjust the substrate height according to the acquired images. Specifically:

[0059] The visual reference calibration block is a rigid structural component fixedly installed on the powder spreading cart. Its position is within the field of view of the industrial camera and is not obstructed. The visual reference calibration block is rectangular, and the distance between the upper surface of the visual reference calibration block and the bottom of the powder spreading cart or the working surface of the scraper is a known fixed value. The upper surface of the calibration reference block has a calibration pattern, which is used by the industrial camera to extract feature points and locate pixel coordinates. The calibration pattern is a regular geometric array, which is used as a height reference in image measurement.

[0060] The industrial camera is fixedly mounted on the support structure of the powder spreading carriage of the 3D printing equipment. The support structure is rigidly connected to the powder spreading carriage, so that the industrial camera moves synchronously with the powder spreading carriage. The field of view of the industrial camera is set to be able to acquire the substrate image and the calibration pattern in the same image at the same time.

[0061] The control system is connected to an industrial camera and a metal 3D printing equipment. It determines the required height of the substrate based on images captured by the industrial camera and sends this value to the metal 3D printing equipment. The control system includes an image processing module, a substrate height determination module, a coarse adjustment module, and a fine adjustment module. The image processing module acquires images from the industrial camera and preprocesses them to obtain substrate and powder-laying images. The image processing module then sends the substrate image to the substrate height determination module to determine the substrate height. The substrate height determination module determines the substrate height. The data is sent to the coarse adjustment module, which then obtains the height deviation. The coarse adjustment module will adjust the height deviation. The control signal is converted and sent to the cylinder lifting mechanism of the metal 3D printing equipment to adjust the substrate height; the image processing module sends the powder spreading image to the fine-tuning module to obtain the substrate height adjustment step size. The fine-tuning module adjusts the substrate height in steps. The cylinder lifting mechanism of the metal 3D printing equipment is converted into a control signal and sent to adjust the height of the substrate.

[0062] Preferably, the substrate height determination module obtains the substrate height. Specifically:

[0063] The substrate height of the metal 3D printing equipment is calibrated by a visual reference calibration block set on the powder spreading carriage. The visual reference calibration block has preset height information. An industrial camera captures an image containing the substrate and the visual reference calibration block. Based on the proportional relationship between the size of the visual reference calibration block and the substrate, and the correspondence between the pixel position of the visual reference calibration block in the image and the actual height, the current substrate height is calculated.

[0064] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0065] 1. This invention introduces a machine vision-based substrate height and powder spreading status monitoring mechanism: a high-resolution industrial camera is integrated into a powder bed metal 3D printing device. By imaging the substrate surface and powder spreading area and analyzing the image features, the automatic identification of substrate height status and powder spreading effect is achieved. Compared with traditional manual observation methods, this method can determine the platform height and powder spreading quality in a non-contact, objective, and quantitative manner, improving the consistency and reliability of the judgment.

[0066] 2. This invention employs a two-stage precision height calibration mechanism: First, by analyzing the bare substrate image, the height deviation of the substrate relative to the processing surface is quickly determined, and the forming piston is controlled to automatically adjust the substrate to a preset position slightly lower than the processing surface, achieving rapid initial height adjustment. Subsequently, during the automatic powder spreading process, by analyzing the image after powder spreading, it is identified whether the powder spreading has reached the first layer processing state, and the substrate height is further fine-tuned to achieve high-precision height calibration. This two-stage adjustment mechanism balances adjustment speed and adjustment accuracy, avoiding the problem of excessively long time consumption in a single fine adjustment process.

[0067] 3. The present invention uses a first-layer powder coating control method based on gradual adjustment of small layer thickness: In the first-layer powder coating stage, there is a highly sensitive coupling relationship between the substrate height and the powder coating layer thickness. Therefore, the present invention achieves precise control of the first-layer powder coating state through machine vision feedback and multiple micro height adjustments. Attached Figure Description

[0068] Figure 1 This is a flowchart of the machine vision-based 3D printing equipment substrate height calibration method of the present invention;

[0069] Figure 2 This is a schematic diagram of the system of the machine vision-based 3D printing equipment substrate height calibration method of the present invention.

[0070] Key reference numerals:

[0071] 1. Industrial camera; 2. Powder spreading vehicle; 3. Vision reference calibration block; 4. Substrate; 5. Cylinder lifting mechanism. Detailed Implementation

[0072] To fully explain the technical content, objectives, and effects of this invention, the embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0073] To address the limitations in accuracy and efficiency of existing technologies where substrate height adjustment relies on manual methods, this invention proposes a machine vision-based substrate height calibration method for 3D printing equipment. Figure 1 As shown, the specific steps include the following:

[0074] S1, initialize the system and parameters.

[0075] Initialize the metal 3D printing equipment and initialize the automatic calibration parameters for substrate height: , , ;in The number of times the substrate height is fine-tuned; This indicates the number of times the comprehensive powder coating quality evaluation index continuously meets the processing conditions of the first layer; This indicates whether the substrate height has been fine-tuned; 0 indicates that it has not been fine-tuned yet.

[0076] S2, acquire the substrate image.

[0077] An industrial camera 1, positioned above the side of the powder spreading vehicle 2, is controlled to capture an initial substrate image containing the substrate and a visual reference calibration block 3. The industrial camera is set at an angle so that the initial substrate image it captures includes the substrate 4 and the visual reference calibration block 3.

[0078] The initial substrate image is processed by distortion correction and perspective transformation to obtain the substrate image, thus eliminating the measurement error caused by the camera's tilt shooting.

[0079] S3, determine the substrate height.

[0080] The current substrate height is obtained from the visual reference calibration block. substrate height This refers to the height of substrate 4 relative to the processing reference surface. Specifically, the distance between the upper surface of the visual reference calibration block 3 and the scraper working surface is a known fixed value. Based on the proportional relationship between the size of the visual reference calibration block 3 and the substrate 4, and based on the correspondence between the pixel position of the visual reference calibration block 3 in the substrate image and its actual height, the substrate height is calculated according to the mapping relationship. That is, the height relative to the working surface of the scraper.

[0081] S4, coarsely adjust the substrate height.

[0082] Current substrate height Height relative to the preset machining reference surface By comparison, the height deviation is obtained. The height of the preset machining reference surface Preset parameters for the equipment or obtain them through calibration.

[0083] Height deviation This serves as the substrate height adjustment amount, which is then converted into a control signal and sent to the cylinder lifting mechanism 5. This signal drives the forming piston to adjust the substrate height to the preset processing reference surface height. Please note the height deviation here. The symbols indicate that substrate 4 needs to be adjusted downwards, and negative symbols indicate that substrate 4 needs to be adjusted upwards. Preferably, the initial substrate height is... All are below the height of the preset machining reference surface. .

[0084] S5, obtain the powder spreading image.

[0085] The powder spreading image obtained after performing one powder spreading operation is as follows:

[0086] After performing a powder spreading operation, the industrial camera 1 is controlled to acquire an initial powder spreading image including the substrate 4 and the visual reference calibration block 3.

[0087] Based on the position of the visual reference calibration block 3 in the initial powder spreading image, the powder spreading area is located, and the powder spreading area is determined by a preset area mapping relationship or a method based on image boundary recognition. The powder spreading area is then cropped from the initial powder spreading image to obtain a cropped powder spreading image.

[0088] The cropped powder-spreading image is preprocessed, including distortion correction, perspective transformation, grayscale conversion, filtering and denoising, and brightness normalization. According to the preprocessing steps, the powder-spreading image is a grayscale image.

[0089] S6 calculates the index based on the powder-spreading image.

[0090] The indicators include powder coverage, powder uniformity, comprehensive powder spreading quality evaluation, and local accumulation. The comprehensive powder spreading quality evaluation indicator is obtained by weighting the powder coverage and powder spreading uniformity indicators.

[0091] S61, extract feature parameters of the powder spreading image.

[0092] Based on the powder-spreading image, gray-level distribution features are extracted. The gray-level distribution features include at least the pixel gray-level value, the gray-level threshold range corresponding to the powder area, the average value and standard deviation of the pixel gray-level value.

[0093] S62, obtain the powder coverage index.

[0094] Powder coverage index Calculated based on the pixel proportion of the powder region in the powder-spreading image:

[0095] (1)

[0096] in, This refers to the powder coverage index. This represents the total number of pixels in the powder-coated image. For pixels grayscale value; This represents the grayscale threshold range corresponding to the powder region. For indicator functions, when belong The value is 1 if it is true, and 0 otherwise.

[0097] S63, obtain the powder spreading uniformity index.

[0098] Powder uniformity index It was calculated based on the statistical characteristics of the gray-scale distribution in the powder-spreading area.

[0099] ; (2)

[0100] in, For the uniformity of powder spreading, This is the average grayscale value of all pixels in the powder-coated image. is the standard deviation of the grayscale values ​​of all pixels in the powder-coated image.

[0101] S64, Obtain comprehensive powder spreading quality evaluation indicators .

[0102] ; (3)

[0103] in, , These are the weighting coefficients, and It is used to balance the requirements of coverage integrity and uniformity.

[0104] S65, obtain local accumulation index.

[0105] Local accumulation is represented by the pile index, initially pile=0 indicating no local accumulation. In this embodiment, a detection window is used for local accumulation detection. The size of the detection window is usually set according to the powder-spreading image. In the powder-spreading image, the detection window traverses the pixels in the image with a step size of one pixel at a time. The average grayscale value of all pixels in the detection window is compared with a preset local accumulation threshold. If it exceeds the preset local accumulation threshold, it indicates that there is local accumulation at the location of the detection window, and pile is set to pile=1 and the detection is exited. Otherwise, the detection window is moved to the next area and the detection continues until the entire powder-spreading image has been detected.

[0106] S7, determine whether to perform substrate height adjustment.

[0107] Based on the comprehensive powdering quality evaluation index If the current powder spreading state meets the first-layer processing conditions, proceed to S10; otherwise, execute S8. Specifically:

[0108] when If the current powder coating condition is deemed to meet the first layer processing requirements, then no substrate height adjustment will be performed, and the process will proceed. Then, jump to S10.

[0109] when If the current powder spreading state does not meet the conditions for the first layer of processing, execute S8.

[0110] in, The first-layer processing judgment threshold is a pre-set quality evaluation threshold.

[0111] S8, determine whether to perform fine-tuning of the substrate height.

[0112] when When the current powder spreading state is determined to be within the critical range of the first layer processing, then... Then execute S9 to enter the fine-tuning substrate height mode; otherwise execute S10.

[0113] S9, fine-tuning the substrate height, specifically:

[0114] S91, step size calculation.

[0115] This application does not adopt a one-time height compensation method, but introduces a small-layer gradual adjustment strategy to limit the height adjustment to a small step size range.

[0116] Based on the comprehensive powdering quality evaluation index With the first layer processing judgment threshold The deviation between them is used to calculate the substrate height adjustment step size. Specifically:

[0117] ; (4)

[0118] in, The nominal layer thickness for a single powder application is a preset parameter of the equipment. It typically refers to the nominal thickness of the metal powder layer formed after a single application of powder by the doctor blade. This thickness directly affects printing accuracy and surface quality. The threshold for determining whether the powder spreading state is in the critical range of the first layer processing; This is the threshold for determining the first layer of processing, used to determine whether the current powder spreading state meets the conditions for the first layer of processing.

[0119] S92, determine the adjustment direction.

[0120] The direction for adjusting the substrate height is determined based on the comprehensive powder coating quality evaluation indicators, specifically as follows:

[0121] First, determine the powder coverage index. If the coverage rate is below the preset threshold, it indicates that the substrate height is too high, and substrate 4 is adjusted downwards, proceeding to step S93; if it is above the threshold, the powder uniformity index is further evaluated. If the substrate height is below the preset uniformity threshold and the local stacking index pile=1, it indicates that the substrate height is too low, and substrate 4 is adjusted upwards; otherwise, it indicates that the substrate height is too high, and substrate 4 is adjusted downwards.

[0122] S93, adjust the substrate height.

[0123] Adjust the step size according to the substrate height Adjust the substrate height and direction, and record the number of adjustments. Specifically:

[0124] Substrate height adjustment step size determined based on S91 Based on the adjustment direction determined by S92, the height adjustment step of the substrate with the specified direction is adjusted. The substrate height adjustment amount is used as the substrate height adjustment amount, and the substrate height adjustment amount is converted into a control signal and sent to the cylinder lifting mechanism 5 to drive the molding piston.

[0125] Perform substrate height adjustment and record the number of adjustments. .

[0126] Preferably, it also includes updating the substrate height, for adjusting the final substrate height and the preset processing reference surface height. The values ​​are compared and used as error values ​​when coarsely adjusting the substrate height, so that the height deviation can be determined more accurately in the next coarse adjustment of the substrate height.

[0127] S10, Loop control and termination condition.

[0128] Determine whether the preset conditions for ending substrate height adjustment are met. If they are met, end the adjustment; otherwise, determine whether the fine-tuning of substrate height has been performed before. If it has, execute S5; otherwise, execute S2. Specifically:

[0129] judge If so, it means continuous. All comprehensive powder spreading quality evaluation indicators meet the requirements. End substrate height adjustment. It is a positive integer.

[0130] Or judgment Is this condition met? If so, it means the preset maximum number of adjustments has been reached. End substrate height adjustment.

[0131] Or judgment If yes, then continue to determine whether. If so, it indicates an abnormal fluctuation in substrate height adjustment, and the substrate height adjustment is terminated; otherwise, it indicates that further coarse adjustment of the substrate height is needed, so execute S2.

[0132] If none of the above three conditions are met, then execute S5 to continue fine-tuning the substrate height.

[0133] like Figure 2 As shown, this invention also discloses a machine vision-based 3D printing equipment substrate height calibration system, including a metal 3D printing equipment, an industrial camera 1, a visual reference calibration block 3, and a control system. The industrial camera 1 and the visual reference calibration block 3 are both fixedly mounted on the powder spreading carriage 2 of the metal 3D printing equipment, enabling them to move synchronously during powder spreading to ensure the stability of the calibration relationship. The control system is connected to the industrial camera 1 and the cylinder lifting mechanism 5 of the metal 3D printing equipment, and is used to adjust the substrate height according to the acquired images, specifically:

[0134] The visual reference calibration block 3 is a rigid structural component fixedly installed on the powder spreading cart 2. Its position must be within the field of view of the industrial camera 1 and unobstructed, preferably located in the non-working interference area of ​​the powder spreading cart 2 near the scraper. The visual reference calibration block 3 is fixed to the powder spreading cart 2 by bolt connection or integral machining to ensure its positional stability during equipment operation. The visual reference calibration block 3 is rectangular, and the distance between its upper surface and the bottom of the powder spreading cart 2 or the working surface of the scraper is a known fixed value. The upper surface of the calibration block has a calibration pattern used by the industrial camera 1 for feature point extraction and pixel coordinate positioning. The calibration pattern is a regular geometric array, preferably a checkerboard pattern or a dot matrix pattern, used as a height reference in image measurement.

[0135] Industrial camera 1 is fixedly mounted on the support structure of powder spreading cart 2. The support structure is rigidly connected to powder spreading cart 2, allowing industrial camera 1 to move synchronously with powder spreading cart 2. The support structure is usually located on one side or at the front of powder spreading cart 2. Industrial camera 1 is set at an inclined angle toward substrate 4, and its field of view is set to simultaneously acquire the substrate surface and visual reference calibration block 3 in the same image. This ensures that the substrate image and calibration pattern are acquired simultaneously in the same image, avoiding errors caused by multiple acquisitions. Through the above structural settings, industrial camera 1 can stably acquire images containing the substrate surface and calibration structure during the movement of powder spreading cart 2, realizing real-time measurement and calibration of substrate height.

[0136] The control system is connected to the industrial camera 1 and the metal 3D printing equipment. It is used to obtain the height value of the substrate 4 that needs to be adjusted based on the image captured by the industrial camera 1 and send it to the metal 3D printing equipment. The control system includes: an image processing module, a substrate height determination module, a coarse adjustment module, and a fine adjustment module.

[0137] The image processing module is used to acquire images from the industrial camera 1 and obtain substrate images and powder coating images through image preprocessing.

[0138] The image processing module sends the substrate image to the substrate height determination module to obtain the substrate height. The substrate height of the metal 3D printing equipment is calibrated by a visual reference calibration block 3 set on the powder spreading carriage 2. Since the visual reference calibration block 3 has preset height information, the industrial camera 1 captures an image containing the substrate 4 and the visual reference calibration block 3. Based on the proportional relationship between the size of the visual reference calibration block 3 and the substrate 4, and based on the correspondence between the pixel position of the visual reference calibration block 3 in the image and the actual height, the current substrate height is calculated.

[0139] The substrate height determination module determines the substrate height. The data is sent to the coarse adjustment module, which then obtains the height deviation. The coarse adjustment module will adjust the height deviation. The cylinder lifting mechanism 5 of the metal 3D printing equipment is converted into a control signal and sent to adjust the height of the substrate.

[0140] The image processing module sends the powder-laying image to the fine-tuning module to obtain the substrate height adjustment step size. The fine-tuning module adjusts the substrate height in steps. The cylinder lifting mechanism 5 of the metal 3D printing equipment is converted into a control signal and sent to adjust the height of the substrate.

[0141] The machine vision-based 3D printing equipment substrate height calibration method and system disclosed in this invention realizes closed-loop control of automatic substrate height calibration and first layer powder laying status judgment, reduces manual intervention, improves the automation and repeatability of the pre-printing preparation process, and provides a reliable foundation for subsequent stable forming.

[0142] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for calibrating the height of a substrate in a 3D printing device based on machine vision, characterized in that: It includes the following steps: S1, initialize the system and parameters; S2, Obtain substrate image: The substrate image includes the substrate and a visual reference calibration block; S3, Determine substrate height: Obtain substrate height based on visual reference calibration block. ; S4, Coarsely adjust substrate height: Adjust the substrate height Height relative to the preset machining reference surface The height deviation was obtained by comparison. As the substrate height adjustment amount, the substrate height is adjusted to the preset processing reference surface height; S5, Obtain powder spreading image: Obtain the powder spreading image after performing one powder spreading operation; S6, Calculate indicators based on the powder spreading image: Indicators include powder coverage indicators. Powder uniformity index Comprehensive powder spreading quality evaluation indicators and the localized accumulation index pile; S7, Determine whether to perform substrate height adjustment: Based on comprehensive powder coating quality evaluation indicators. Determine whether the current powder spreading status meets the first layer processing conditions. If it does, proceed to S10; otherwise, execute S8. S8, Determine whether to perform fine-tuning of substrate height: When If the current powder spreading state is determined to be in the critical range of the first layer processing, execute S9; otherwise, execute S10. S9, fine-tuning the substrate height; S9 specifically includes S91-S93: S91, Step Length Calculation: Based on the comprehensive powder spreading quality evaluation index With the first layer processing judgment threshold The deviation between them is used to calculate the substrate height adjustment step size. Specifically: ； in, The number of times the substrate height is finely adjusted. The nominal layer thickness for a single application of powder. The threshold for determining whether the powder spreading state is in the critical range of the first layer processing; The threshold for determining the first layer of processing; S92, Determine the adjustment direction: Determine the adjustment direction of the substrate height based on the comprehensive powder coating quality evaluation index; S93, Adjust substrate height: Adjust the step size according to the substrate height. Adjust the orientation and height of the substrate, and record the number of adjustments; S10, Loop control and termination condition; Determine whether the preset conditions for ending substrate height adjustment are met. If they are met, end the adjustment. If not, determine whether the substrate height has been fine-tuned. If it has, execute S5. If it has not, execute S2.

2. The machine vision-based 3D printing equipment substrate height calibration method according to claim 1, characterized in that: The initialization parameters in S1 include at least the following: Initialize the automatic calibration parameters for substrate height: , , ;in The number of times the substrate height is fine-tuned; This indicates the number of times the comprehensive powder coating quality evaluation index continuously meets the processing conditions of the first layer; This indicates whether the substrate height has been fine-tuned; 0 indicates that it has not been fine-tuned yet.

3. The machine vision-based 3D printing equipment substrate height calibration method according to claim 1, characterized in that: S2 specifically refers to: An industrial camera positioned above the side of the powder spreading vehicle is used to capture an initial substrate image containing the substrate and a visual reference calibration block. The industrial camera is set at an angle so that the initial substrate image it captures includes the substrate and the visual reference calibration block. Distortion correction and perspective transformation processing are performed on the captured initial substrate image to obtain the substrate image.

4. The machine vision-based 3D printing equipment substrate height calibration method according to claim 3, characterized in that: S3 specifically refers to: The distance between the upper surface of the visual reference calibration block and the working surface of the scraper is a known fixed value. Based on the proportional relationship between the size of the visual reference calibration block and the substrate, and based on the correspondence between the pixel position of the visual reference calibration block in the substrate image and its actual height, the substrate height is calculated according to the mapping relationship. .

5. The machine vision-based 3D printing equipment substrate height calibration method according to claim 1, characterized in that: S5 specifically refers to: After performing a powder spreading operation, control the industrial camera to capture the initial powder spreading image, which includes the substrate and the visual reference calibration block; Based on the position of the visual reference calibration block in the initial powder spreading image, the powder spreading area is located, and the powder spreading area is determined by a preset area mapping relationship or a method based on image boundary recognition. The powder spreading area is then cropped from the initial powder spreading image to obtain a cropped powder spreading image. The cropped powder-spreading image is preprocessed, including distortion correction, perspective transformation, grayscale conversion, filtering and denoising, and brightness normalization to obtain the powder-spreading image.

6. The machine vision-based 3D printing equipment substrate height calibration method according to claim 5, characterized in that: S6 specifically refers to: S61, Extract feature parameters of the powder spreading image; Based on the powder-spreading image, gray-level distribution features are extracted. The gray-level distribution features include at least the pixel gray-level value, the gray-level threshold range corresponding to the powder area, the average value and standard deviation of the pixel gray-level value. S62, obtain the powder coverage index; Powder coverage index Calculated based on the pixel proportion of the powder region in the powder-spreading image: ； in, This refers to the powder coverage index. This represents the total number of pixels in the powder-coated image. For pixels grayscale value; This represents the grayscale threshold range corresponding to the powder region. For indicator functions, when belong The value is 1 if it is true, and 0 otherwise. S63, obtain the powder spreading uniformity index; Powder uniformity index The following calculations were made based on the statistical characteristics of the grayscale distribution in the powder-coated area: ； in, For the uniformity of powder spreading, This is the average grayscale value of all pixels in the powder-coated image. This represents the standard deviation of the grayscale values ​​of all pixels in the powder-coated image. S64, Obtain comprehensive powder spreading quality evaluation indicators ; ； in, , These are the weighting coefficients, and This is used to balance the requirements of coverage integrity and uniformity; S65, obtain local accumulation index; The local accumulation index is represented by "pile". Initially, pile=0 indicates that there is no local accumulation. Local accumulation detection is performed using a detection window. In the powder-spreading image, the detection window traverses the pixels in the powder-spreading image with a step size of one pixel at a time. The average gray value of all pixels in the detection window is compared with the preset local accumulation threshold. If it exceeds the preset local accumulation threshold, it indicates that there is local accumulation at the location of the detection window. In this case, pile is set to pile=1 and the detection is exited. Otherwise, the detection window is moved to the next area and the detection continues until the entire powder-spreading image has been detected.

7. The machine vision-based 3D printing equipment substrate height calibration method according to claim 1, characterized in that: S92 specifically refers to: First determine the powder coverage index If the coverage rate is below a preset threshold, the substrate is adjusted downwards and step S93 is executed; if it is above the threshold, the powder uniformity index is further evaluated. If the substrate is below the preset uniformity threshold and the local stacking index pile=1, then the substrate is adjusted upwards; otherwise, the substrate is adjusted downwards.

8. The machine vision-based 3D printing equipment substrate height calibration method according to claim 1, characterized in that: S10, Loop Control and Termination Condition, specifically: Determine continuity Do all the comprehensive powder spreading quality evaluation indicators meet the requirements? Once the desired height is achieved, the substrate height adjustment process ends. It is a positive integer; Alternatively, it can determine whether the preset maximum number of adjustments has been reached; if so, the substrate height adjustment will end. Or judgment If yes, continue to determine whether the substrate height has been fine-tuned before. If yes, it means that there is an abnormal fluctuation and the substrate height adjustment ends. If no, it means that the substrate height still needs to be coarsely adjusted and then execute S2. If none of the above three conditions are met, then execute S5 to continue fine-tuning the substrate height.

9. A machine vision-based 3D printing equipment substrate height calibration system, the calibration system being used in the machine vision-based 3D printing equipment substrate height calibration method of claim 1, characterized in that: The calibration system includes: metal 3D printing equipment, industrial camera, visual reference calibration block and control system; The industrial camera and vision reference calibration block are both fixedly mounted on the powder spreading carriage of the metal 3D printing equipment, allowing them to move synchronously during the powder spreading process. The control system is connected to the industrial camera and the cylinder lifting mechanism of the metal 3D printing equipment, and is used to adjust the substrate height according to the acquired images. Specifically: The visual reference calibration block is a rigid structural component fixedly installed on the powder spreading cart. Its position is within the field of view of the industrial camera and is not obstructed. The visual reference calibration block is rectangular, and the distance between the upper surface of the visual reference calibration block and the bottom of the powder spreading cart or the working surface of the scraper is a known fixed value. The upper surface of the calibration reference block has a calibration pattern, which is used by the industrial camera to extract feature points and locate pixel coordinates. The calibration pattern is a regular geometric array, which is used as a height reference in image measurement. The industrial camera is fixedly mounted on the support structure of the powder spreading carriage of the 3D printing equipment. The support structure is rigidly connected to the powder spreading carriage, so that the industrial camera moves synchronously with the powder spreading carriage. The field of view of the industrial camera is set to be able to acquire the substrate image and the calibration pattern in the same image at the same time. The control system is connected to an industrial camera and a metal 3D printing equipment. It determines the required height of the substrate based on images captured by the industrial camera and sends this value to the metal 3D printing equipment. The control system includes an image processing module, a substrate height determination module, a coarse adjustment module, and a fine adjustment module. The image processing module acquires images from the industrial camera and preprocesses them to obtain substrate and powder-laying images. The image processing module then sends the substrate image to the substrate height determination module to determine the substrate height. The substrate height determination module determines the substrate height. The data is sent to the coarse adjustment module, which then obtains the height deviation. The coarse adjustment module will adjust the height deviation. The control signal is converted and sent to the cylinder lifting mechanism of the metal 3D printing equipment to adjust the substrate height; the image processing module sends the powder spreading image to the fine-tuning module to obtain the substrate height adjustment step size. The fine-tuning module adjusts the substrate height in steps. The cylinder lifting mechanism of the metal 3D printing equipment is converted into a control signal and sent to adjust the height of the substrate.

10. The machine vision-based 3D printing equipment substrate height calibration system according to claim 9, characterized in that: The substrate height determination module obtains the substrate height. Specifically: The substrate height of the metal 3D printing equipment is calibrated by a visual reference calibration block set on the powder spreading carriage. The visual reference calibration block has preset height information. An industrial camera captures an image containing the substrate and the visual reference calibration block. Based on the proportional relationship between the size of the visual reference calibration block and the substrate, and the correspondence between the pixel position of the visual reference calibration block in the image and the actual height, the current substrate height is calculated.

Images