Intelligent rust removal method, device, equipment and system of laser cleaning machine

Abstract

The application discloses a kind of laser cleaning machine intelligent rust removal method, device, equipment and system, it is related to laser rust removal field, method includes: by controlling camera to the surface of target metal piece photographing obtains metal surface image, rust area is identified and rust rough grade is determined to image analysis;Utilize sensor to collect the environmental parameter of target metal piece application scene, calculate out environmental influence index total;Spectrometer is detected simultaneously to rust area by control, and spectrum detection data is obtained, rust type is determined and chemical composition complexity index is obtained by analysis;Chemical composition complexity index, rust rough grade, rust type and environmental influence index are weighted and summed to calculate out comprehensive rust degree index by synthesis;According to comprehensive rust degree index, the laser cleaning parameter of laser cleaning machine is determined, and laser cleaning machine is controlled to the rust area of target metal piece and carries out laser rust removal.The application can effectively improve the rust removal effect of metal rust.

Description

Technical Field

[0001] This invention relates to the field of laser rust removal technology, and in particular to an intelligent rust removal method, apparatus, equipment and system for laser cleaning machines. Background Technology

[0002] In the field of metal surface treatment, traditional metal rust removal methods mainly include mechanical rust removal and chemical rust removal. Mechanical rust removal typically uses methods such as sandblasting and shot blasting to remove rust from the metal surface through physical impact. However, this method can easily damage the metal surface and creates a harsh working environment with significant noise and dust pollution. Chemical rust removal utilizes chemical reagents such as acids and alkalis to react with the rust and remove it. However, these chemical reagents may corrode the metal substrate and also cause environmental pollution problems.

[0003] With technological advancements, several methods for removing rust from metals using lasers have emerged. However, current laser rust removal techniques typically rely on experience or preset, fixed parameters, lacking accurate assessment and intelligent control of the degree of rust. Specifically, when removing rust from target metal parts, the extent of rust is generally assessed through simple visual observation or coarse inspection methods, followed by direct application of a laser cleaning machine. This approach fails to consider the impact of factors such as the roughness of the rust, the type of rust, and environmental parameters of the application environment on the degree of rust removal.

[0004] Therefore, existing laser rust removal technology lacks accurate judgment and intelligent control of the degree of metal corrosion. It relies solely on experience or fixed parameters for operation, which can easily lead to poor rust removal effect, over-rust removal, or incomplete rust removal, resulting in an unsatisfactory rust removal effect. Summary of the Invention

[0005] This invention provides an intelligent rust removal method, apparatus, equipment, and system for laser cleaning machines, which can effectively improve the rust removal effect on metal corrosion.

[0006] One embodiment of the present invention provides an intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion, comprising:

[0007] The camera is controlled to take pictures of the surface of the target metal part to obtain an image of the metal surface;

[0008] Image analysis is performed on the metal surface image to identify the rusted areas of the target metal part;

[0009] Image analysis is performed on the corroded area to determine the corrosion roughness level of the corroded area;

[0010] The environmental parameters of the application scenario of the target metal part are collected by sensors, and the sum of environmental impact indices affecting the degree of corrosion of the rusted area is calculated based on the environmental parameters.

[0011] The spectral analyzer is controlled to detect the rusted area of ​​the target metal part to obtain spectral detection data;

[0012] The spectral detection data is analyzed to determine the corrosion type of the corrosion area of ​​the target metal part, and the chemical composition complexity index of the corrosion area is obtained.

[0013] The comprehensive corrosion degree index of the rusted area is calculated by weighting and summing the chemical composition complexity index, corrosion roughness level, corrosion type, and environmental impact index affecting the corrosion degree of the rusted area.

[0014] Based on the comprehensive corrosion index, the laser cleaning parameters of the laser cleaning machine are determined;

[0015] Based on the determined laser cleaning parameters, the laser cleaning machine is controlled to perform laser rust removal on the rusted area of ​​the target metal part.

[0016] As an improvement to the above solution, the step of performing image analysis on the corroded area to determine the corrosion roughness level of the corroded area includes:

[0017] The corroded area in the metal surface image is cropped to obtain a corroded area image;

[0018] The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area.

[0019] The rust roughness level of the preprocessed rust area image is calculated using the following formula:

[0020] ;

[0021] in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

[0022] As an improvement to the above solution, the step of analyzing the spectral detection data to determine the corrosion type of the rusted area of ​​the target metal part includes:

[0023] Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data;

[0024] The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula:

[0025] ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element.

[0026] The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

[0027] As an improvement to the above scheme, the weighted summation based on the chemical composition complexity index of the rusted area, the rust roughness level, the rust type, and the environmental impact index affecting the degree of rust in the rusted area includes:

[0028] Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion;

[0029] Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula:

[0030] ;

[0031] in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

[0032] Another embodiment of the present invention provides an intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion, comprising:

[0033] The first control module is used to control the camera to take pictures of the surface of the target metal part to obtain an image of the metal surface;

[0034] The first analysis module is used to perform image analysis on the metal surface image and identify the rust area of ​​the target metal part;

[0035] The second analysis module is used to perform image analysis on the rusted area to determine the rust roughness level of the rusted area;

[0036] The second control module is used to collect environmental parameters of the application scenario of the target metal part through sensors, and calculate the sum of environmental impact indices that affect the degree of corrosion of the rusted area based on the environmental parameters;

[0037] The third control module is used to control the spectrometer to detect the rusted area of ​​the target metal part and obtain spectral detection data;

[0038] The third analysis module is used to analyze the spectral detection data, determine the corrosion type of the corrosion area of ​​the target metal part, and analyze the chemical composition complexity index of the corrosion area.

[0039] The calculation module is used to calculate the comprehensive corrosion degree index of the rusted area by weighted summation of the chemical composition complexity index, the corrosion roughness level, the corrosion type, and the environmental impact index affecting the corrosion degree of the rusted area.

[0040] The determination module is used to determine the laser cleaning parameters of the laser cleaning machine based on the comprehensive corrosion degree index;

[0041] The fourth control module is used to control the laser cleaning machine to perform laser rust removal on the rusted area of ​​the target metal part according to the determined laser cleaning parameters.

[0042] As an improvement to the above solution, the second analysis module is specifically used for:

[0043] The corroded area in the metal surface image is cropped to obtain a corroded area image;

[0044] The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area.

[0045] The rust roughness level of the preprocessed rust area image is calculated using the following formula:

[0046] ;

[0047] in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

[0048] As an improvement to the above solution, the third analysis module is specifically used for:

[0049] Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data;

[0050] The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula:

[0051] ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element.

[0052] The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

[0053] As an improvement to the above solution, the calculation module is specifically used for:

[0054] Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion;

[0055] Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula:

[0056] ;

[0057] in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

[0058] Another embodiment of the present invention provides an intelligent rust removal control device for a laser cleaning machine based on the degree of metal corrosion, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the intelligent rust removal control method for a laser cleaning machine based on the degree of metal corrosion described in the above-described embodiment of the invention.

[0059] Another embodiment of the present invention provides an intelligent rust removal control system for a laser cleaning machine based on the degree of metal corrosion, including a camera, a spectrometer, and the intelligent rust removal control device for a laser cleaning machine based on the degree of metal corrosion described in the above embodiment of the invention; the intelligent rust removal control device for the laser cleaning machine is communicatively connected to the camera and to the spectrometer.

[0060] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0061] First, a camera is used to capture images of the target metal surface. These images are then analyzed to identify rusted areas and determine the rust roughness level. Next, sensors collect environmental parameters of the target metal part's application environment, calculating the total environmental impact index. Simultaneously, a spectrometer is used to detect the rusted area, obtaining spectral data to analyze and determine the rust type and chemical composition complexity index. Then, a comprehensive rust severity index is calculated by weighted summation of the chemical composition complexity index, rust roughness level, rust type, and environmental impact index. Based on this comprehensive rust severity index, the laser cleaning parameters of the laser cleaning machine are determined, and finally, the laser cleaning machine is used to remove rust from the target metal part. This invention effectively solves the problem of existing laser rust removal technologies lacking accurate judgment and intelligent control of metal rust severity. By determining the comprehensive rust severity index through multi-dimensional analysis, laser cleaning parameters are precisely set, achieving efficient and accurate rust removal from the target metal part's rusted areas. This avoids damage to the metal substrate caused by over- or incomplete rust removal, while fully considering the impact of environmental factors on the rust severity. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating an intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion, provided in an embodiment of the present invention.

[0063] Figure 2 This is a schematic diagram of the structure of an intelligent rust removal control device for a laser cleaning machine based on the degree of metal corrosion, provided in an embodiment of the present invention.

[0064] Figure 3 This is a schematic diagram of the structure of an intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion, provided in an embodiment of the present invention;

[0065] Figure 4 This is a schematic diagram of the structure of an intelligent rust removal control system for a laser cleaning machine based on the degree of metal corrosion, provided in an embodiment of the present invention. Detailed Implementation

[0066] 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.

[0067] See Figure 1 This is a flowchart illustrating an intelligent rust removal control method for a laser cleaning machine based on the degree of metal corrosion, according to an embodiment of the present invention. The intelligent rust removal control method for a laser cleaning machine based on the degree of metal corrosion includes steps S10 to S18:

[0068] S10, control the camera to take pictures of the surface of the target metal part to obtain an image of the metal surface;

[0069] S11, perform image analysis on the metal surface image to identify the rusted area of ​​the target metal part;

[0070] S12, perform image analysis on the corroded area to determine the corrosion roughness level of the corroded area;

[0071] S13, collect environmental parameters of the application scenario of the target metal part through sensors, and calculate the sum of environmental impact indices that affect the degree of corrosion of the rusted area based on the environmental parameters;

[0072] S14, control the spectrometer to detect the rusted area of ​​the target metal part to obtain spectral detection data;

[0073] S15, Analyze the spectral detection data to determine the corrosion type of the corrosion area of ​​the target metal part, and analyze the chemical composition complexity index of the corrosion area.

[0074] S16. The comprehensive corrosion degree index of the rusted area is calculated by weighted summation of the chemical composition complexity index, the corrosion roughness level, the corrosion type, and the environmental impact index affecting the corrosion degree of the rusted area.

[0075] S17. Determine the laser cleaning parameters of the laser cleaning machine based on the comprehensive corrosion index.

[0076] S18, according to the determined laser cleaning parameters, control the laser cleaning machine to perform laser rust removal on the rusted area of ​​the target metal part.

[0077] In this embodiment of the invention, firstly, a camera is used to photograph the surface of the target metal part to obtain an image of the metal surface. The image is analyzed to identify rusted areas and determine the rust roughness level. Next, sensors are used to collect environmental parameters of the application scenario of the target metal part, and the sum of environmental impact indices is calculated. Simultaneously, a spectrometer is controlled to detect the rusted area to obtain spectral detection data, which is analyzed to determine the rust type and obtain a chemical composition complexity index. Then, a comprehensive rust degree index is calculated by combining the chemical composition complexity index, rust roughness level, rust type, and the sum of environmental impact indices. Based on the comprehensive rust degree index, the laser cleaning parameters of the laser cleaning machine are determined. Finally, the laser cleaning machine is controlled to perform laser rust removal on the rusted area of ​​the target metal part. This embodiment of the invention effectively solves the problem of existing laser rust removal technologies lacking accurate judgment and intelligent control of the degree of metal rust. By determining the comprehensive rust degree index through multi-dimensional analysis, laser cleaning parameters are precisely set, achieving efficient and accurate rust removal of the rusted area of ​​the target metal part. This avoids damage to the metal substrate caused by over-rust removal or incomplete rust removal, while fully considering the influence of environmental factors on the degree of rust.

[0078] It is understood that existing image analysis methods can be used to analyze the metal surface image in order to identify the rusted areas of the target metal part.

[0079] As an example, through extensive experiments and data analysis, the corresponding combinations of laser cleaning parameters for different comprehensive corrosion index ranges were determined. For instance, the comprehensive corrosion index was divided into three levels: low, medium, and high, each corresponding to different parameters such as laser power, scanning speed, and pulse frequency. For a low comprehensive corrosion index, a lower laser power (e.g., 500W), a higher scanning speed (e.g., 1000mm / s), and a higher pulse frequency (e.g., 50kHz) might be selected. This is because areas with low corrosion require less energy input, while a higher scanning speed can improve rust removal efficiency and avoid overtreatment. For a medium comprehensive corrosion index, a moderate laser power (e.g., 1000W), a moderate scanning speed (e.g., 500mm / s), and a moderate pulse frequency (e.g., 30kHz) could be selected. This ensures rust removal effectiveness while minimizing damage to the metal surface. For a high comprehensive corrosion index, a higher laser power (e.g., 1500W), a lower scanning speed (e.g., 200mm / s), and a lower pulse frequency (e.g., 20kHz) might be selected. Areas with high levels of rust require a larger energy input to effectively remove the rust, but a lower scanning speed ensures that the laser can fully target the rusted area, improving the quality of rust removal. After calculating the overall rust level index of the target metal part's rusted area, suitable laser cleaning parameters for that rust level are determined by consulting a pre-established correspondence table or using an algorithm model.

[0080] In addition, the process of controlling the laser cleaning machine to remove rust can be as follows:

[0081] 1. Equipment Setup: Set the laser power, scanning speed, pulse frequency, and other parameters of the laser cleaning machine to the specified laser cleaning parameters. Ensure that the optical path system, control system, and cooling system of the laser cleaning machine are in normal working order. Adjust the working distance and focusing position of the laser cleaning machine so that the laser can be accurately focused on the rusted area to achieve the best rust removal effect.

[0082] 2. Rust Removal Operation: The laser cleaning machine is activated to remove rust from the rusted areas of the target metal parts. During the rust removal process, the laser beam irradiates the rusted area, causing the rust to vaporize, decompose, or detach due to the instantaneous high temperature, thus achieving the purpose of rust removal. The laser cleaning machine can use continuous scanning or spot scanning, selecting the appropriate scanning mode according to the shape and size of the rusted area. For example, for large areas of rust, continuous scanning mode can be used to improve rust removal efficiency; for complex-shaped or small areas of rust, spot scanning mode can be used to ensure that the laser accurately targets each rust point.

[0083] 3. Real-time Monitoring and Adjustment: During laser rust removal, sensors can monitor the working status and rust removal effect of the laser cleaning machine in real time. For example, temperature sensors can be used to monitor temperature changes on the metal surface to avoid damage to the metal substrate due to excessive heat; optical sensors can be used to monitor the surface quality after rust removal to ensure that the rust removal effect meets requirements. If the rust removal effect is unsatisfactory or abnormalities occur, the laser cleaning parameters can be adjusted in time or the rust removal operation can be stopped for inspection and adjustment. For example, if rust removal is found to be incomplete, the laser power can be increased or the scanning speed can be decreased appropriately; if overheating is found on the metal surface, the laser power can be decreased or the scanning speed can be increased.

[0084] Assuming the target metal component is a part of a large steel bridge structure, the overall corrosion index of a certain rusted area is calculated to be moderate based on the preceding steps. According to a pre-established correlation, laser cleaning parameters of 1000W laser power, 500mm / s scanning speed, and 30kHz pulse frequency are selected. During laser rust removal, the laser cleaning machine parameters are set to the above values, and the working distance and focus position are adjusted. After starting the laser cleaning machine, the rusted area is continuously scanned. During the rust removal process, a temperature sensor monitors the temperature of the metal surface in real time. When the temperature approaches a critical value, the laser power is appropriately reduced or the scanning speed is increased to prevent the metal substrate from overheating. Simultaneously, an optical sensor monitors the surface quality after rust removal. If residual rust is detected, the laser power is appropriately increased or the scanning speed is appropriately decreased to improve the rust removal effect.

[0085] Through the above process, the laser cleaning parameters can be accurately determined based on the comprehensive corrosion degree index, and the laser cleaning machine can be controlled to perform efficient and precise laser rust removal on the rusted areas of the target metal parts.

[0086] As an improvement to the above solution, the step of performing image analysis on the corroded area to determine the corrosion roughness level of the corroded area includes:

[0087] The corroded area in the metal surface image is cropped to obtain a corroded area image;

[0088] The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area.

[0089] The rust roughness level of the preprocessed rust area image is calculated using the following formula:

[0090] ;

[0091] in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

[0092] In this embodiment of the invention, not only are traditional factors such as the total number of pixels, neighborhood size, pixel grayscale value, and average grayscale value considered, but fractal dimension, adjustment parameters, and grayscale gradient feature values ​​are also introduced. This comprehensive consideration of multiple factors can more comprehensively and accurately describe the surface roughness of the rusted area. Simultaneously, the introduction of fractal dimension can better reflect the complexity and irregularity of the rusted area surface. Calculating the fractal dimension using box counting provides a new perspective and method for evaluating the rust roughness level. Furthermore, the addition of grayscale gradient feature values ​​enriches the description of surface roughness. Calculating the gradient amplitude of pixel grayscale values ​​as feature values ​​can capture the grayscale changes between pixels and their neighboring pixels in the rusted area image, further reflecting the surface roughness. Therefore, this embodiment of the invention provides an important basis for determining the comprehensive rust degree index by accurately calculating the rust roughness level. This is because the accuracy of the comprehensive rust degree index directly affects the setting of laser cleaning parameters for the laser cleaning machine. By accurately calculating the rust roughness level, laser cleaning parameters can be set more precisely, improving the accuracy of rust removal and avoiding over-rust removal or incomplete rust removal.

[0093] As an improvement to the above solution, the step of analyzing the spectral detection data to determine the corrosion type of the rusted area of ​​the target metal part includes:

[0094] Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data;

[0095] The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula:

[0096] ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element.

[0097] The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

[0098] In this embodiment of the invention, feature extraction is first performed on the spectral detection data of the rusted area of ​​the target metal part to obtain peak features, band features, and slope features. Then, these features are fused using a specific feature fusion formula, which incorporates weighting coefficients, bias terms, nonlinear transformation functions, and Hadamard product operations to capture the interaction information between different features. Finally, the fused spectral features are input into a convolutional neural network-based rust type classification model to determine the rust type of the rusted area. Specifically, this embodiment of the invention comprehensively considers peak features, band features, and slope features, analyzing spectral data from multiple perspectives, thus improving the accuracy of rust type judgment. Furthermore, during feature fusion, not only are the weights of different feature vectors considered, but the nonlinear transformation function also enhances the feature representation capability. The Hadamard product operation is used to capture the interaction information between features, making the fused features more representative (the Hadamard product operation multiplies element-wise, effectively capturing the interaction relationship between peak, band, and slope features, providing more information for rust type judgment). Moreover, the nonlinear transformation function processes different feature vectors, extracting higher-level feature representations and enhancing feature discriminative power. In summary, this invention, through precise feature extraction, innovative feature fusion, and a classification model employing convolutional neural networks, can accurately determine the rust type of a target metal part's rusted area. This provides crucial information for calculating the comprehensive rust severity index, enabling precise parameter settings for the laser cleaning machine, improving rust removal efficiency and quality, and avoiding poor rust removal results due to inaccurate rust type identification. Furthermore, it fully leverages the advantages of convolutional neural networks, enhancing the overall intelligence level of the rust removal process.

[0099] As an improvement to the above solution, the step of basing the calculation on the chemical composition complexity index of the rusted area, the rust roughness level, the rust type, and the sum of the environmental impact index affecting the degree of rust in the rusted area includes:

[0100] Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion;

[0101] Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula:

[0102] ;

[0103] in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

[0104] In this embodiment of the invention, firstly, the degree of corrosion influence of each corrosion type is determined based on the corrosion type of the target metal part's rusted area. This step helps to more specifically quantify the contribution of different corrosion types to the overall corrosion degree. Then, a comprehensive corrosion degree index is calculated by comprehensively considering multiple key factors. These include the corrosion roughness level, which reflects the surface roughness of the rusted area; the degree of corrosion influence of each corrosion type, clarifying the specific role of different corrosion types; the chemical composition complexity index, reflecting the impact of the diversity of chemical composition in the rusted area on corrosion; and the sum of environmental impact indices, considering the effect of environmental parameters in the application scenario of the target metal part on corrosion. Therefore, this embodiment of the invention comprehensively integrates the influence of multiple aspects such as corrosion roughness level, corrosion type, chemical composition complexity, and environmental parameters on the corrosion degree. Compared with traditional methods, it no longer relies solely on a single factor for evaluation, but comprehensively considers multiple perspectives, making the evaluation results more accurate and comprehensive. For example, this invention not only considers the physical characteristics of the rusted area (rust roughness level) but also deeply analyzes the influence of chemical factors (complexity of chemical composition) and external environmental factors (environmental parameters) on rust. Furthermore, it combines the specific influence degree of each rust type, providing richer information for a comprehensive understanding of the rust condition. In addition, determining the degree of rust influence based on the rust type provides more specific considerations for comprehensive evaluation. Different rust types may have different corrosion mechanisms and development rates. By determining the degree of rust influence for each type, the overall rust situation can be assessed more specifically. For example, different types such as oxidative rust and electrochemical rust may have different degrees of damage to metal parts and different development trends. By determining their respective degrees of rust influence, different weights can be assigned in the comprehensive evaluation. In summary, this invention provides an accurate basis for determining the laser cleaning parameters of a laser cleaning machine by accurately calculating the comprehensive rust degree index. Specifically, it enables a comprehensive and accurate assessment of the rusted area, avoiding the limitations of single-factor assessment. For example, traditional methods, considering only the roughness level of rust, may overlook the influence of rust type, chemical composition, and environmental factors, leading to inaccurate laser cleaning parameter settings and affecting rust removal effectiveness. This technical solution, however, comprehensively considers multiple factors, allowing for more accurate determination of parameters such as laser cleaning machine power, scanning speed, and pulse frequency, improving the targeting and effectiveness of rust removal. Furthermore, it allows for flexible adjustments based on different actual conditions, enhancing the adaptability and practicality of the solution. For instance, under different metal materials, rust environments, and application requirements, adjusting weighting coefficients and preset adjustment indices makes the calculation of the comprehensive rust degree index more consistent with reality, thus better guiding laser rust removal operations.

[0105] Understandably, preset adjustment indices are set for different factors, such as preset adjustment indices for rust roughness level, preset adjustment indices for the degree of influence of rust type, and preset adjustment indices for chemical composition complexity. These adjustment indices can flexibly adjust the weight of each factor in the comprehensive index according to actual conditions, adapting to different metal materials and rusting environments. For example, for certain special metal materials, the influence of chemical composition complexity on rust may be more significant; in this case, the value can be increased to improve the weight of the chemical composition complexity index in the comprehensive rust degree index. The values ​​of the adjustment indices can be determined through the analysis and experimental verification of a large number of actual cases to ensure that they accurately reflect the true impact of each factor on the degree of rust under different conditions.

[0106] It should be noted that, This is the sum of environmental impact indices affecting the degree of corrosion in the corroded area, allowing for a more detailed consideration of the influence of environmental factors on the degree of corrosion. For example, environmental parameters may include humidity, temperature, and pH levels. Different environmental parameters may have varying degrees of impact on corrosion. By calculating the environmental impact index for each environmental parameter separately and then summing them, the combined effect of environmental factors on corrosion can be more accurately reflected. For instance, assuming the target metal part is an outdoor iron structure, consider three environmental parameters: humidity, temperature, and salinity. Humidity can be measured using a humidity sensor. For example, when the relative humidity is below 50%, the impact on corrosion is small, with an environmental impact index of 0.2; 50%-70% is 0.4; and above 70% is 0.6. Temperature is measured using a temperature sensor. The environmental impact index is 0.1 for 0-25°C, 0.3 for 25°C-35°C, and 0.5 for above 35°C. Salinity can be determined by measuring the salt content in the surrounding soil or air. The environmental impact index is 0.1 for low salinity, 0.3 for moderate salinity, and 0.5 for high salinity. Assuming a relative humidity of 65%, a temperature of 32°C, and moderate salinity are measured at a certain moment, the environmental impact index for humidity is 0.4, for temperature it is 0.3, and for salinity it is 0.3. With three environmental parameters (n=3), the total environmental impact index E = 0.4 + 0.3 + 0.3 = 1. This reflects the combined impact of these environmental parameters on the degree of corrosion in the rusted area of ​​the iron structure.

[0107] As an example, existing principal component analysis (PCA) algorithms can be used to extract the spectral features of the spectral detection data, and the chemical composition complexity index of the rusted region can be obtained by assigning values ​​based on the importance of the spectral features. For example, suppose there is a series of spectral detection data of the rusted regions of a target metal part, where each data point contains spectral intensity values ​​at multiple wavelengths. First, these data are preprocessed, including removing outliers and normalizing them. For example, the spectral intensity values ​​can be normalized to the [0,1] interval for subsequent analysis. The spectral features of the spectral detection data are extracted using the PCA algorithm: For the preprocessed spectral data matrix (assuming the number of rows is the number of samples and the number of columns is the number of wavelengths), its covariance matrix is ​​calculated. The covariance matrix reflects the correlation between different wavelengths. Eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and corresponding eigenvectors. The eigenvalues ​​represent the variance along the direction of the corresponding eigenvector. The eigenvectors are arranged in descending order of eigenvalues. Typically, the first few eigenvectors with larger eigenvalues ​​are selected as principal components, as these principal components can explain most of the data variance. Assuming the first k principal components are selected, the original data is projected onto these k principal components to obtain dimensionality-reduced spectral data. Furthermore, the importance of each principal component is determined based on its eigenvalues. Larger eigenvalues ​​indicate a greater contribution of the corresponding principal component to explaining the data variance, thus warranting higher importance. For example, the eigenvalues ​​can be normalized so that their sum is 1. Then, the proportion of each eigenvalue to the total sum of eigenvalues ​​can be used as the importance weight of the corresponding principal component. The chemical composition complexity index is calculated: for each sample's dimensionality-reduced spectral data, a weighted sum is obtained based on the importance weights of the principal components, resulting in a comprehensive value. This comprehensive value can be used as the chemical composition complexity index for that sample. Assuming the dimensionality-reduced spectral data is... ,in The sample is in the first The projection values ​​on each principal component, and the importance weights of the principal components are... The chemical composition complexity index It can be calculated as follows: .

[0108] See Figure 2 This is a schematic diagram of a laser cleaning machine intelligent rust removal control device based on the degree of metal corrosion, according to an embodiment of the present invention. The intelligent rust removal control device for the laser cleaning machine based on the degree of metal corrosion includes:

[0109] The first control module 10 is used to control the camera to take pictures of the surface of the target metal part to obtain an image of the metal surface;

[0110] The first analysis module 11 is used to perform image analysis on the metal surface image and identify the rust area of ​​the target metal part;

[0111] The second analysis module 12 is used to perform image analysis on the rusted area and determine the rust roughness level of the rusted area.

[0112] The second control module 13 is used to collect environmental parameters of the application scenario of the target metal part through sensors, and calculate the sum of environmental impact indices that affect the degree of corrosion of the rusted area based on the environmental parameters.

[0113] The third control module 14 is used to control the spectrometer to detect the rusted area of ​​the target metal part and obtain spectral detection data;

[0114] The third analysis module 15 is used to analyze the spectral detection data, determine the corrosion type of the corrosion area of ​​the target metal part, and analyze the chemical composition complexity index of the corrosion area.

[0115] The calculation module 16 is used to calculate the comprehensive corrosion degree index of the rusted area by weighted summation of the chemical composition complexity index, the corrosion roughness level, the corrosion type, and the environmental impact index affecting the corrosion degree of the rusted area.

[0116] The determination module 17 is used to determine the laser cleaning parameters of the laser cleaning machine based on the comprehensive corrosion degree index;

[0117] The fourth control module 18 is used to control the laser cleaning machine to perform laser rust removal on the rusted area of ​​the target metal part according to the determined laser cleaning parameters.

[0118] In this embodiment of the invention, firstly, a camera is used to photograph the surface of the target metal part to obtain an image of the metal surface. The image is analyzed to identify rusted areas and determine the rust roughness level. Next, sensors are used to collect environmental parameters of the application scenario of the target metal part, and the sum of environmental impact indices is calculated. Simultaneously, a spectrometer is controlled to detect the rusted area to obtain spectral detection data, which is analyzed to determine the rust type and obtain a chemical composition complexity index. Then, a comprehensive rust degree index is calculated by combining the chemical composition complexity index, rust roughness level, rust type, and the sum of environmental impact indices. Based on the comprehensive rust degree index, the laser cleaning parameters of the laser cleaning machine are determined. Finally, the laser cleaning machine is controlled to perform laser rust removal on the rusted area of ​​the target metal part. This embodiment of the invention effectively solves the problem of existing laser rust removal technologies lacking accurate judgment and intelligent control of the degree of metal rust. By determining the comprehensive rust degree index through multi-dimensional analysis, laser cleaning parameters are precisely set, achieving efficient and accurate rust removal of the rusted area of ​​the target metal part. This avoids damage to the metal substrate caused by over-rust removal or incomplete rust removal, while fully considering the influence of environmental factors on the degree of rust.

[0119] As an improvement to the above solution, the second analysis module is specifically used for:

[0120] The corroded area in the metal surface image is cropped to obtain a corroded area image;

[0121] The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area.

[0122] The rust roughness level of the preprocessed rust area image is calculated using the following formula:

[0123] ;

[0124] in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

[0125] As an improvement to the above solution, the third analysis module is specifically used for:

[0126] Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data;

[0127] The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula:

[0128] ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element.

[0129] The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

[0130] As an improvement to the above solution, the calculation module is specifically used for:

[0131] Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion;

[0132] Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula:

[0133] ;

[0134] in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

[0135] It is understood that the solutions of the relevant embodiments of the intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion can be referred to the relevant embodiments of the intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion described above, and will not be repeated here.

[0136] See Figure 3 This is a schematic diagram of an intelligent rust removal control device for a laser cleaning machine based on the degree of metal corrosion, provided in an embodiment of the present invention. This intelligent rust removal control device for a laser cleaning machine based on the degree of metal corrosion includes: a processor 100, a memory 101, and a computer program stored in the memory and executable on the processor, such as an intelligent rust removal control program for a laser cleaning machine based on the degree of metal corrosion. When the processor executes the computer program, it implements the steps in the various embodiments of the intelligent rust removal control method for a laser cleaning machine based on the degree of metal corrosion described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above. See also... Figure 4This is a schematic diagram of an intelligent rust removal control system for a laser cleaning machine based on the degree of metal corrosion, provided in an embodiment of the present invention. The intelligent rust removal control system for a laser cleaning machine based on the degree of metal corrosion includes a camera 1, a spectrometer 2, and the intelligent rust removal control device 3 for a laser cleaning machine based on the degree of metal corrosion described in the above embodiment; the intelligent rust removal control device 3 is communicatively connected to the camera 1 and to the spectrometer 2.

[0137] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion.

[0138] The intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of an intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion and does not constitute a limitation on the device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the intelligent rust removal control device may also include input / output devices, network access devices, buses, etc.

[0139] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the intelligent rust removal control equipment for laser cleaning machines based on the degree of metal corrosion, connecting all parts of the equipment via various interfaces and lines.

[0140] The memory can be used to store the computer programs and / or modules. The processor, by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory, realizes various functions of the intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0141] The modules / units integrated into the intelligent rust removal control equipment for laser cleaning machines based on the degree of metal corrosion, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium may be appropriately added to or subtracted from the content as required by the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium may not include electrical carrier signals and telecommunication signals.

[0142] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0143] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for intelligent rust removal control of a laser cleaning machine based on the degree of metal corrosion, characterized in that, include: The camera is controlled to take pictures of the surface of the target metal part to obtain an image of the metal surface; Image analysis is performed on the metal surface image to identify the rusted areas of the target metal part; Image analysis is performed on the corroded area to determine the corrosion roughness level of the corroded area; The environmental parameters of the application scenario of the target metal part are collected by sensors, and the sum of environmental impact indices affecting the degree of corrosion of the rusted area is calculated based on the environmental parameters. The spectral analyzer is controlled to detect the rusted area of ​​the target metal part to obtain spectral detection data; The spectral detection data is analyzed to determine the corrosion type of the corrosion area of ​​the target metal part, and the chemical composition complexity index of the corrosion area is obtained. The comprehensive corrosion degree index of the rusted area is calculated based on the sum of the chemical composition complexity index, the corrosion roughness level, the corrosion type, and the environmental impact index affecting the corrosion degree of the rusted area. Based on the comprehensive corrosion index, the laser cleaning parameters of the laser cleaning machine are determined; Based on the determined laser cleaning parameters, the laser cleaning machine is controlled to perform laser rust removal on the rusted area of ​​the target metal part; The sum of the chemical composition complexity index of the rusted area, the rust roughness grade, the rust type, and the environmental impact index affecting the degree of rust in the rusted area includes: Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion; Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula: ； in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

2. The intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion as described in claim 1, characterized in that, The step of performing image analysis on the corroded area to determine the corrosion roughness level of the corroded area includes: The corroded area in the metal surface image is cropped to obtain a corroded area image; The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area. The rust roughness level of the preprocessed rust area image is calculated using the following formula: ； in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

3. The intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion as described in claim 1, characterized in that, The step of analyzing the spectral detection data to determine the corrosion type of the rusted area of ​​the target metal part includes: Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data; The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula: ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element. The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

4. A smart rust removal control device for laser cleaning machines based on the degree of metal corrosion, characterized in that, include: The first control module is used to control the camera to take pictures of the surface of the target metal part to obtain an image of the metal surface; The first analysis module is used to perform image analysis on the metal surface image and identify the rust area of ​​the target metal part; The second analysis module is used to perform image analysis on the rusted area to determine the rust roughness level of the rusted area; The second control module is used to collect environmental parameters of the application scenario of the target metal part through sensors, and calculate the sum of environmental impact indices that affect the degree of corrosion of the rusted area based on the environmental parameters; The third control module is used to control the spectrometer to detect the rusted area of ​​the target metal part and obtain spectral detection data; The third analysis module is used to analyze the spectral detection data, determine the corrosion type of the corrosion area of ​​the target metal part, and analyze the chemical composition complexity index of the corrosion area. The calculation module is used to calculate the comprehensive corrosion degree index of the rusted area based on the chemical composition complexity index, the corrosion roughness level, the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the rusted area. The determination module is used to determine the laser cleaning parameters of the laser cleaning machine based on the comprehensive corrosion degree index; The fourth control module is used to control the laser cleaning machine to perform laser rust removal on the rusted area of ​​the target metal part according to the determined laser cleaning parameters; The calculation module is specifically used for: Based on the type of corrosion in the corroded area, determine the degree of corrosion impact of the type of corrosion; Based on the chemical composition complexity index of the corroded area, the corrosion roughness grade, the corrosion impact degree of the corrosion type, and the sum of the environmental impact index affecting the corrosion degree of the corroded area, the comprehensive corrosion degree index of the corroded area is calculated using the following formula: ； in, These are the weighting coefficients, and their sum is 1. It is the roughness level of the corrosion area; It is a preset adjustment index for the roughness level of rust; It refers to the degree of impact of the type of rust. It is a preset adjustment index for the degree of rust impact of different rust types, used to adjust their weight in the comprehensive rust degree index according to the actual situation; It is an index of the complexity of the chemical composition of the corroded area; It is a preset adjustment index for the complexity of chemical composition to adapt to different metal materials and corrosion environments; It is the sum of environmental impact indices that affect the degree of corrosion in the corroded area; It refers to the number of environmental parameters; It is the first The environmental parameter is an environmental impact index that influences the degree of corrosion in the corroded area.

5. The intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion as described in claim 4, characterized in that, The second analysis module is specifically used for: The corroded area in the metal surface image is cropped to obtain a corroded area image; The median filtering algorithm is used to denoise the image of the rusted area, and the histogram equalization algorithm is used to enhance the contrast of the image of the rusted area, so as to obtain the preprocessed image of the rusted area. The rust roughness level of the preprocessed rust area image is calculated using the following formula: ； in, It is the total number of pixels in the preprocessed image of the rusted area; It is the neighborhood size; It is a pixel grayscale value; It is the average gray value of the rusted area; It is the fractal dimension, calculated using the box counting method. Specifically, the pre-processed image of the rusted area is overlaid on a grid of different sizes, the number of non-empty boxes is counted, and then the fractal dimension is calculated through log-regression analysis. and It involves adjusting parameters through experiments. Specifically, an initial value is given, and then the degree of matching between the calculated rust roughness level and the actual rust condition is observed. The parameter values ​​are gradually adjusted until the result that meets the user's needs is obtained. It is a feature value used to describe the gray-level gradient of surface roughness. Specifically, it calculates the gradient magnitude of pixel gray-level values ​​as a feature value. The gradient magnitude is obtained by calculating the gray-level difference between each pixel and its neighboring pixels in the preprocessed rust area image. It is the total number of feature values ​​of the grayscale gradient.

6. The intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion as described in claim 4, characterized in that, The third analysis module is specifically used for: Feature extraction is performed on the spectral detection data to obtain peak features, band features, and slope features of the spectral data; The peak feature, the band feature, and the slope feature are fused, and the fused spectral feature is calculated using the following feature fusion formula: ;in, It is a vector of peak features, and the set of vectors of peak features is... , Indicates the first A vector of peak features; It is a vector of band features, and the set of vectors of band features is... , Indicates the first A vector of band characteristics; These are vectors representing slope features; the set of vectors representing slope features is... , Indicates the first A vector of slope characteristics; These are the weighting coefficients, and their sum is 1. It is a bias term; These are the nonlinear transformation functions for the vectors of peak characteristics, the vectors of band characteristics, and the vectors of slope characteristics, respectively. The Hadamard product, representing the peak, band, and slope feature vectors, is used to capture the interaction information between different features by multiplying them element by element. The fused spectral features are input into a corrosion type classification model that is a convolutional neural network, and the corrosion type of the corrosion region of the target metal part is output.

7. A laser cleaning machine intelligent rust removal control device based on the degree of metal corrosion, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the intelligent rust removal control method for laser cleaning machines based on the degree of metal corrosion as described in any one of claims 1 to 3.

8. An intelligent rust removal control system for a laser cleaning machine based on the degree of metal corrosion, characterized in that, It includes a camera, a spectrometer, and the intelligent rust removal control device for laser cleaning machines based on the degree of metal corrosion as described in claim 7; the intelligent rust removal control device for laser cleaning machines is communicatively connected to the camera and to the spectrometer.

Images