A laser engraving method for a flower roller

Abstract

A laser engraving method of a flower roller comprises the following steps: S1, determining a reference laser power and a reference absorption rate; S2, collecting image data of a surface of the flower roller in real time through a coaxial vision device; S3, processing the image data to extract an impurity influence parameter of each processing area of the surface of the flower roller; S4, calculating an actual laser absorption rate of a current processing area according to the impurity influence parameter; S5, calculating a laser power compensation coefficient K; and S6, compensating the standard laser power according to the laser power compensation coefficient K, and performing laser engraving on the current processing area with the compensated actual laser power P. The method solves the problem that the engraving quality is inconsistent in the laser processing process due to the changes of the surface oxidation and roughness of the flower roller.

Description

Technical Field

[0001] This invention relates to a laser engraving method for flower rollers, belonging to the field of laser engraving technology. Background Technology

[0002] Embossing rollers, also known as texturing rollers, are widely used for surface texturing of materials such as paper, leather, and plastic films. The quality of their surface engraving directly affects the clarity of the pattern and the printing quality of the final product. Laser engraving technology, due to its advantages such as non-contact processing, high precision, and good flexibility, has become the mainstream process for engraving patterns on the surface of embossing rollers.

[0003] In laser engraving, a laser beam acts on the surface of a pattern roller, causing the material to vaporize or melt through a photothermal effect, thus forming the engraved pattern. Theoretically, given a laser power, scanning speed, and spot size, the amount of material removed depends on the coupling efficiency between the laser energy and the material, i.e., the material's laser absorptivity. However, among the material properties affecting the quality of laser engraving, the most important is the material's laser absorptivity, followed by thermophysical parameters such as thermal conductivity and coefficient of linear expansion. In addition, the surface condition of the material also has a significant impact on the absorptivity.

[0004] In the practice of laser engraving rollers, for roller surfaces with impurities such as oxide scale and oil stains, existing technologies typically employ surface pretreatment methods such as mechanical grinding and chemical etching. For example, a grinding wheel is used to polish a smooth metal surface, or acidic substances are used to etch away a thin layer of the metal surface to improve the material's absorption efficiency for laser light. For engraved rollers, post-processing often involves grinding and polishing to remove remelted material produced by laser engraving and reduce surface roughness.

[0005] However, the following limitations still exist:

[0006] Firstly, pretreatment methods such as grinding and polishing only address the overall condition before engraving and cannot resolve real-time absorption rate fluctuations caused by localized residual impurities (e.g., oxide impurities), heating oxidation, or differences in substrate microstructure during the engraving process. Specifically, grinding and polishing pretreatment methods can only improve the initial state of the roller surface. In actual engraving, they face the following challenges: first, localized residual impurities left from pretreatment; second, new surface state changes introduced into the surrounding areas to be engraved during laser engraving, such as oxidation; and third, inherent differences in the substrate microstructure of the raw material, leading to changes in the microstructure of the surrounding areas during laser engraving, manifested as variations in surface roughness. In short, even with pretreatment, the degree of residual oxides and differences in surface microstructure in different areas still exist during actual engraving, causing continuous fluctuations in laser absorption rate during the engraving process, thus affecting the consistency of engraving depth and processing accuracy.

[0007] Secondly, grinding is a material removal process. For patterned rollers with fine engravings or those that need to retain specific surface morphology, grinding will damage the outline accuracy and microstructure of the pattern. In addition, grinding, polishing and other processes require additional independent processes and re-clamping, which introduces positioning errors and reduces production efficiency.

[0008] Therefore, there is an urgent need for a method that can identify the surface condition in real time and compensate the laser power online during the engraving process, so as to achieve uniform engraving of the entire roller surface without adding processes or damaging the engraved pattern. Summary of the Invention

[0009] To address the shortcomings of existing laser engraving techniques for patterned rollers, such as fluctuations in real-time absorption rate caused by localized residual impurities, heating oxidation, or differences in substrate structure, which affect the consistency of engraving depth and processing accuracy, the present invention aims to provide a laser engraving method for patterned rollers. This method achieves uniform engraving across the entire roller surface by online identification of surface impurities in each processing area and real-time compensation of laser power, without adding any additional steps or damaging the already engraved pattern.

[0010] A laser engraving method for a pattern roller includes the following steps:

[0011] Step S1: Determine the reference laser power and benchmark absorption The reference laser power The reference absorptivity is the laser power required to achieve the target engraving depth on a standard clean surface of the flower roller substrate. The intrinsic absorption rate of the laser on the standard clean surface of the flower roller substrate;

[0012] Step S2: Real-time image data of the flower roller surface is acquired using a coaxial vision device;

[0013] Step S3: Process the image data and extract the impurity influence parameters of each processing area on the surface of the flower roller. The impurity influence parameters include at least one of the following: oxide coverage area ratio, oxide thickness, and surface roughness.

[0014] Step S4: Calculate the actual laser absorption rate of the current processing area based on the impurity influence parameters. The calculation formula is as follows:

[0015] ,in, This represents the actual laser absorption rate. As the baseline absorption rate, These are the oxide coverage area ratio coefficient, oxide scale thickness coefficient, and surface roughness coefficient, respectively. This represents the percentage of the area covered by oxidation. For oxide scale thickness, For surface roughness;

[0016] Step S5: Calculate the laser power compensation coefficient K based on the ratio of the actual laser absorptivity to the intrinsic absorptivity.

[0017] Step S6: Compensate the standard laser power according to the laser power compensation coefficient K to obtain the compensated actual laser power P, and use the compensated actual laser power P to perform laser engraving on the current processing area.

[0018] Preferably, the reference absorption rate Obtained through the following methods:

[0019] Obtain the refractive index of the flower roller substrate at the laser working wavelength. and extinction coefficient ;

[0020] Using Fresnel's formula Obtained through calculation.

[0021] Preferably, in step S3, the method for extracting the oxidation coverage area ratio is as follows:

[0022] The image data is binarized based on a threshold segmentation algorithm to distinguish between oxidized and non-oxidized regions.

[0023] The proportion of pixels in the oxidized area to the total number of pixels is used as the oxidized area ratio.

[0024] Preferably, in step S3, the method for extracting the oxide layer thickness is as follows:

[0025] Multispectral images of the flower roller surface are acquired using the coaxial vision device;

[0026] Based on the reflectance characteristics of the multispectral images in different bands, the chemical composition of the surface oxides is identified;

[0027] The thickness of the oxide layer is calculated based on the extinction coefficients corresponding to the chemical components and the degree of reflectance attenuation of the multispectral image.

[0028] Preferably, the method for extracting the surface roughness R2 is as follows:

[0029] A linear laser stripe is projected onto the surface of the pattern roller through the line structured light projector in the coaxial vision device, and the linear laser stripe extends along the axial direction of the pattern roller.

[0030] The camera in the coaxial vision device acquires the deformation image of the linear laser stripes on the surface of the patterned roller from a direction at a preset angle to the projection direction of the linear structured light.

[0031] Based on the lateral offset of the laser stripes at each pixel position in the deformed image Combined with pre-calibrated offset-height conversion coefficients Calculate the height value at the corresponding position on the surface of the flower roller. Generate a height map of the surface of the flower roller. ;

[0032] Surface roughness is calculated using the following formula: ,

[0033] in, For sampling length, This is the arithmetic mean of the surface profile height.

[0034] Preferably, in step S2: the optical axis of the coaxial vision device is coaxially set with the laser processing optical axis, so that the center of the image data acquisition field of view coincides with the position of the laser focus on the surface of the pattern roller; the frame rate of the image data acquisition is not less than 10 times the rotation frequency of the pattern roller.

[0035] Preferably, the following post-processing steps are also included:

[0036] After completing the surface engraving of the flower roller, a surface image of the finished flower roller is acquired;

[0037] Based on the surface image of the finished flower roller, extract the actual engraving depth distribution data;

[0038] The actual carving depth distribution data is compared with the target carving depth to calculate the carving depth deviation value;

[0039] When the engraving depth deviation exceeds a preset threshold, the reference laser power is adjusted according to the deviation value. And store it in the process parameter database.

[0040] In summary, the present invention has the following beneficial effects:

[0041] 1. This invention acquires and processes image data of the roller surface in real time, extracts multi-dimensional impurity influence parameters such as the proportion of oxide coverage area, oxide thickness and surface roughness, and incorporates multiple factors such as surface chemical state (oxide composition), physical morphology (roughness) and macroscopic coverage (oxidation area) into a unified compensation calculation framework to achieve uniform engraving on the entire roller surface without adding processes or damaging the engraved pattern.

[0042] 2: This invention, through post-processing steps, makes the reference laser power in the database continuously approach the optimal value under various working conditions, thereby gradually improving the engraving quality of subsequent batches of flower rollers and continuously improving processing consistency. Attached Figure Description

[0043] Figure 1 This is a flowchart of the laser engraving method for the flower roller in Example 1. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0045] Example 1:

[0046] A laser engraving method for a patterned roller includes the following steps:

[0047] Step S1: Determine the reference laser power and benchmark absorption Reference laser power The reference absorptivity is the laser power required to achieve the target engraving depth on a standard clean surface of the flower roller substrate. The intrinsic absorption rate of the laser on the standard clean surface of the flower roller substrate is given.

[0048] In this embodiment, the substrate of the patterned roller is taken as 45# steel, and its benchmark absorption rate is determined. The specific method is:

[0049] Using Fresnel's formula Calculated, where, where The refractive index of the roller substrate at the laser working wavelength is given. The extinction coefficient of the patterned roller substrate at the laser working wavelength.

[0050] Using a 1064nm laser wavelength as the working wavelength, the refractive index and extinction coefficient of the pattern roller substrate are common optical parameters. It is 3.81. It is 4.44;

[0051] Substituting into Fresnel's formula:

[0052]

[0053] That is, the reference absorption rate of No. 45 steel for 1064nm laser is 36%.

[0054] Determine its reference laser power The specific method is:

[0055] Engraving experiments were conducted on a standard clean surface of 45# steel (no oxidation, no oxide scale, roughness Ra≤0.4µm) using different laser powers. The experimental conditions were: scanning speed 500mm / s, spot diameter 50µm, target engraving depth 30µm. Engraving was performed at laser powers of 160W, 180W, 200W, 220W, and 240W, and the actual engraving depth was measured at each power. The results showed that the engraving depth was 22µm at 160W, 26µm at 180W, 30µm at 200W, 33µm at 220W, and 37µm at 240W. The minimum laser power of 200W, which achieved the target engraving depth of 30µm, was determined as the baseline laser power. .

[0056] Step S2: Real-time image data of the roller surface is acquired using a coaxial vision device. The coaxial vision device is a mature acquisition device on the market; its coaxial configuration ensures that the center point of the image captured by the camera is the current position of the laser focus.

[0057] Step S3: Process the image data to extract the impurity influence parameters of each processing area on the surface of the flower roller;

[0058] Specifically, this step includes three parallel subtasks:

[0059] 1. Oxidation coverage area ratio Extraction:

[0060] First, the acquired grayscale image is binarized using the Otsu thresholding algorithm to automatically determine the optimal grayscale threshold for distinguishing between oxidized regions (darker color, lower grayscale value) and normal metal regions (higher grayscale value).

[0061] Then, the proportion of pixels with grayscale values ​​below the threshold is used as the proportion of the oxide coverage area. .

[0062] Specifically, in this embodiment, the image size of the current processing area captured by the industrial camera is 1920×1080 pixels, with a total pixel count of = 2,073,600.

[0063] Statistical analysis of the grayscale values ​​of each pixel yields a histogram of the overall image's grayscale distribution. The optimal segmentation threshold is determined using the Otsu's algorithm, which iterates through all possible grayscale thresholds. (0 to 255), calculate the foreground region for each threshold. With background area Between-class variance The threshold at which the inter-class variance reaches its maximum value is determined as the optimal segmentation threshold. The formula for calculating the inter-class variance is: in, These represent the proportions of foreground and background pixels in the entire image, respectively. These are the average gray values ​​of the foreground and background regions, respectively. This represents the average grayscale value of the entire image.

[0064] Calculations show that the optimal segmentation threshold corresponding to the maximum inter-class variance in this embodiment is: =108 (Grayscale value range is 0 to 255, where 0 is pure black and 255 is pure white).

[0065] At the optimal segmentation threshold At a grayscale value of 108, the image is binarized pixel by pixel: pixels with a grayscale value ≤ 108 are marked as 1 (oxidized region), and pixels with a grayscale value > 108 are marked as 0 (normal metal region).

[0066] Count the number of pixels marked as 1 =725760, which accounts for the proportion of the total number of pixels:

[0067] This means that approximately 35% of the current processing area is covered by rust.

[0068] This ratio will change accordingly when processing areas with varying degrees of rust coverage.

[0069] 2. Oxide Scale Thickness Extraction:

[0070] The first step is multispectral image acquisition. Multispectral images of the patterned roller surface are acquired sequentially in the visible light band (450nm), near-infrared band (850nm), and short-wave infrared band (1550nm) using a coaxial vision device. Image acquisition for each band is completed at the same rotation angle of the patterned roller to ensure that the images of the three bands correspond precisely in space.

[0071] The second step is the identification of the chemical composition of the oxides. Oxides with different chemical compositions have characteristic reflectance spectra: It exhibits a high absorption coefficient in the visible light band (especially in the 450nm blue light region), a significantly reduced reflectivity, and a reddish-brown characteristic. It exhibits a characteristic absorption peak in the near-infrared band (around 850 nm), where the reflectance shows a significant dip compared to adjacent bands. The measured multispectral reflectance curves are shown below. The optical parameters are pre-stored in an oxide optical parameter database (using an optical constant database known in the art, such as the RefractiveIndex.INFO database, or Palik's *Handbook of Optical Constants of Solids*). , The standard spectra of the components were compared with those of other possible oxides. A linear combination of the standard spectra of each component was fitted using the least squares method.

[0072] ,

[0073]

[0074] The third step is to calculate the oxide layer thickness. Based on the identified oxide chemical composition, the relevant parameters are retrieved from the oxide optical parameter database. Extinction coefficient at 1550nm wavelength and In this embodiment, , Calculate the absorption coefficient of each component using the following formulas.

[0075] ,

[0076] by For example, ,

[0077] The reflectance of the oxide-coated areas was extracted from the 1550nm band image. and reflectivity of areas without oxide coating In this embodiment, That is, the reflectivity decreased by about 40%.

[0078] Calculated according to the reflectivity attenuation form of Lambert-Beer law The corresponding oxide layer thickness:

[0079]

[0080] Calculate using the same method. thickness .

[0081] Finally, the weighted average thickness of the oxide scale is calculated using the relative content ratio of each component as the weight:

[0082] .

[0083] That is, the oxide scale thickness in the current processing area is approximately 0.69 μm.

[0084] 3. Surface roughness Extraction:

[0085] A linear laser stripe is projected onto the surface of the pattern roller using a line structured light projector in the coaxial vision device. This stripe extends along the axial direction of the pattern roller. Simultaneously, an image of the linear laser stripe on the surface of the pattern roller is captured by a camera in the coaxial vision device from a predetermined angle (30° in this embodiment) to the projection direction.

[0086] The principle behind its height calculation is as follows: When the surface of the roller is an ideally smooth plane, the laser stripes appear as a straight baseline in the camera image; when the surface has unevenness, the stripes will show a corresponding lateral offset in the image. If the surface height at a certain point changes relative to the baseline plane... The image pixel position of the stripe at that point is offset relative to the baseline. Pixels. Offset With height change The relationship between the two is linear, and the scaling factor is determined by the laser projection angle, camera shooting angle, and lens magnification. It can be obtained through calibration—that is, by using a standard step block with a known height difference for calibration, establishing a mapping relationship between pixel offset and height change, and obtaining the conversion factor. (In this embodiment, the calibration value is 0.5 μm / pixel). Therefore, the surface height value... .

[0087] After the camera acquires the image of the deformed stripes, the center position of the stripes is extracted pixel by pixel, and the lateral offset of each point is obtained by comparing it with the baseline position. Multiply by the conversion factor This allows us to obtain the height value at the corresponding position on the surface of the flower roller, thereby generating a complete height map along the axial direction. ,in The coordinates are along the axis of the flower roller. This is the height value at that location.

[0088] At sampling length Within a range of 4mm (this sampling length is selected according to the recommendations of international standard ISO 4287, which can represent the statistical characteristics of the roller surface and avoids introducing the influence of low-frequency waviness), the arithmetic mean roughness is calculated according to the formula:

[0089] ,in, The arithmetic mean of the surface profile height within the sampling length: In this embodiment, after acquiring the laser stripe deformation image, the surface height distribution within the sampling length is obtained through the above calculations. Substituting into the formula, we get .

[0090] Step S4: Calculate the actual laser absorption rate of the current processing area based on the impurity influence parameters. .

[0091] In this embodiment, a simplified linear model is specifically used for calculation.

[0092] , These are the oxide coverage area ratio coefficient, oxide scale thickness coefficient, and surface roughness coefficient, respectively. Regarding... The confirmation of these three coefficients will be detailed below.

[0093] Step S5, based on the actual laser absorption rate Compared with the benchmark absorption rate The ratio of the two values ​​is used to calculate the laser power compensation coefficient K, which is calculated using the formula K= / .

[0094] Step S6: Adjust the reference laser power according to the laser power compensation coefficient K. Compensation is performed to obtain the actual laser power P, and the current processing area is engraved using the actual laser power P. The formula for calculating the actual laser power P is: .

[0095] Step S7: Repeat steps S2 to S6 until the engraving of the entire flower roller surface is completed.

[0096] The above method extracts multi-dimensional impurity influence parameters such as oxide coverage area ratio, oxide thickness, and surface roughness in step S3, and establishes a quantitative mathematical mapping from impurity influence parameters to laser power compensation coefficient in steps S4 to S6. This incorporates multiple factors such as surface chemical state (oxide composition), physical morphology (roughness), and macroscopic coverage (oxidation area) into a unified compensation calculation framework, achieving uniform engraving across the entire roller surface without adding any processes or damaging the engraved pattern.

[0097] The following will introduce Confirmation of these three coefficients:

[0098] Step 1, Standard Sample Preparation: Take several sets of standard samples of the same material as the roller to be processed. Different oxide coverage area ratios (S), oxide thicknesses (T), and surface roughness (R) are obtained for each set of samples using the following methods:

[0099] Changing the oxidation coverage area ratio S: After coating the sample surface with an anti-oxidation mask, the sample is placed in a heating furnace and heated in air. By controlling the proportion of the mask opening area (0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100% respectively), samples with different S values ​​are obtained.

[0100] Changing the oxide scale thickness T: In an open atmospheric environment, samples with different T values ​​were obtained by controlling the heating time (5 min, 10 min, 20 min, 40 min, and 60 min) and heating temperature (200℃, 300℃, 400℃, and 500℃).

[0101] Changing the surface roughness R: Grinding with sandpaper of different grits (#80, #120, #240, #400, #600, and #800 respectively) to obtain samples with different R values;

[0102] Of course, to ensure the statistical significance of the calibration equation, the total number of samples should not be less than 30.

[0103] The second step is to measure the actual laser absorptivity of each group of samples: using the same laser wavelength and spot size as in step S1, the surface of each standard sample is irradiated with the same laser power, and the actual laser absorptivity of each group of samples is measured by calorimetry. Specifically, the sample is placed in an insulated container, and the reference laser power is used. Irradiate the sample surface for a fixed time Δt, measure the temperature rise ΔT of the sample before and after irradiation, and calculate the absorptivity using the following formula:

[0104] ,

[0105] in, The specific heat capacity of the sample material. The mass of the sample (weighed using a precision balance before measurement). The reference laser power is Δt, the irradiation time is Δt, and the temperature difference before and after irradiation is ΔT (measured by a thermocouple attached to the back of the sample).

[0106] The third step is multiple linear regression calibration: using the measured actual laser absorption rate... With S, T, and R of each group of samples in the first step as the dependent variable, a multiple linear regression fit is performed according to the following formula:

[0107] ,in, The baseline absorption rate (i.e., the one mentioned above) ), , , These are the oxide coverage area ratio coefficient, oxide scale thickness coefficient, and surface roughness coefficient, respectively, which are the above-mentioned... .

[0108] Regression analysis will directly yield the following results. , , The values ​​and their confidence intervals are then determined through goodness-of-fit (i.e., The value, as an aside, The reliability of the model is verified by assuming that R in the model is not the surface roughness R.

[0109] If calibrated =0.15, = , = The goodness of fit obtained by this calibration method This demonstrates that the linear model has good fitting accuracy. , , The value can be used as the oxide coverage area ratio coefficient, oxide scale thickness coefficient, and surface roughness coefficient.

[0110] Example 2:

[0111] The main difference between this embodiment and Embodiment 1 is the addition of a post-processing closed-loop optimization step, specifically including:

[0112] Step S8: Extract the actual engraving depth distribution data based on the surface image of the finished flower roller;

[0113] Step S9: Compare the actual carving depth distribution data with the target carving depth, and calculate the carving depth deviation value;

[0114] Step S10: When the engraving depth deviation value exceeds a preset threshold, the reference laser power is corrected according to the deviation value. And store it in the process parameter database.

[0115] With this design, the reference laser power in the database By continuously approaching the optimal value under various working conditions, the engraving quality of subsequent batches of flower rollers gradually improves, and the consistency of processing continues to improve.

[0116] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the concept and scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the design concept of the present invention should fall within the protection scope of the present invention. The technical content for which protection is sought in this invention has been fully described in the claims.

Claims

1. A laser engraving method for a patterned roller, characterized in that, Includes the following steps Step S1: Determine the reference laser power and benchmark absorption The reference laser power The reference absorptivity is the laser power required to achieve the target engraving depth on a standard clean surface of the flower roller substrate. The intrinsic absorption rate of the laser on the standard clean surface of the flower roller substrate; Step S2: Real-time image data of the flower roller surface is acquired using a coaxial vision device; Step S3: Process the image data and extract the impurity influence parameters of each processing area on the surface of the flower roller. The impurity influence parameters include at least one of the following: oxide coverage area ratio, oxide thickness, and surface roughness. Step S4: Calculate the actual laser absorption rate of the current processing area based on the impurity influence parameters. The calculation formula is as follows: ,in, This represents the actual laser absorption rate. As the baseline absorption rate, These are the oxide coverage area ratio coefficient, oxide scale thickness coefficient, and surface roughness coefficient, respectively. This represents the percentage of the area covered by oxidation. For oxide scale thickness, For surface roughness; Step S5: Calculate the laser power compensation coefficient K based on the ratio of the actual laser absorptivity to the intrinsic absorptivity. Step S6: Compensate the standard laser power according to the laser power compensation coefficient K to obtain the compensated actual laser power P, and use the compensated actual laser power P to perform laser engraving on the current processing area.

2. The laser engraving method for a patterned roller according to claim 1, characterized in that, The benchmark absorption rate Obtained through the following methods: Obtain the refractive index of the flower roller substrate at the laser working wavelength. and extinction coefficient ; Using Fresnel's formula Obtained through calculation.

3. The laser engraving method for a patterned roller according to claim 1, characterized in that, In step S3, the method for extracting the oxidation coverage area ratio is as follows: The image data is binarized based on a threshold segmentation algorithm to distinguish between oxidized and non-oxidized regions. The proportion of pixels in the oxidized area to the total number of pixels is used as the oxidized area ratio.

4. The laser engraving method for a patterned roller according to claim 1, characterized in that, In step S3, the method for extracting the oxide scale thickness is as follows: Multispectral images of the flower roller surface are acquired in multiple bands using the coaxial vision device. The measured multispectral reflectance curves were compared with standard spectral curves in the oxide optical parameter database to identify the chemical composition of the surface oxides and the relative content ratio of each component. Based on the identified oxide chemical composition, the extinction coefficient of the corresponding component in the infrared band is retrieved from the oxide optical parameter database. Calculate the absorption coefficient ; Obtain the reflectance of the oxide-coated and oxide-free regions in the infrared band. According to the formula Calculate the thickness of the oxide layer Furthermore, when multiple oxide components are identified, the thickness of the oxide layer... Take the weighted average of the calculation results for each component.

5. The laser engraving method for a patterned roller according to claim 1, characterized in that, The method for extracting the surface roughness R2 is as follows: A linear laser stripe is projected onto the surface of the pattern roller through the line structured light projector in the coaxial vision device, and the linear laser stripe extends along the axial direction of the pattern roller. The camera in the coaxial vision device acquires the deformation image of the linear laser stripes on the surface of the patterned roller from a direction at a preset angle to the projection direction of the linear structured light. Based on the lateral offset of the laser stripes at each pixel position in the deformed image Combined with pre-calibrated offset-height conversion coefficients Calculate the height value at the corresponding position on the surface of the flower roller. Generate a height map of the surface of the flower roller. ; Surface roughness is calculated using the following formula: , in, For sampling length, It is the arithmetic mean of the surface profile height.

6. The laser engraving method for a patterned roller according to claim 1, characterized in that, In step S2: the optical axis of the coaxial vision device is coaxially set with the laser processing optical axis, so that the center of the field of view for image data acquisition coincides with the position of the laser focus on the surface of the pattern roller; the frame rate of image data acquisition is not less than 10 times the rotation frequency of the pattern roller.

7. The laser engraving method for a patterned roller according to claim 1, characterized in that, It also includes the following post-processing steps: After completing the surface engraving of the flower roller, a surface image of the finished flower roller is acquired; Based on the surface image of the finished flower roller, extract the actual engraving depth distribution data; The actual carving depth distribution data is compared with the target carving depth to calculate the carving depth deviation value; When the engraving depth deviation exceeds a preset threshold, the reference laser power is adjusted according to the deviation value. And store it in the process parameter database.

Images