A UV curing light source dynamic compensation method and system based on visual intelligence

Abstract

The application relates to the field of visual intelligence, and discloses a UV curing light source dynamic compensation method and system based on visual intelligence, which comprises the following steps: setting an adjustable UV light source array for curing materials; collecting real-time images of the material surface during the curing process of the adjustable UV light source array; pre-processing the real-time images to obtain target images, obtaining material surface morphology characteristics based on the target images, and extracting micro-geometric structure variation of a curing reaction area based on the material surface morphology characteristics; calculating a radiation energy distribution map for compensating the UV light source according to the micro-geometric structure variation, generating light source control instructions according to the radiation energy distribution map; and sending the light source control instructions to the adjustable UV light source array, and adjusting the output intensity and irradiation angle of each light emitting unit in the adjustable UV light source array. The application can realize dynamic compensation of the UV light source.

Description

Technical Field

[0001] This invention relates to the field of visual intelligence technology, specifically to a method and system for dynamic compensation of UV curing light sources based on visual intelligence. Background Technology

[0002] UV curing light sources are core equipment for advanced curing technology that use ultraviolet light as an energy source. They irradiate materials containing photosensitive components by emitting ultraviolet light of a specific wavelength, triggering rapid photopolymerization or cross-linking reactions and achieving an instantaneous transformation from liquid to solid state. Compared with traditional thermal curing methods, UV curing light sources have advantages such as high efficiency, low energy consumption, and environmental friendliness. They can complete surface or overall curing in a very short time, significantly improving production speed and finished product quality.

[0003] Existing dynamic compensation strategies are mostly based on simple linear and empirical models. For example, after a photoelectric sensor detects a certain percentage decrease in the output intensity of the light source, the driving current is linearly increased according to a preset ratio for compensation. This strategy has a fundamental flaw: it simplifies the curing process into an idealized, linear energy input and output relationship, ignoring the inherent complex physicochemical characteristics of the material's curing reaction. The energy absorption efficiency of a material is strongly correlated with various dynamic factors such as the degree of curing, real-time temperature, surface morphology, and reaction byproducts, making it a highly nonlinear process. Therefore, a single-dimensional, linear compensation based solely on the light source intensity cannot respond to the actual needs of the material surface, leading to insufficient or over-compensation, resulting in quality problems such as uneven curing, surface defects, or internal stress accumulation. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for dynamic compensation of UV curing light sources based on visual intelligence, thereby solving the above-mentioned technical problems.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A dynamic compensation method for UV curing light sources based on visual intelligence includes the following steps:

[0007] A modulated UV light source array is set up to cure the material, and real-time images of the material surface are acquired during the curing process of the modulated UV light source array.

[0008] The real-time image is preprocessed to obtain the target image, the surface morphology features of the material are obtained based on the target image, and the micro-geometric structure changes of the curing reaction area are extracted based on the surface morphology features of the material.

[0009] The radiation energy distribution map for compensating the UV light source is calculated based on the changes in the micro-geometric structure, and a light source control command is generated based on the radiation energy distribution map.

[0010] The light source control command is sent to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array.

[0011] As a further aspect of the present invention: the preprocessing includes performing illumination uniformity correction and feature enhancement processing on the acquired real-time images;

[0012] The illumination uniformity correction includes:

[0013] Identify overexposed and underexposed areas in real-time images caused by ambient light or UV reflected light, and reconstruct pixel gray values ​​for the overexposed and underexposed areas using a pre-established image gray-scale equalization model.

[0014] The feature enhancement includes: using an edge-preserving filter to enhance the contour information of the micro-texture of the material surface, while suppressing high-frequency noise introduced during image acquisition.

[0015] As a further aspect of the present invention: obtaining the surface morphological features of the material includes:

[0016] A 3D model corresponding to the target image is constructed using a 3D reconstruction algorithm. Based on the 3D model, the height value of each point on the material surface relative to the reference plane, the curvature value reflecting the surface unevenness, and the gradient vector representing the texture direction are calculated.

[0017] The set of height, curvature, and gradient vectors constitutes the surface morphology of a material.

[0018] As a further aspect of the present invention: extracting the changes in microscopic geometric structure includes:

[0019] By comparing the baseline surface morphology characteristics of the material to be cured in its initial state with the calculated surface morphology characteristics, the differences in surface undulations, crack propagation trends, and leveling states caused by the curing reaction can be identified.

[0020] The spatial distribution and evolution intensity of the surface undulation changes, crack propagation trends, or leveling state differences are quantified by calculating the surface curvature distribution gradient and the local deformation vector field.

[0021] The surface curvature distribution gradient and the local deformation vector field together constitute the changes in the microscopic geometry.

[0022] As a further aspect of the present invention: calculating the radiation energy distribution map includes:

[0023] A mapping relationship model is established. The input of the mapping relationship model is the change in micro-geometric structure, and the output is the UV radiation energy compensation value and its spatial projection position. The energy compensation values ​​and their spatial projection positions at all positions are integrated to generate the radiation energy distribution map.

[0024] In the process of calculation, the mapping relationship model calculates the actual UV energy absorption efficiency based on the pre-stored reflection, scattering and absorption characteristics of the material surface to UV energy.

[0025] By comparing the actual UV energy absorption efficiency with the energy absorption efficiency of the ideal curing process, the UV energy value and energy projection position required to maintain uniform curing are derived.

[0026] As a further aspect of the present invention: the generation of light source control instructions includes:

[0027] The radiation energy distribution map is analyzed into intensity control coefficients and angle deflection parameters corresponding to each light-emitting unit in the modulated UV light source array;

[0028] The driving current of the light-emitting unit is adjusted according to the intensity control coefficient, and the orientation of the optical lens of the light-emitting unit is adjusted by controlling the micro motor according to the angle deflection parameter.

[0029] As a further aspect of the present invention: the modulated UV light source array is composed of multiple independently controlled ultraviolet light-emitting units, each of which is equipped with a lens group driven by a micro motor and a current-adjustable drive circuit.

[0030] According to the light source control command, the driving current of all or part of the light-emitting units and the deflection angle of the lens group are adjusted synchronously so that the UV energy is concentrated and projected onto the curing reaction area that needs compensation.

[0031] A visual intelligence-based dynamic compensation system for UV curing light sources, characterized in that it comprises:

[0032] Acquisition module: Set up a modulated UV light source array for curing materials, and acquire real-time images of the material surface during the curing process of the modulated UV light source array;

[0033] Analysis module: preprocesses the real-time image to obtain the target image, obtains the surface morphology features of the material based on the target image, and extracts the micro-geometric structure changes of the curing reaction area based on the surface morphology features of the material.

[0034] Generation module: Calculates the radiation energy distribution map for compensating the UV light source based on the changes in the micro-geometric structure, and generates light source control commands based on the radiation energy distribution map;

[0035] Control module: Sends the light source control command to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array.

[0036] The beneficial effects of this invention compared to the prior art are as follows:

[0037] 1) By acquiring and preprocessing real-time images of the material surface, this invention can effectively eliminate the influence of uneven ambient lighting or imaging interference, ensuring the accuracy of subsequent feature extraction; the processed images can clearly present the fine texture, undulation and contour features of the material surface, thereby ensuring the accuracy of morphological feature recognition and three-dimensional reconstruction.

[0038] 2) By establishing the correlation between the surface morphology characteristics of the material and the changes in the micro-geometric structure of the curing reaction region, this invention can identify and quantify the surface undulations, crack propagation trends, and leveling differences of the material during the curing process in real time. With the accurate capture of these dynamic characteristics, this invention can fully reflect the nonlinear evolution characteristics of the material surface caused by the curing reaction and convert them into quantitative parameters that can be used for energy distribution calculation. Thus, this invention can not only identify the development trend of surface defects, but also effectively characterize the dynamic evolution of the material surface during the curing process, thereby ensuring that the compensation decision is highly matched with the actual reaction state of the material.

[0039] 3) This invention maps changes in microscopic geometry into a radiation energy distribution map and further generates light source control commands, enabling dynamic adjustment of the modulated UV light source array unit by unit. This process allows the output intensity and irradiation angle of the light source to be finely adjusted according to the needs of different areas on the material surface, thereby ensuring that the spatial distribution of energy projection is completely consistent with the compensation target. With the help of this dynamic adjustment mechanism, this invention can maintain the uniformity and consistency of the curing process as a whole, avoid the occurrence of local over-curing or under-curing, and finally obtain a curing effect with high surface flatness and good structural integrity. Attached Figure Description

[0040] The invention will now be further described with reference to the accompanying drawings.

[0041] Figure 1 This is a flowchart illustrating a dynamic compensation method for UV curing light sources based on visual intelligence, according to the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figure 1 As shown, this invention is a dynamic compensation method for UV curing light sources based on visual intelligence, comprising the following steps:

[0044] An adjustable ultraviolet light source array is set up. This light source array serves as an energy input device for curing materials. During operation, the light source array can continuously output ultraviolet radiation to meet the energy requirements of the material in the curing reaction.

[0045] While the ultraviolet light source array is curing the material, an image acquisition device is set up to observe the surface state of the material in real time. The image acquisition process can be carried out simultaneously with the light source irradiation and the material curing reaction to ensure that the state of the material surface at different stages of curing can be completely recorded.

[0046] The real-time image is preprocessed to obtain the target image, the surface morphology features of the material are obtained based on the target image, and the micro-geometric structure changes of the curing reaction area are extracted based on the surface morphology features of the material.

[0047] In a preferred embodiment of the present invention, the preprocessing process first performs illumination uniformity correction and feature enhancement on the acquired real-time image. The illumination uniformity correction step is as follows: by analyzing the brightness distribution in the real-time image, local overexposed areas and underexposed areas caused by external ambient light or ultraviolet reflected light are identified. For these areas, the gray values ​​of the pixels are reconstructed using a pre-established image gray-level equalization model. This model calculates the distribution of the gray-level histogram of the entire image and equalizes the pixel frequency in different intervals to make the gray-level distribution more balanced, compressing the brightness of overexposed areas and increasing the brightness of underexposed areas, so that the overall gray values ​​are more uniformly distributed in the spatial range.

[0048] After completing the illumination uniformity correction, feature enhancement processing is performed. This process uses an edge-preserving filter, which weights each pixel's neighborhood during smoothing. The weight calculation considers both the spatial distance between pixels and the grayscale differences between neighboring pixels. When the grayscale difference is small, the pixels are easier to smooth, while the boundaries are kept unblurred where the grayscale difference is large. Therefore, it can remove local noise while preserving the edge structure, making the contours of the material surface texture more prominent. In this process, high-frequency noise is judged as a discontinuous feature due to the large grayscale variation between adjacent pixels. Therefore, it receives a lower weight in the weighting calculation and is gradually weakened, thereby effectively suppressing noise interference in the acquired image. This results in a final image with clear texture contours while avoiding excessive random interference signals.

[0049] In a preferred embodiment, obtaining the surface morphological features of the material includes:

[0050] The target image is transformed into a three-dimensional model corresponding to the material surface by using a three-dimensional reconstruction algorithm. The principle of three-dimensional reconstruction is to match and fuse images from different angles or positions to deduce the depth information of each point in space, thereby forming a three-dimensional data structure that can describe the surface morphology.

[0051] On this 3D model, the height value of each surface point relative to the reference plane can be calculated. The method is to establish a fixed reference plane as a reference, extract the vertical component of the 3D coordinates of any point in the model, and compare the difference between the component and the position of the reference plane. The difference is the height value of the point.

[0052] The curvature value is calculated based on the local curvature of the surface in three-dimensional space. Specifically, a point is selected and a set of adjacent points in a predetermined neighborhood are determined. The local surface is fitted by the three-dimensional coordinate relationship of these points, and then the curvature of the principal direction at that point is solved on the surface. During this process, the spatial distance and relative position between adjacent points are collected. The normal vector field of the local surface is derived from this information. Then, the rate of change of the normal vector in the neighborhood is examined. If the normal vector changes rapidly, it indicates that the local curvature of the surface is significant and the curvature at that point is large. If the normal vector changes slowly, the surface is relatively flat.

[0053] The gradient vector is calculated based on the spatial distribution of surface height values. The height of a point is compared with the height of its surrounding points to obtain the rate of change of height in each direction. The rate of change is then combined into a vector form, with its direction pointing to the direction of the greatest height change and its magnitude representing the magnitude of the height change. This reflects the extension direction and strength of the surface texture.

[0054] The height value, curvature value, and gradient vector are combined to form a complete description of the surface morphology of the material.

[0055] It should be noted that the extraction of changes in microscopic geometry includes:

[0056] The process of extracting changes in microscopic geometric structure first requires establishing a basis for comparison.

[0057] Before a material is cured, it has relatively stable surface morphology characteristics. These characteristics can usually be obtained through optical measurement, three-dimensional reconstruction or laser scanning to form surface data under a reference state. After the curing reaction proceeds, the material surface will undergo certain changes, such as local height fluctuations, the generation and propagation of microcracks and differences in surface flatness. By comparing the surface morphology characteristics collected after curing with the initial reference state, the geometric morphological differences caused during the curing process can be identified.

[0058] Curvature itself reflects the magnitude and direction of surface bending. By calculating the distribution of curvature values ​​in space, the overall undulation pattern of the material surface can be observed. Furthermore, if the spatial trend of curvature is extracted, a curvature distribution gradient is formed. Calculating this gradient usually requires first fitting or interpolating the surface height field to obtain a continuous and smooth surface representation. Then, the curvature is calculated on this surface, and its magnitude and direction of change are compared between different locations. In this way, it is possible to reveal which areas have drastic surface undulations and which areas are relatively flat, thereby capturing the macroscopic and microscopic differences caused by the curing reaction.

[0059] Local deformation vector field refers to mapping the initial surface morphology to the cured surface morphology point by point, and determining the displacement direction and amount of each surface point through matching or registration methods. Commonly used methods include digital image correlation methods, optical interferometry, or registration calculation based on 3D point clouds. The basic idea of ​​these methods is to find the correspondence between the reference surface and the target surface, and to calculate the deformation at each position by tracking local textures, surface feature points, or using the displacement of interference fringes. The final result can be represented as a vector distribution map of the entire field, where each vector represents the offset and deformation experienced by that point during the curing process.

[0060] By combining the surface curvature distribution gradient with the local deformation vector field, the geometric changes of the material surface can be comprehensively characterized from two aspects. The curvature distribution gradient reveals the changing trends of the overall undulation and detailed texture, while the local deformation vector field provides information on the motion and deformation of each point. Through this complementary approach, the spatial distribution characteristics and evolution intensity of the surface during the curing process can be quantitatively described. The combination of the two constitutes a complete quantity of microscopic geometric structure changes. This quantity of changes is not only a numerical representation, but also a comprehensive description that unifies macroscopic surface features and microscopic local changes under the same framework.

[0061] The radiation energy distribution map for compensating the UV light source is calculated based on the changes in the micro-geometric structure, and a light source control command is generated based on the radiation energy distribution map.

[0062] It should be noted that the calculation of the radiation energy distribution map includes:

[0063] When establishing the mapping relationship model, it is first necessary to clarify the scope and objective of the model, namely, to dynamically compensate the energy distribution of the ultraviolet light source according to the changes in the material surface during the curing process. The input of the model is the change in micro-geometric structure obtained from the previous steps. This change has comprehensively characterized the changes in surface undulation, crack propagation trend, and leveling state caused by the curing reaction. By inputting these geometric features into the mapping relationship model, the model can output the corresponding ultraviolet radiation energy compensation value and the spatial projection position of the compensation effect.

[0064] The energy compensation values ​​and their coordinate information at all locations are integrated to generate a radiation energy distribution map. This distribution map not only reflects the magnitude of the compensation amount but also indicates the specific areas where the light source needs adjustment, providing precise guidance for the intensity control and angle adjustment of the light source array, thereby ensuring the overall uniformity and stability of the curing process;

[0065] Understandably, the model's calculations rely not only on the input geometric changes but also on externally stored material optical property data. This data includes the material surface's reflection, scattering, and absorption characteristics of ultraviolet energy, pre-stored during initialization through experimental calibration or database methods. Reflection characteristics describe the proportion of energy reflected by the material surface under ultraviolet irradiation; scattering characteristics reflect the degree of angular diffusion of incident energy within the surface and internal structure; and absorption characteristics determine the effective proportion of energy actually absorbed within the material and converted into driving forces for chemical reactions. By utilizing these optical properties and combining them with the input microscopic geometric changes, the model can accurately assess the actual absorption efficiency of ultraviolet energy under current conditions during calculations.

[0066] After obtaining the actual absorption efficiency, the model also needs to be compared with the target energy absorption efficiency corresponding to the ideal curing process; the energy absorption efficiency of the ideal curing process refers to the energy utilization required to achieve uniform curing of the material surface under theoretical conditions, which is usually set in advance by the process designer based on experience, experimental or simulation results.

[0067] The comparison results can reveal the gap between the current actual curing process and the ideal curing state. When the actual absorption efficiency is lower than the target value, it indicates that there are areas of insufficient energy on the material surface, and compensation energy needs to be added at the corresponding locations. When the difference between the actual absorption efficiency and the target value is small, only a small correction is needed. Through this comparison, the model can derive the amount of ultraviolet energy required to maintain uniform curing and its spatial projection location.

[0068] Understandably, the process of calculating the current absorption efficiency involves multiple steps. First, the curvature and micro-uneven state of the surface are determined based on the changes in the micro-geometric structure. These geometric features affect the local incident angle distribution of the incident light, thereby indirectly changing the intensity of reflection and scattering.

[0069] The model calls the reflection parameters in the material library, combines the surface angle distribution to estimate the reflected energy loss at different locations, and then uses the scattering parameters and surface texture features to calculate the energy diffusion of incident light in the surrounding area.

[0070] Finally, the absorption coefficient is applied to calculate the proportion of remaining energy deposited inside the material, thereby obtaining the actual energy absorption efficiency at each surface location.

[0071] In another preferred embodiment of the present invention, the generation of light source control instructions includes:

[0072] The radiation energy distribution map is analyzed point by point, and the energy compensation value and spatial coordinates in the distribution map are mapped to the light-emitting unit of the modulated ultraviolet light source array. Since each light-emitting unit has a fixed coverage area and corresponding projection direction in the array, the surface coordinates can be mapped one-to-one with the irradiation range of the unit.

[0073] During the analysis process, the amount of additional energy required for each surface point is first determined based on the compensation energy value of that point. Then, the energy value is converted into the intensity control coefficient of the corresponding light-emitting unit by combining the geometric relationship between that position and the light source unit. The intensity control coefficient is used to determine the adjustment range of the driving current. When the current increases, the radiation intensity of the light source increases, and when the current decreases, the radiation intensity decreases.

[0074] At the same time, it is also necessary to deduce the angle direction of energy projection from the relationship between the spatial coordinates and surface normal vectors in the distribution map. By comparing the angle between the projection position and the light source emission path, the range of angles that the lens needs to rotate can be determined. Then, the angle change is converted into the angle deflection parameter of the corresponding light-emitting unit. The angle deflection parameter controls the micro motor to drive the lens group to adjust the direction, so that the energy can accurately cover the target area. The layout map itself contains the energy demand and spatial coordinate information of each position, and the current and angle of the light-emitting unit are directly related to the energy projection distribution. The energy demand can be converted into the current adjustment amount, and the relationship between the projection position coordinates and the geometric layout of the light source can be converted into the angle deflection amount. Thus, the mapping from the distribution map to specific control commands is realized.

[0075] The light source control command is sent to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array;

[0076] It is important to note that the modulated ultraviolet light source array consists of multiple independently controllable light-emitting units. Each light-emitting unit is equipped with a lens group driven by a micro-motor and a current-adjustable drive circuit. According to the light source control command, all or some of the light-emitting units are synchronously adjusted. The drive current is adjusted according to the intensity control coefficient to change the intensity of the ultraviolet radiation output by the unit. The lens group rotates according to the angle deflection parameter to change the projection direction of the beam in space.

[0077] Multiple light-emitting units perform adjustment operations at the same time, and can selectively participate in the regulation according to the distribution of the compensation area, so that the output of different units can form a continuous coverage in the target area; through the dual coordinated adjustment of intensity and angle, the ultraviolet energy is concentrated on the area that needs to be compensated in the curing reaction, thereby achieving precise energy projection and dynamic compensation in the spatial range.

[0078] A visual intelligence-based dynamic compensation system for UV curing light sources includes:

[0079] Acquisition module: Set up a modulated UV light source array for curing materials, and acquire real-time images of the material surface during the curing process of the modulated UV light source array;

[0080] Analysis module: preprocesses the real-time image to obtain the target image, obtains the surface morphology features of the material based on the target image, and extracts the micro-geometric structure changes of the curing reaction area based on the surface morphology features of the material.

[0081] Generation module: Calculates the radiation energy distribution map for compensating the UV light source based on the changes in the micro-geometric structure, and generates light source control commands based on the radiation energy distribution map;

[0082] Control module: Sends the light source control command to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array.

[0083] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art based on the actual situation. Meanwhile, other thresholds used for judgment, including but not limited to faster and slower, can be set based on the experience of those skilled in the art and are not restricted here.

[0084] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A dynamic compensation method for UV curing light sources based on visual intelligence, characterized in that, Includes the following steps: A modulated UV light source array is set up to cure the material, and real-time images of the material surface are acquired during the curing process of the modulated UV light source array. The real-time image is preprocessed to obtain the target image, the surface morphology features of the material are obtained based on the target image, and the micro-geometric structure changes of the curing reaction area are extracted based on the surface morphology features of the material. The radiation energy distribution map for compensating the UV light source is calculated based on the changes in the micro-geometric structure, and a light source control command is generated based on the radiation energy distribution map. The light source control command is sent to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array; The calculation of the radiation energy distribution map includes: A mapping relationship model is established. The input of the mapping relationship model is the change in micro-geometric structure, and the output is the UV radiation energy compensation value and its spatial projection position. The energy compensation values ​​and their spatial projection positions at all positions are integrated to generate the radiation energy distribution map. In the process of calculation, the mapping relationship model calculates the actual UV energy absorption efficiency based on the pre-stored reflection, scattering and absorption characteristics of the material surface to UV energy. By comparing the actual UV energy absorption efficiency with the energy absorption efficiency of the ideal curing process, the UV energy value and energy projection position required to maintain uniform curing are derived. The generated light source control instructions include: The radiation energy distribution map is analyzed into intensity control coefficients and angle deflection parameters corresponding to each light-emitting unit in the modulated UV light source array; The driving current of the light-emitting unit is adjusted according to the intensity control coefficient, and the orientation of the optical lens of the light-emitting unit is adjusted by controlling the micro motor according to the angle deflection parameter.

2. The method for dynamic compensation of UV curing light source based on visual intelligence according to claim 1, characterized in that, The preprocessing includes performing illumination uniformity correction and feature enhancement on the acquired real-time images; The illumination uniformity correction includes: Identify overexposed and underexposed areas in real-time images caused by ambient light or UV reflected light, and reconstruct pixel gray values ​​for the overexposed and underexposed areas using a pre-established image gray-scale equalization model. The feature enhancement includes: using an edge-preserving filter to enhance the contour information of the micro-texture of the material surface, while suppressing high-frequency noise introduced during image acquisition.

3. The method for dynamic compensation of UV curing light source based on visual intelligence according to claim 2, characterized in that, Obtaining material surface morphological features includes: A 3D model corresponding to the target image is constructed using a 3D reconstruction algorithm. Based on the 3D model, the height value of each point on the material surface relative to the reference plane, the curvature value reflecting the surface unevenness, and the gradient vector representing the texture direction are calculated. The set of height, curvature, and gradient vectors constitutes the surface morphology of a material.

4. The method for dynamic compensation of UV curing light source based on visual intelligence according to claim 3, characterized in that, Extracting changes in microscopic geometry includes: By comparing the baseline surface morphology characteristics of the material to be cured in its initial state with the calculated surface morphology characteristics, the differences in surface undulations, crack propagation trends, and leveling states caused by the curing reaction can be identified. The spatial distribution and evolution intensity of the surface undulation changes, crack propagation trends, or leveling state differences are quantified by calculating the surface curvature distribution gradient and the local deformation vector field. The surface curvature distribution gradient and the local deformation vector field together constitute the changes in the microscopic geometry.

5. The method for dynamic compensation of UV curing light source based on visual intelligence according to claim 1, characterized in that, The modulated UV light source array consists of multiple independently controlled ultraviolet light-emitting units, each equipped with a lens group driven by a micro motor and a current-adjustable drive circuit. According to the light source control command, the driving current of all or part of the light-emitting units and the deflection angle of the lens group are adjusted synchronously so that the UV energy is concentrated and projected onto the curing reaction area that needs compensation.

6. A dynamic compensation system for UV curing light sources based on visual intelligence, characterized in that, include: Acquisition module: Set up a modulated UV light source array for curing materials, and acquire real-time images of the material surface during the curing process of the modulated UV light source array; Analysis module: preprocesses the real-time image to obtain the target image, obtains the surface morphology features of the material based on the target image, and extracts the micro-geometric structure changes of the curing reaction area based on the surface morphology features of the material. Generation module: Calculates the radiation energy distribution map for compensating the UV light source based on the changes in the micro-geometric structure, and generates light source control commands based on the radiation energy distribution map; Control module: Sends the light source control command to the modulated UV light source array to adjust the output intensity and irradiation angle of each light-emitting unit in the modulated UV light source array; The calculation of the radiation energy distribution map includes: A mapping relationship model is established. The input of the mapping relationship model is the change in micro-geometric structure, and the output is the UV radiation energy compensation value and its spatial projection position. The energy compensation values ​​and their spatial projection positions at all positions are integrated to generate the radiation energy distribution map. In the process of calculation, the mapping relationship model calculates the actual UV energy absorption efficiency based on the pre-stored reflection, scattering and absorption characteristics of the material surface to UV energy. By comparing the actual UV energy absorption efficiency with the energy absorption efficiency of the ideal curing process, the UV energy value and energy projection position required to maintain uniform curing are derived. The generated light source control instructions include: The radiation energy distribution map is analyzed into intensity control coefficients and angle deflection parameters corresponding to each light-emitting unit in the modulated UV light source array; The driving current of the light-emitting unit is adjusted according to the intensity control coefficient, and the orientation of the optical lens of the light-emitting unit is adjusted by controlling the micro motor according to the angle deflection parameter.

Images