Adaptive dimming method and system based on surface detection

Abstract

The present application provides an adaptive dimming method and system based on surface detection, which relates to the field of data processing technology. The method includes: dividing the defect detection surface through laser scanning modeling and grid curvature analysis, and then obtaining an adaptive dimming strategy through dimming trajectory analysis and testing. Finally, the target adaptive dimming strategy is generated through the shortest trajectory connection fitting. The present application can solve the technical problem that the existing technology of curved surface component detection based on photoelectric detection has inaccurate scanning of curved surface defects, resulting in some defects being hidden, further affecting the accuracy of photoelectric detection and recognition. The surface of the object is scanned by a photoelectric sensor, a three-dimensional model is established, and a curvature change analysis is performed on multiple grid three-dimensional models. Based on the analysis results, the detection surface of the component three-dimensional model is divided, which can focus on the area with significant curvature change, realize accurate and efficient component defect detection, and photoelectric detection greatly improves the detection accuracy.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an adaptive dimming method and system based on surface detection. Background Art

[0002] Photoelectric detection technology is used to detect the surface of an object. During the detection process, due to the different relative positions of the object to be measured with respect to the photoelectric detection equipment, the light reflection characteristics of the same surface defects are different, thus affecting the accuracy of the measurement results.

[0003] In summary, the existing technology for detecting curved surface parts based on photoelectric detection has the technical problem of inaccurate scanning of curved surface defects, which causes some defects to be hidden, further affecting the accuracy of photoelectric detection and recognition. Summary of the invention

[0004] The purpose of this application is to provide an adaptive dimming method and system based on surface detection, so as to solve the technical problem that the existing technology of curved surface component detection based on photoelectric detection has inaccurate scanning of curved surface defects, resulting in some defects being hidden, further affecting the accuracy of photoelectric detection and recognition.

[0005] In view of the above problems, the present application provides an adaptive dimming method and system based on surface detection.

[0006] In a first aspect, the present application provides an adaptive dimming method based on surface detection, which is implemented by an adaptive dimming system based on surface detection, wherein the method comprises: performing laser scanning on a component to be tested to obtain component point cloud data, and performing surface modeling based on the component point cloud data to obtain a component three-dimensional model; presetting a first grid density, and using the first grid density as a constraint, gridding the component three-dimensional model to obtain a plurality of grid three-dimensional models; performing curvature change analysis on the plurality of grid three-dimensional models, and dividing the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; pre-building a dimming system; and The method comprises the following steps: a trajectory analysis channel is provided, and the K defect detection surfaces are synchronized to the dimming trajectory analysis channel for dimming trajectory analysis to obtain K dimming trajectory parameters; a standard defect component of the component to be tested is interactively obtained, and a dimming test is performed on the standard defect component to obtain multiple dimming luminance parameters of various component apparent defects; the multiple dimming luminance parameters are fitted to the K dimming trajectory parameters to obtain K adaptive dimming strategies; the K dimming trajectory parameters are fitted with the shortest trajectory according to the component three-dimensional model to obtain a target dimming trajectory, and the K adaptive dimming strategies are connected according to the target dimming trajectory to generate a target adaptive dimming strategy.

[0007] In a second aspect, the present application further provides an adaptive dimming system based on surface detection, which is used to execute the adaptive dimming method based on surface detection as described in the first aspect, wherein the system includes: a three-dimensional model acquisition module, the three-dimensional model acquisition module is used to perform laser scanning on the component to be tested to obtain component point cloud data, and perform surface modeling based on the component point cloud data to obtain a component three-dimensional model; a grid processing module, the grid processing module is used to preset a first grid density, and use the first grid density as a constraint to perform grid processing on the component three-dimensional model to obtain multiple grid three-dimensional models; a defect detection surface acquisition module, the defect detection surface acquisition module is used to perform curvature change analysis on the multiple grid three-dimensional models, and divide the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; a dimming trajectory parameter acquisition module, the dimming trajectory parameter The data acquisition module is used to pre-build a dimming trajectory analysis channel, and synchronize the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters; the dimming test module is used to interactively obtain the standard defective component of the component to be tested, and perform dimming test on the standard defective component to obtain multiple dimming luminance parameters of various component apparent defects; the adaptive dimming strategy acquisition module is used to fit the multiple dimming luminance parameters to the K dimming trajectory parameters to obtain K adaptive dimming strategies; the target dimming trajectory acquisition module is used to perform the shortest trajectory connection fitting on the K dimming trajectory parameters according to the component three-dimensional model to obtain a target dimming trajectory, and connect the K adaptive dimming strategies according to the target dimming trajectory to generate a target adaptive dimming strategy.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The component point cloud data is obtained by laser scanning the component to be tested, and surface modeling is performed based on the component point cloud data to obtain a component three-dimensional model; a first grid density is preset, and the component three-dimensional model is gridded with the first grid density as a constraint to obtain multiple grid three-dimensional models; a curvature change analysis is performed on the multiple grid three-dimensional models, and the component three-dimensional model is divided into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; a dimming trajectory analysis channel is pre-constructed, and the K defect detection surfaces are synchronized to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters; a standard defective component of the component to be tested is interactively obtained, and a dimming test is performed on the standard defective component to obtain multiple dimming photometric parameters of various component surface defects; the multiple dimming photometric parameters are synchronized to the dimming trajectory analysis channel to obtain K dimming trajectory parameters. The method is characterized in that the method comprises the following steps: fitting the K dimming trajectory parameters to the K dimming trajectory parameters in the shortest trajectory according to the component three-dimensional model to obtain K adaptive dimming strategies; and obtaining a target dimming trajectory by connecting the K adaptive dimming strategies according to the target dimming trajectory. The method solves the technical problem that the prior art performs curved component detection based on photoelectric detection, and the curved surface defects are inaccurately scanned, resulting in some defects being hidden, and further affecting the accuracy of photoelectric detection and recognition. The method uses a photoelectric sensor to scan the surface of an object, establishes a three-dimensional model, performs curvature change analysis on multiple grid three-dimensional models, and divides the component three-dimensional model into detection surfaces based on the analysis results. The method can focus on areas with significant curvature changes, thereby achieving accurate and efficient component defect detection. Photoelectric detection greatly improves the detection accuracy.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A schematic diagram of the process of the adaptive dimming method based on surface detection in this application;

[0013] Figure 2 This is a schematic diagram of the structure of the adaptive dimming system based on surface detection in this application.

[0014] Description of reference numerals:

[0015] A three-dimensional model acquisition module 11, a grid processing module 12, a defect detection surface acquisition module 13, a dimming trajectory parameter acquisition module 14, a dimming test module 15, an adaptive dimming strategy acquisition module 16, and a target dimming trajectory acquisition module 17. DETAILED DESCRIPTION

[0016] The present application provides an adaptive dimming method and system based on surface detection, which solves the technical problem that the prior art performs curved surface component detection based on photoelectric detection, resulting in inaccurate scanning of curved surface defects, causing some defects to be hidden, and further affecting the accuracy of photoelectric detection and recognition. The application uses a photoelectric sensor to scan the surface of an object, establishes a three-dimensional model, performs curvature change analysis on multiple grid three-dimensional models, and divides the component three-dimensional model into detection surfaces based on the analysis results. It can focus on areas with significant curvature changes, thereby achieving accurate and efficient component defect detection. Photoelectric detection greatly improves the detection accuracy.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] Embodiment 1

[0019] Please see attached Figure 1 The present application provides an adaptive dimming method based on surface detection, wherein the method is applied to an adaptive dimming system based on surface detection, and the method specifically comprises the following steps:

[0020] Step 1: Perform laser scanning on the component to be tested to obtain component point cloud data, and perform surface modeling based on the component point cloud data to obtain a component three-dimensional model;

[0021] Specifically, according to the size, accuracy requirements and scanning environment of the part to be measured, select a suitable model of laser scanner. Ensure that the light in the scanning site is stable and avoid strong direct light; clean the site and remove obstacles that may block or reflect the laser. Install the software supporting the laser scanner on the computer and make necessary configurations and settings. Connect the laser scanner to the computer via a data cable or wireless network. Set parameters such as scanning density and range in the software to ensure that sufficient details of the surface of the part can be captured. Aim the laser scanner at the part to be measured and start the scanning program. During the scanning process, the position and direction of the scanner can be adjusted as needed to ensure that all parts of the part are evenly covered. After the scan is completed, save the captured point cloud data to the computer. Use a special point cloud processing software to open the saved point cloud data file. Use filtering algorithms to remove noise and outliers in the point cloud data to improve data quality. If multiple scans are performed, the point cloud data under different scanning perspectives need to be registered to ensure that they are in the same coordinate system. Under the premise of maintaining geometric features, simplify the point cloud data to reduce the data volume and improve processing efficiency. According to the characteristics of the point cloud data and modeling requirements, select a suitable surface modeling method. You can use reconstruction based on triangular meshes, implicit surface reconstruction, etc. Before modeling, set some parameters, such as mesh resolution, smoothness, etc., to control the accuracy and detail of the model. Input the processed point cloud data into the modeling algorithm to generate a three-dimensional model of the component.

[0022] Step 2: presetting a first grid density, and using the first grid density as a constraint, performing gridding processing on the component three-dimensional model to obtain a plurality of grid three-dimensional models;

[0023] Specifically, determine the appropriate mesh density based on application requirements, such as accuracy requirements, computing resource limitations, etc. Mesh density is usually expressed as the number of meshes per unit area or unit volume, or as the average distance between meshes. In the modeling software processing tool, find the mesh density setting option. Enter the preset first mesh density value to ensure that the value meets the application requirements. Select a suitable meshing algorithm based on the geometric characteristics and complexity of the component three-dimensional model. Including algorithms based on Delaunay triangulation, algorithms based on Voronoi diagrams, etc. During the meshing process, ensure that the generated mesh meets the preset first mesh density. Adjust the algorithm parameters or iterate multiple times to obtain a mesh that meets the density requirements. Input the component three-dimensional model into the selected meshing algorithm. Based on the output of the algorithm, obtain multiple mesh three-dimensional models constrained by the first mesh density.

[0024] Step 3: performing curvature change analysis on the multiple grid three-dimensional models, and dividing the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces;

[0025] Specifically, for each mesh solid model, all its mesh vertices are traversed. The curvature calculation formula, such as Gaussian curvature, mean curvature, etc., is used to calculate the curvature value of each vertex. These curvature values ​​reflect the degree of curvature of the model surface at that point. The curvature values ​​between adjacent vertices are compared to calculate the curvature change. Identify areas with significant curvature changes, which may be defects, edges or feature points on the surface of the component. A curvature change threshold can be set to screen out potential defect areas. According to the results of the curvature change analysis, determine the appropriate detection surface division strategy. You can choose to use methods such as division based on curvature threshold and division based on region growth. Use the selected division strategy to divide the surface of the component solid model into multiple detection surfaces. Each detection surface should contain mesh vertices with similar curvature characteristics. Ensure that the divided detection surface can cover the entire surface of the component without omission. Optimize and adjust the initially divided detection surface to remove small or irregularly shaped patches. Adjacent small patches can be merged, or complex-shaped areas can be re-divided. Set the criteria for identifying defect detection surfaces. These criteria can be based on parameters such as curvature value, curvature variation, shape complexity, etc. Apply the defect judgment criteria to filter out defect detection surfaces that meet the conditions from the divided detection surfaces. One or more thresholds can be set to determine whether the detection surface is a defect surface. Mark and save the K defect detection surfaces that are filtered out.

[0026] Step 4: pre-construct a dimming trajectory analysis channel, and synchronize the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters;

[0027] Specifically, according to the characteristics of the defect detection surface and the purpose of dimming analysis, the functions and performance requirements required by the analysis channel are clarified. A channel architecture is constructed that can receive defect detection surface data, perform dimming trajectory analysis and output results. The architecture may include components such as data input interface, preprocessing module, analysis algorithm library, result output interface, etc. The constructed analysis channel is fully tested and verified to ensure that it meets the design requirements and performance indicators. If the data format of the defect detection surface is incompatible with the analysis channel, data format conversion is required. The converted defect detection surface data is transmitted to the dimming trajectory analysis channel through the data input interface. Ensure the real-time and stability of data transmission to avoid data loss or confusion. Verify the received defect detection surface data to ensure the correctness and validity of the data. Perform necessary preprocessing operations such as denoising and smoothing as needed to improve the accuracy of the analysis. According to the characteristics of the defect detection surface and the analysis requirements, select a suitable dimming trajectory analysis algorithm. Time series-based analysis can be used. Input the preprocessed defect detection surface data into the selected analysis algorithm. Run the algorithm to perform dimming trajectory analysis on each defect detection surface. K parameters related to the dimming trajectory are extracted from the analysis results, such as brightness change rate, lighting uniformity, etc.

[0028] Step 5: interactively obtain a standard defective component of the component to be tested, and perform a dimming test on the standard defective component to obtain multiple dimming photometric parameters of various component apparent defects;

[0029] Specifically, it should be understood that different defects have different reflection characteristics for different lights, that is, different sensitivities. Based on this, this embodiment uses photoelectric detection technology to realize the identification of component apparent defects. Based on this, this embodiment calls and obtains standard defective components of the same model as the component to be tested that have been produced historically. The standard defective components cover all types of apparent defects that may exist in the component to be tested. The apparent defects of the standard defective components may include cracks, depressions, protrusions, uneven colors, etc. Prepare a stable light source and lighting conditions to ensure the consistency of the test environment. Dimming test is a key link in photoelectric detection, including adjusting the intensity or wavelength of the detection light source to improve the detection sensitivity of different defects. By accurately adjusting the light source, the detection system can more keenly capture the subtle changes caused by the defect. For example, if a certain wavelength of light has a stronger reflection or absorption characteristic for the defect, then using this wavelength of light for detection will significantly increase the probability of finding the defect. If the light source intensity is too strong, the scattered light will be too strong, and it may even cover the scattered light of the defect, making the defect difficult to identify. On the contrary, if the light source intensity is too weak, the scattered light of the defect will also become weak, and the defect cannot be detected. A moderate light source intensity needs to be selected. Secondly, different defects may react differently to different wavelengths. For example, at certain wavelengths, defects may be more easily detected because at these wavelengths, the contrast between the defects and the surrounding materials is higher. Therefore, different light source intensities and wavelengths can be selected for testing. For each combination of light source intensity and wavelength, observe and record the detection effect of the defect. Analyze the test data to find the combination of light source intensity and wavelength that makes the defect most obvious and easiest to capture. Adjust the light source parameters according to the test results to obtain the best detection effect. Extract the dimming photometric parameters related to various defects. This embodiment adjusts the intensity or wavelength of the detection light source to improve the detection sensitivity of different defects, and obtains high detection sensitivity dimming photometric parameters for various types of defects.

[0030] Step 6: Fit the multiple dimming luminance parameters to the K dimming trajectory parameters to obtain K adaptive dimming strategies;

[0031] Specifically, the obtained multiple dimming photometric parameters are integrated with the K dimming trajectory parameters. Ensure that the two sets of parameters are consistent in data format and unit for subsequent processing. Clean the integrated data to remove invalid, duplicate or abnormal values. Filter data within a specific range as needed to improve the accuracy and efficiency of fitting. Normalize the data to eliminate the influence of dimension and order of magnitude. Match multiple dimming photometric parameters with K dimming parameter trajectories one by one to obtain corresponding fitting results. According to the fitting results, formulate corresponding adaptive dimming strategies for each dimming trajectory. The strategy may include adjusting the intensity or number of wavelengths of the light source according to the real-time trajectory.

[0032] Step 7: Perform shortest trajectory connection fitting on the K dimming trajectory parameters according to the component 3D model to obtain a target dimming trajectory, and connect the K adaptive dimming strategies according to the target dimming trajectory to generate a target adaptive dimming strategy.

[0033] Specifically, the characteristics of K dimming trajectory parameters are analyzed, including the starting point, end point, path length, curvature, etc. Determine which parameters are critical for trajectory connection, and consider how to optimize these parameters to achieve the shortest trajectory. Based on the component three-dimensional model, a mathematical model for trajectory connection is constructed. The model should consider constraints such as trajectory continuity, smoothness, and length minimization. Select a suitable optimization algorithm, such as dynamic programming, genetic algorithm, or path planning algorithm, to solve the shortest trajectory connection problem. Use the selected optimization algorithm to connect and fit the K dimming trajectory parameters. Through iterative calculation, continuously optimize the trajectory path until the shortest trajectory requirements are met. Verify the fitted target dimming trajectory to ensure that it meets the preset accuracy and feasibility requirements. If necessary, fine-tune the trajectory to further improve its quality and applicability. According to the characteristics of the target dimming trajectory, a strategy connection method is formulated. The target adaptive dimming strategy can be that the position of the photoelectric detection device remains unchanged, the position of the object to be tested changes, the relative position of the photoelectric detection device and the object to be tested changes, and the object to be tested is illuminated along the obtained path. Dimming of photoelectric detection equipment based on detection trajectory should ensure that the dimming strategy can smoothly transition when the trajectory switches to avoid abrupt changes in light source intensity or wavelength. Generate a target adaptive dimming strategy by combining the target dimming trajectory and K adaptive dimming strategies. The target strategy intelligently selects appropriate dimming parameters based on the real-time trajectory position and adjustment requirements. Test the effect of the target adaptive dimming strategy in a real environment. Collect test data and analyze the performance of the strategy and possible problems. Optimize the strategy based on the test results to improve its adaptability and robustness.

[0034] Furthermore, step three of this application also includes:

[0035] Extracting principal curvature parameters and normal vector parameters of the plurality of mesh three-dimensional models respectively to obtain a plurality of mesh principal curvatures and a plurality of mesh normal vectors;

[0036] The plurality of grid principal curvatures and the plurality of grid normal vectors constitute the curvature change analysis result;

[0037] The curvature change analysis results are compared with the principal curvatures of adjacent grids and the normal vectors of adjacent grids according to the connection relationship of the multiple grid three-dimensional models, and the multiple grid three-dimensional models are locally merged based on the comparison results to obtain the K defect detection surfaces.

[0038] Specifically, ensure that there are data for multiple mesh stereo models, which can be obtained through 3D scanning, modeling software, etc. Preprocess the mesh data, such as removing duplicate vertices, optimizing the mesh structure, etc. Calculate the principal curvature of each vertex of each mesh stereo model. The principal curvature can be obtained by analyzing the local surface shape around the vertex, usually using the method of fitting quadratic surfaces, such as quadratic equations. The extracted principal curvature parameters reflect the degree of curvature of the model at that point. Calculate the normal vector of each vertex or patch of each mesh stereo model. The normal vector can be obtained by analyzing the adjacent relationship of vertices or patches, usually using the vector cross product method. The extracted normal vector parameters reflect the orientation of the model at that point. Integrate the extracted multiple mesh principal curvatures and multiple mesh normal vectors together. These data together constitute the curvature change analysis results for subsequent comparison and merging operations. Analyze the connection relationship between multiple mesh stereo models to determine which meshes are adjacent. This can be achieved by checking the boundaries between meshes, shared vertices or patches, etc. Compare the principal curvatures and normal vectors of adjacent meshes. The purpose of the comparison is to find out the areas where the curvature and normal vector change greatly at the joints. These areas may be potential defects or parts that need to be merged. According to the comparison results, the criteria for local merging are formulated. For example, a threshold can be set. When the difference in the main curvature or normal vector of adjacent meshes exceeds this threshold, they are merged. According to the formulated criteria, the adjacent meshes that meet the conditions are locally merged. The merger can be achieved by adjusting the vertices, edges and patches of the mesh to ensure that the merged model remains continuous and smooth at the joints. The merged model is verified to ensure that it meets the preset geometric and topological constraints. If necessary, the merged results are fine-tuned to further improve the quality of the model. After the above steps, K defect detection surfaces that have been locally merged will be obtained. These detection surfaces are obtained based on the analysis of the changes in curvature and normal vectors, and may contain potential defect information.

[0039] Furthermore, the present application also includes:

[0040] Presetting a curvature acquisition interval, and acquiring spatial points of a first grid three-dimensional model based on the curvature acquisition interval to obtain a first spatial point set, wherein the first grid three-dimensional model is obtained by calling a model of the plurality of grid three-dimensional models;

[0041] Performing curvature analysis on the first spatial point set based on the first grid three-dimensional model to obtain a first principal curvature parameter set and a first normal vector set;

[0042] Serializing the first principal curvature parameter set and performing extreme value deviation calculation to obtain a first principal curvature deviation coefficient, and performing normal vector consistency calculation on the first normal vector set to obtain a first normal vector deviation coefficient;

[0043] Preset a principal curvature deviation threshold and a normal vector deviation threshold, determine whether the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold, and determine whether the first normal vector deviation coefficient satisfies the normal vector deviation threshold;

[0044] If the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold and the first normal vector deviation coefficient satisfies the normal vector deviation threshold, data selection is performed on the first principal curvature parameter set and the first normal vector set to obtain the first mesh principal curvature and the first mesh normal vector.

[0045] Specifically, according to the accuracy requirements of the model analysis, a curvature acquisition interval is pre-set, and this interval determines the density of point acquisition on the grid stereo model. The first grid stereo model is called, and the model is obtained by merging or selecting multiple grid stereo models. Based on the set curvature acquisition interval, spatial points are uniformly collected on the first grid stereo model to form a first spatial point set. Curvature analysis is performed on each point in the first spatial point set, and the corresponding principal curvature parameters are calculated to form a first principal curvature parameter set. Similarly, normal vector calculation is performed on each spatial point to obtain the normal vector of the point to form a first normal vector set. The first principal curvature parameter set is serialized, and the extreme value deviation of the first principal curvature parameter set is calculated, for example, the difference between the maximum and minimum values ​​or the standard deviation, etc., to obtain the first principal curvature deviation coefficient, which reflects the amplitude of the curvature change. The consistency rate of the normal vectors in the first normal vector set is calculated to check their consistency or direction changes. By calculating the angle or direction difference between adjacent normal vectors and counting the average value or maximum / minimum value of these differences, the first normal vector deviation coefficient is obtained, which reflects the degree of change of the normal vector. According to the actual application requirements and the tolerable curvature or normal vector change range, the main curvature deviation threshold and the normal vector deviation threshold are pre-set. Compare the first main curvature deviation coefficient with the preset main curvature deviation threshold. At the same time, compare the first normal vector deviation coefficient with the preset normal vector deviation threshold. If the first main curvature deviation coefficient satisfies, that is, is less than or equal to the main curvature deviation threshold, and the first normal vector deviation coefficient also satisfies the normal vector deviation threshold, it means that the curvature and normal vector changes of the first grid stereo model under the specified acquisition interval are relatively consistent, and the model surface is smooth and there is no significant geometric feature mutation. In this case, data election is performed on the first main curvature parameter set and the first normal vector set, and a representative main curvature and normal vector as the representative of the entire set can be selected by voting mechanism, average value calculation, median selection and other methods, which are respectively called the first grid main curvature and the first grid normal vector.

[0046] Furthermore, the present application also includes:

[0047] Performing data aggregation on a plurality of mesh principal curvatures and a plurality of mesh normal vectors of the plurality of mesh three-dimensional models to obtain M types of aggregated mesh principal curvatures-aggregated mesh normal vectors;

[0048] Referring to the M types of aggregated grid principal curvatures-aggregated grid normal vectors, the plurality of grid three-dimensional models are grouped to obtain M groups of grid three-dimensional models;

[0049] According to the connection relationship between the plurality of grid 3D models, locally merge adjacent grid 3D models in the M groups of grid 3D models to obtain M groups of locally merged models, where M is a positive integer;

[0050] Perform group label stripping on the M groups of local merged models to obtain K local merged models, where K is a positive integer greater than M;

[0051] The K local merged models are projected onto the component to be tested to obtain the K defect detection surfaces.

[0052] Specifically, a clustering algorithm such as K-means, DBSCAN, etc. or a statistical method such as the mean, median, etc. is used to aggregate the mesh principal curvature and mesh normal vector data. The purpose of aggregation is to combine the curvature and normal vector to form M aggregated mesh principal curvatures-aggregated mesh normal vectors. The M aggregated mesh principal curvatures-aggregated mesh normal vectors obtained by reference aggregation. Each aggregation result represents a specific geometric feature. According to the similarity of the aggregation results, multiple mesh stereo models are divided into M groups. Each group of mesh stereo models after grouping has similar curvature and normal vector features. The connection relationship between the mesh stereo models in each group is analyzed. Determine which meshes are adjacent, that is, share boundaries or vertices. Locally merge the adjacent mesh stereo models in each group. Boundary-based merging algorithms, vertex-based merging algorithms, etc. can be used. The grouping labels are stripped from the merged model to obtain a local merged model without labels. K local merged models without labels are projected onto the part to be tested. The projection can be achieved by spatial transformation, alignment algorithms, etc. The purpose of the projection is to match the local merged model with the geometric shape of the part to be tested. After projection, K local merged models and the corresponding areas of the tested parts form K defect detection surfaces.

[0053] Furthermore, the present application also includes:

[0054] If the first principal curvature deviation coefficient does not satisfy the principal curvature deviation threshold, or the first normal vector deviation coefficient does not satisfy the normal vector deviation threshold, calling a preset second mesh density, wherein the second mesh density is twice the first mesh density;

[0055] Taking the second grid density as a constraint, performing gridding processing on the first grid three-dimensional model to obtain a plurality of secondary grid three-dimensional models;

[0056] Performing curvature change analysis on the multiple secondary grid three-dimensional models to obtain secondary curvature change analysis results;

[0057] The first grid three-dimensional model is divided into detection surfaces based on the secondary curvature change analysis result to obtain H secondary defect detection surfaces, where H is a positive integer.

[0058] Specifically, since the analysis results under the first grid density do not meet the requirements, it means that the first grid stereo model is not a smooth surface. According to the preset rules, the second grid density is selected, which is twice the first grid density. The second grid density will provide higher resolution so as to more accurately capture the geometric features of the model. With the second grid density as a constraint, the first grid stereo model is prepared to be meshed. The first grid stereo model is refined to generate multiple secondary grid stereo models. These secondary grid stereo models have a higher grid density and can more finely represent the surface shape of the model. Collect relevant data of multiple secondary grid stereo models. Perform curvature change analysis on these secondary grid stereo models, including calculation of the main curvature and normal vector. By comparing the curvature and normal vector of adjacent grids, the area with significant curvature change is identified. Integrate the analysis data to form the secondary curvature change analysis result. This result contains more detailed curvature change information of the model surface. Based on the secondary curvature change analysis results, determine the basis for dividing the detection surface. Usually, the area with significant curvature change or inconsistent normal vector is selected as the division boundary. According to the division basis, the first grid stereo model is divided into the detection surface. Each divided detection surface corresponds to one or more secondary mesh three-dimensional models, and these detection surfaces can more accurately reflect the geometric feature changes of the model. After division, H secondary defect detection surfaces are obtained. These detection surfaces are identified at a higher mesh density and are therefore more likely to contain defect information of the model.

[0059] Furthermore, the present application also includes:

[0060] Interactively obtain multiple sample defect detection surfaces, multiple sample dimming trajectory parameters and multiple sample curvature identifiers of associated mapping;

[0061] The plurality of sample defect detection surfaces, the plurality of sample dimming trajectory parameters and the plurality of sample curvature identifiers are stored in association with each other to generate the dimming trajectory analysis channel;

[0062] Synchronizing the K principal curvatures of the K defect detection surfaces to the dimming trajectory analysis channel for curvature data comparison and corresponding sample dimming trajectory parameter call to obtain K standard dimming trajectory parameters;

[0063] Obtaining K model center points of the K local merged models by calling the component stereo model;

[0064] The K standard dimming trajectory parameters are optimized according to the K model center points to obtain the K dimming trajectory parameters.

[0065] Specifically, multiple sample defect detection surfaces are interactively obtained, and these detection surfaces are model surface areas known to contain defects. At the same time, sample dimming trajectory parameters corresponding to these sample defect detection surfaces are collected, that is, the optimal lighting adjustment parameters for highlighting defects under different lighting conditions. It is also necessary to obtain the sample curvature identification of each sample defect detection surface to identify the curvature characteristics of the detection surface. Ensure that a correct association mapping relationship is established between each sample defect detection surface, the corresponding sample dimming trajectory parameters and the sample curvature identification. The associated mapped multiple sample defect detection surfaces, multiple sample dimming trajectory parameters and multiple sample curvature identifications are associated and stored, which can be achieved using a database or a specific data structure. Based on the stored data, a dimming trajectory analysis channel is generated. This channel can receive new defect detection surface data and match the corresponding sample data according to its curvature characteristics, thereby outputting the corresponding dimming trajectory parameters. K main curvatures of K defect detection surfaces are synchronized to the dimming trajectory analysis channel. In the dimming trajectory analysis channel, the synchronized main curvature is compared with the stored sample curvature identification to find the most similar sample data. According to the comparison results, the sample dimming trajectory parameters associated with similar sample data are called as the initial value or reference of K standard dimming trajectory parameters. According to the three-dimensional model of the component, the K model center points of K local merged models are called and obtained. These center points are the position references of the local merged models in three-dimensional space. Combined with the position information of the K model center points, the K standard dimming trajectory parameters are optimized. The optimization process can consider factors such as the relative position of the illumination direction and the model surface, the distribution of illumination intensity, etc., to ensure that the illumination can best highlight the defect features. After optimization, K dimming trajectory parameters are obtained, which are the optimal dimming solutions for K defect detection surfaces.

[0066] Furthermore, step seven of this application also includes:

[0067] The K dimming trajectory parameters include K trajectory starting points and K trajectory end points;

[0068] Calling and obtaining a first dimming trajectory parameter from the K dimming trajectory parameters, wherein the first dimming trajectory parameter includes a first trajectory starting point and a first trajectory end point;

[0069] Calling the first trajectory end point and the K-1 trajectory starting points of the remaining K-1 dimming trajectory parameters based on the component stereo model, K-1 turning line distances;

[0070] Serialize the K-1 line-turning distances and perform minimum value call, and connect the dimming trajectory parameter corresponding to the minimum value with the first dimming trajectory parameter;

[0071] Similarly, the target dimming trajectory is obtained by performing shortest trajectory connection fitting on the K dimming trajectory parameters.

[0072] Specifically, ensure that there are K dimming trajectory parameters, each parameter includes a trajectory start point and a trajectory end point. Select one of the K dimming trajectory parameters as the first dimming trajectory parameter, and record the first trajectory start point and the first trajectory end point. Based on the component three-dimensional model, calculate the turn-line distance from the first trajectory end point to the K-1 trajectory start points of the remaining K-1 dimming trajectory parameters. The turn-line distance can be understood as the minimum distance or shortest path required to move from one trajectory end point to another trajectory start point. Sort the calculated K-1 turn-line distances so that the minimum value can be called later. Call the dimming trajectory parameter corresponding to the minimum value from the serialized turn-line distance. This dimming trajectory parameter will be selected to connect with the first dimming trajectory parameter because it provides the shortest path from the first trajectory end point to the next trajectory start point. Connect the trajectory end point of the first dimming trajectory parameter with the trajectory start point of the dimming trajectory parameter corresponding to the called minimum value. Update the end point of the currently connected trajectory to the new current trajectory end point. With the new current trajectory end point as the starting point, repeat the above steps of calculating the turn-line distance, serializing, calling the minimum value, and connecting the dimming trajectory parameters. Iterate this process until all K dimming trajectory parameters are connected to the target dimming trajectory. After the above iterative process, a shortest and continuous target dimming trajectory fitted by K dimming trajectory parameters will eventually be obtained. Based on the target dimming trajectory, the intensity or wavelength of the light source of the photoelectric detection device and the relative position of the surface of the component to be tested and the photoelectric detection device are adjusted. Photoelectric detection technology accurately identifies and distinguishes various defects on the surface of components based on differences in the reflection, absorption and scattering characteristics of light on the surface of the component. Once there are defects such as cracks, scratches, dents or color changes on the surface of the component, these defects will change the reflection properties of the light, thereby affecting the readings of the photoelectric detection system.

[0073] In summary, the surface detection-based adaptive dimming method provided by the present application has the following technical effects:

[0074] The component point cloud data is obtained by laser scanning the component to be tested, and surface modeling is performed based on the component point cloud data to obtain a component three-dimensional model; a first grid density is preset, and the component three-dimensional model is gridded with the first grid density as a constraint to obtain multiple grid three-dimensional models; a curvature change analysis is performed on the multiple grid three-dimensional models, and the component three-dimensional model is divided into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; a dimming trajectory analysis channel is pre-constructed, and the K defect detection surfaces are synchronized to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters; a standard defect component of the component to be tested is interactively obtained, and a dimming test is performed on the standard defect component to obtain multiple dimming photometric parameters of various component surface defects; the multiple dimming photometric parameters are fitted to the K dimming trajectory parameters to obtain K automatic Adaptive dimming strategy; according to the component three-dimensional model, the K dimming trajectory parameters are connected and fitted with the shortest trajectory to obtain the target dimming trajectory, and the K adaptive dimming strategies are connected according to the target dimming trajectory to generate a target adaptive dimming strategy. The existing technology does not accurately scan surface defects, resulting in some defects being hidden, further affecting the accuracy of surface recognition. The surface of the object is scanned by a photoelectric sensor, a three-dimensional model is established, and a curvature change analysis is performed on multiple grid three-dimensional models. The component three-dimensional model is divided into detection surfaces based on the analysis results, which can focus on areas with significant curvature changes, thereby achieving accurate and efficient component defect detection. Photoelectric detection greatly improves the detection accuracy and can realize non-contact measurement, thereby avoiding direct contact with the object being measured, reducing wear and interference, improving the accuracy and reliability of measurement, and being able to quickly respond to changes in optical signals.

[0075] Embodiment 2

[0076] Based on the same inventive concept as the adaptive dimming method based on surface detection in the aforementioned embodiment, the present application also provides an adaptive dimming system based on surface detection, see the attached Figure 2 , the system comprising:

[0077] A three-dimensional model acquisition module 11, which is used to perform laser scanning on the component to be measured to obtain component point cloud data, and perform surface modeling based on the component point cloud data to obtain a three-dimensional model of the component;

[0078] A grid processing module 12, the grid processing module 12 is used to preset a first grid density, and use the first grid density as a constraint to perform grid processing on the component three-dimensional model to obtain a plurality of grid three-dimensional models;

[0079] A defect detection surface acquisition module 13, the defect detection surface acquisition module 13 is used to perform curvature change analysis on the multiple grid three-dimensional models, and divide the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces;

[0080] A dimming trajectory parameter acquisition module 14, wherein the dimming trajectory parameter acquisition module 14 is used to pre-build a dimming trajectory analysis channel, and synchronize the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters;

[0081] A dimming test module 15, the dimming test module 15 is used to interactively obtain a standard defective component of the component to be tested, and perform a dimming test on the standard defective component to obtain multiple dimming photometric parameters of various component apparent defects;

[0082] An adaptive dimming strategy acquisition module 16, wherein the adaptive dimming strategy acquisition module 16 is used to fit the multiple dimming luminance parameters to the K dimming trajectory parameters to obtain K adaptive dimming strategies;

[0083] The target dimming trajectory acquisition module 17 is used to perform shortest trajectory connection fitting on the K dimming trajectory parameters according to the component three-dimensional model to obtain a target dimming trajectory, and connect the K adaptive dimming strategies according to the target dimming trajectory to generate a target adaptive dimming strategy.

[0084] Furthermore, the defect detection surface acquisition module 13 in the system is also used for:

[0085] Extracting principal curvature parameters and normal vector parameters of the plurality of mesh three-dimensional models respectively to obtain a plurality of mesh principal curvatures and a plurality of mesh normal vectors;

[0086] The plurality of grid principal curvatures and the plurality of grid normal vectors constitute the curvature change analysis result;

[0087] The curvature change analysis results are compared with the principal curvatures of adjacent grids and the normal vectors of adjacent grids according to the connection relationship of the multiple grid three-dimensional models, and the multiple grid three-dimensional models are locally merged based on the comparison results to obtain the K defect detection surfaces.

[0088] Furthermore, the system further comprises a first principal curvature and normal vector acquisition module, wherein the first principal curvature and normal vector acquisition module is used for:

[0089] Presetting a curvature acquisition interval, and acquiring spatial points of a first grid three-dimensional model based on the curvature acquisition interval to obtain a first spatial point set, wherein the first grid three-dimensional model is obtained by calling a model of the plurality of grid three-dimensional models;

[0090] Performing curvature analysis on the first spatial point set based on the first grid three-dimensional model to obtain a first principal curvature parameter set and a first normal vector set;

[0091] Serializing the first principal curvature parameter set and performing extreme value deviation calculation to obtain a first principal curvature deviation coefficient, and performing normal vector consistency calculation on the first normal vector set to obtain a first normal vector deviation coefficient;

[0092] Preset a principal curvature deviation threshold and a normal vector deviation threshold, determine whether the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold, and determine whether the first normal vector deviation coefficient satisfies the normal vector deviation threshold;

[0093] If the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold and the first normal vector deviation coefficient satisfies the normal vector deviation threshold, data selection is performed on the first principal curvature parameter set and the first normal vector set to obtain the first mesh principal curvature and the first mesh normal vector.

[0094] Furthermore, the system further comprises a defect detection surface acquisition module, wherein the defect detection surface acquisition module is used to:

[0095] Performing data aggregation on a plurality of mesh principal curvatures and a plurality of mesh normal vectors of the plurality of mesh three-dimensional models to obtain M types of aggregated mesh principal curvatures-aggregated mesh normal vectors;

[0096] Referring to the M types of aggregated grid principal curvatures-aggregated grid normal vectors, the plurality of grid three-dimensional models are grouped to obtain M groups of grid three-dimensional models;

[0097] According to the connection relationship between the plurality of grid 3D models, locally merge adjacent grid 3D models in the M groups of grid 3D models to obtain M groups of locally merged models, where M is a positive integer;

[0098] Perform group label stripping on the M groups of local merged models to obtain K local merged models, where K is a positive integer greater than M;

[0099] The K local merged models are projected onto the component to be tested to obtain the K defect detection surfaces.

[0100] Furthermore, the system further comprises a secondary defect detection surface acquisition module, and the secondary defect detection surface acquisition module is used to:

[0101] If the first principal curvature deviation coefficient does not satisfy the principal curvature deviation threshold, or the first normal vector deviation coefficient does not satisfy the normal vector deviation threshold, calling a preset second mesh density, wherein the second mesh density is twice the first mesh density;

[0102] Taking the second grid density as a constraint, performing gridding processing on the first grid three-dimensional model to obtain a plurality of secondary grid three-dimensional models;

[0103] Performing curvature change analysis on the multiple secondary grid three-dimensional models to obtain secondary curvature change analysis results;

[0104] The first grid three-dimensional model is divided into detection surfaces based on the secondary curvature change analysis result to obtain H secondary defect detection surfaces, where H is a positive integer.

[0105] Furthermore, the system further includes a dimming trajectory parameter acquisition module, and the dimming trajectory parameter acquisition module is used to:

[0106] Interactively obtain multiple sample defect detection surfaces, multiple sample dimming trajectory parameters and multiple sample curvature identifiers of associated mapping;

[0107] The plurality of sample defect detection surfaces, the plurality of sample dimming trajectory parameters and the plurality of sample curvature identifiers are stored in association with each other to generate the dimming trajectory analysis channel;

[0108] Synchronizing the K principal curvatures of the K defect detection surfaces to the dimming trajectory analysis channel for curvature data comparison and corresponding sample dimming trajectory parameter call to obtain K standard dimming trajectory parameters;

[0109] Obtaining K model center points of the K local merged models by calling the component stereo model;

[0110] The K standard dimming trajectory parameters are optimized according to the K model center points to obtain the K dimming trajectory parameters.

[0111] Furthermore, the adaptive dimming strategy acquisition module 16 in the system is also used for:

[0112] The K dimming trajectory parameters include K trajectory starting points and K trajectory end points;

[0113] Calling and obtaining a first dimming trajectory parameter from the K dimming trajectory parameters, wherein the first dimming trajectory parameter includes a first trajectory starting point and a first trajectory end point;

[0114] Calling the first trajectory end point and the K-1 trajectory starting points of the remaining K-1 dimming trajectory parameters based on the component stereo model, K-1 turning line distances;

[0115] Serialize the K-1 line-turning distances and perform minimum value call, and connect the dimming trajectory parameter corresponding to the minimum value with the first dimming trajectory parameter;

[0116] Similarly, the target dimming trajectory is obtained by performing shortest trajectory connection fitting on the K dimming trajectory parameters.

[0117] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The surface detection-based adaptive dimming method and specific examples in the first embodiment are also applicable to the surface detection-based adaptive dimming system in the present embodiment. Through the above detailed description of the surface detection-based adaptive dimming method, those skilled in the art can clearly understand the surface detection-based adaptive dimming system in the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method description.

[0118] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0119] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. The adaptive dimming method based on surface detection is characterized in that: The method comprises: Performing laser scanning on the component to be measured to obtain component point cloud data, and performing surface modeling based on the component point cloud data to obtain a component three-dimensional model; Preset a first grid density, and use the first grid density as a constraint to perform gridding processing on the component three-dimensional model to obtain a plurality of grid three-dimensional models; Performing curvature change analysis on the multiple grid three-dimensional models, and dividing the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; Pre-constructing a dimming trajectory analysis channel, and synchronizing the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis, and obtaining K dimming trajectory parameters; Interactively obtaining a standard defective component of the component to be tested, and performing a dimming test on the standard defective component to obtain a plurality of dimming photometric parameters of various component apparent defects; Fitting the multiple dimming luminance parameters to the K dimming trajectory parameters to obtain K adaptive dimming strategies; Performing shortest trajectory connection fitting on the K dimming trajectory parameters according to the component stereo model to obtain a target dimming trajectory, and connecting the K adaptive dimming strategies according to the target dimming trajectory to generate a target adaptive dimming strategy; Performing curvature change analysis on the multiple grid three-dimensional models, and dividing the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces, the method further comprising: Extracting principal curvature parameters and normal vector parameters of the plurality of mesh three-dimensional models respectively to obtain a plurality of mesh principal curvatures and a plurality of mesh normal vectors; The plurality of grid principal curvatures and the plurality of grid normal vectors constitute the curvature change analysis result; The curvature change analysis results are compared with the principal curvatures of adjacent grids and the normal vectors of adjacent grids according to the connection relationship of the multiple grid three-dimensional models, and the multiple grid three-dimensional models are locally merged based on the comparison results to obtain the K defect detection surfaces.

2. The method according to claim 1, characterized in that Extracting principal curvature parameters and normal vector parameters from the plurality of mesh three-dimensional models respectively to obtain a plurality of mesh principal curvatures and a plurality of mesh normal vectors, the method further comprising: Presetting a curvature acquisition interval, and acquiring spatial points of a first grid three-dimensional model based on the curvature acquisition interval to obtain a first spatial point set, wherein the first grid three-dimensional model is obtained by calling a model of the plurality of grid three-dimensional models; Performing curvature analysis on the first spatial point set based on the first grid three-dimensional model to obtain a first principal curvature parameter set and a first normal vector set; Serializing the first principal curvature parameter set and performing extreme value deviation calculation to obtain a first principal curvature deviation coefficient, and performing normal vector consistency calculation on the first normal vector set to obtain a first normal vector deviation coefficient; Preset a principal curvature deviation threshold and a normal vector deviation threshold, determine whether the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold, and determine whether the first normal vector deviation coefficient satisfies the normal vector deviation threshold; If the first principal curvature deviation coefficient satisfies the principal curvature deviation threshold and the first normal vector deviation coefficient satisfies the normal vector deviation threshold, data selection is performed on the first principal curvature parameter set and the first normal vector set to obtain the first mesh principal curvature and the first mesh normal vector.

3. The method according to claim 2, characterized in that Comparing the curvature change analysis results with the principal curvatures of adjacent grids and the normal vectors of adjacent grids according to the connection relationship of the multiple grid three-dimensional models, and locally merging the multiple grid three-dimensional models based on the comparison results to obtain the K defect detection surfaces, the method further comprising: Performing data aggregation on a plurality of mesh principal curvatures and a plurality of mesh normal vectors of the plurality of mesh three-dimensional models to obtain M types of aggregated mesh principal curvatures-aggregated mesh normal vectors; Referring to the M types of aggregated grid principal curvatures-aggregated grid normal vectors, the plurality of grid three-dimensional models are grouped to obtain M groups of grid three-dimensional models; According to the connection relationship between the plurality of grid 3D models, locally merge adjacent grid 3D models in the M groups of grid 3D models to obtain M groups of locally merged models, where M is a positive integer; Perform group label stripping on the M groups of local merged models to obtain K local merged models, where K is a positive integer greater than M; The K local merged models are projected onto the component to be tested to obtain the K defect detection surfaces.

4. The method according to claim 3, characterized in that The method further comprises: If the first principal curvature deviation coefficient does not satisfy the principal curvature deviation threshold, or the first normal vector deviation coefficient does not satisfy the normal vector deviation threshold, calling a preset second mesh density, wherein the second mesh density is twice the first mesh density; Taking the second grid density as a constraint, performing gridding processing on the first grid three-dimensional model to obtain a plurality of secondary grid three-dimensional models; Performing curvature change analysis on the multiple secondary grid three-dimensional models to obtain secondary curvature change analysis results; The first grid three-dimensional model is divided into detection surfaces based on the secondary curvature change analysis result to obtain H secondary defect detection surfaces, where H is a positive integer.

5. The method according to claim 4, characterized in that Pre-constructing a dimming trajectory analysis channel, and synchronizing the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis, and obtaining K dimming trajectory parameters, the method further includes: Interactively obtain multiple sample defect detection surfaces, multiple sample dimming trajectory parameters and multiple sample curvature identifiers of associated mapping; The plurality of sample defect detection surfaces, the plurality of sample dimming trajectory parameters and the plurality of sample curvature identifiers are stored in association with each other to generate the dimming trajectory analysis channel; Synchronizing the K principal curvatures of the K defect detection surfaces to the dimming trajectory analysis channel for curvature data comparison and corresponding sample dimming trajectory parameter call to obtain K standard dimming trajectory parameters; Obtaining K model center points of the K local merged models by calling the component stereo model; The K standard dimming trajectory parameters are optimized according to the K model center points to obtain the K dimming trajectory parameters.

6. The method according to claim 1, characterized in that According to the component stereo model, the K dimming trajectory parameters are connected and fitted with the shortest trajectory to obtain a target dimming trajectory, and according to the target dimming trajectory, the K adaptive dimming strategies are connected to generate a target adaptive dimming strategy. The method further includes: The K dimming trajectory parameters include K trajectory starting points and K trajectory end points; Calling and obtaining a first dimming trajectory parameter from the K dimming trajectory parameters, wherein the first dimming trajectory parameter includes a first trajectory starting point and a first trajectory end point; Calling the first trajectory end point and the K-1 trajectory starting points of the remaining K-1 dimming trajectory parameters based on the component stereo model, K-1 turning line distances; Serialize the K-1 line-turning distances and perform minimum value call, and connect the dimming trajectory parameter corresponding to the minimum value with the first dimming trajectory parameter; Similarly, the target dimming trajectory is obtained by performing shortest trajectory connection fitting on the K dimming trajectory parameters.

7. Adaptive dimming system based on surface detection, characterized in that: For implementing the steps of the method according to any one of claims 1 to 6, the system comprises: A three-dimensional model acquisition module, which is used to perform laser scanning on the component to be measured to obtain component point cloud data, and perform surface modeling based on the component point cloud data to obtain a three-dimensional model of the component; A grid processing module, the grid processing module is used to preset a first grid density, and use the first grid density as a constraint to perform grid processing on the component three-dimensional model to obtain a plurality of grid three-dimensional models; A defect detection surface acquisition module, the defect detection surface acquisition module is used to perform curvature change analysis on the multiple grid three-dimensional models, and divide the component three-dimensional model into detection surfaces based on the curvature change analysis results to obtain K defect detection surfaces; A dimming trajectory parameter acquisition module, wherein the dimming trajectory parameter acquisition module is used to pre-build a dimming trajectory analysis channel, and synchronize the K defect detection surfaces to the dimming trajectory analysis channel to perform dimming trajectory analysis to obtain K dimming trajectory parameters; A dimming test module, the dimming test module is used to interactively obtain a standard defective component of the component to be tested, and perform a dimming test on the standard defective component to obtain multiple dimming photometric parameters of various component apparent defects; An adaptive dimming strategy acquisition module, the adaptive dimming strategy acquisition module is used to fit the multiple dimming light intensity parameters to the K dimming trajectory parameters to obtain K adaptive dimming strategies; A target dimming trajectory acquisition module is used to perform shortest trajectory connection fitting on the K dimming trajectory parameters according to the component three-dimensional model to obtain a target dimming trajectory, and to connect the K adaptive dimming strategies according to the target dimming trajectory to generate a target adaptive dimming strategy.

Images