High-precision length measurement method and system based on laser technology

Abstract

The present invention discloses a high-precision length measurement method and system based on laser technology, including obtaining working data and status data of laser measurement, calculating laser edge coordinates, the first laser measurement length, and ranging scale according to the laser scan data, processing the camera image to obtain pixel edge coordinates, matching the pixel edge coordinates and the laser edge coordinates to obtain the first edge coordinates, fusing the first edge coordinates and the second laser measurement length to obtain the laser image length, constructing a high-precision length measurement model according to the comprehensive measurement length, the ranging scale, the transmission influence factor, and the first edge coordinates, and inputting the working data and status data to be measured into the high-precision length measurement model to obtain the measurement length. This method can not only improve the accuracy of laser length measurement, but also has good interpretability, and can be directly applied to a high-precision length measurement system based on laser technology.

Description

Technical Field

[0001] The present invention relates to the field of length measurement, and particularly to a high-precision length measurement method and system based on laser technology. Background Art

[0002] With the development of modern industry and technology, the requirements for the accuracy of spatial three-dimensional coordinates in fields such as major equipment manufacturing, space technology, and national defense industry are increasing day by day. Due to its unique advantages such as non-contact, high precision, high speed, and high reliability, laser technology is playing an increasingly important role in the field of high-precision length measurement.

[0003] Although traditional laser measurement technologies meet the accuracy requirements to a certain extent, they still face many challenges in complex environments, dynamic measurements, and long-distance measurements, and need to be adjusted in combination with effective auxiliary measurement methods. At the same time, laser measurement has a large amount of sampling and rich data content, which puts higher requirements on the data processing ability of the measurement system. With the rapid development of computer vision technology, image recognition technology has been widely used in length measurement. Through the processing of images or videos, various functions such as target detection, target tracking, image segmentation, and image recognition can be realized, which can be used as an effective length auxiliary measurement method. By fully considering factors such as transmission influence, image conversion, and status data, combining laser technology, image recognition technology, and machine learning technology, a precise, efficient, and fast high-precision length measurement method and system based on laser technology are designed to overcome the deficiencies of existing length measurement methods and systems, achieve efficient and high-precision length measurement, and provide strong technical support for fields such as industrial automation and precision manufacturing. Summary of the Invention

[0004] The purpose of the present invention is to provide a high-precision length measurement method and system based on laser technology.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0006] The present invention includes the following steps:

[0007] Obtain the working data and status data of laser measurement, and preprocess the working data and the status data; the working data includes laser scan data and camera images; the status data includes first status data and second status data;

[0008] Calculate the laser edge coordinates, the first laser measurement length, and the ranging scale according to the laser scan data, input the first status data into the transmission influence function to obtain the transmission influence factor, and determine the second laser measurement length according to the transmission influence factor and the first laser measurement length;

[0009] Process the camera image to obtain pixel edge coordinates, convert the pixel edge coordinates into world edge coordinates according to the second state data, and match the world edge coordinates with the laser edge coordinates to obtain the first edge coordinates;

[0010] Fuse the first edge coordinates and the second laser measurement length to obtain the laser image length, and adjust the laser image length using the ranging scale to obtain the comprehensive measurement length;

[0011] Construct a high-precision length measurement model based on the comprehensive measurement length, the ranging scale, the transmission influence factor, and the first edge coordinates, and input the work data and state data to be measured into the high-precision length measurement model to obtain the measurement length.

[0012] Further, the method for calculating the laser edge coordinates, the first laser measurement length, and the ranging scale includes:

[0013] Use a laser scanner and a camera as a pair of scanning groups, set more than three pairs of scanning groups at different spatial positions, and obtain the machine work data and state data;

[0014] Determine the reference center and coordinate system of the surface of the workpiece to be measured. When laser scanning, project the spectral identifier as a mark on the reference center and edge of the surface of the workpiece to be measured, calculate the relative coordinates of the edge point and the reference center as the relative laser edge coordinates, and determine the laser edge coordinates based on the mean, variance, and skewness of the relative laser edge coordinates of multiple scanning groups;

[0015] Sort the laser scanning data in time series, set the time point when the phase of the laser scanning data changes as the edge time point, calculate the interval of the edge time points to obtain the scanning time, determine the rotation angle of the laser scanner based on the scanning time and the rotation speed of the laser scanner, calculate the edge distance from the scanning point to the edge point based on the round-trip time of the laser beam corresponding to the edge time point, and calculate the first laser measurement length based on the edge distance and the rotation angle of the laser scanner;

[0016] Calculate the ranging scale from the scanning point to the centroid of the surface of the reference object based on the round-trip time of the laser beam corresponding to the time point when the laser scans to the centroid mark on the surface of the reference object.

[0017] Further, the method for obtaining the transmission influence factor includes:

[0018] Input the first state data into the transmission influence function to obtain the transmission influence factor, and the expression is:

[0019]

[0020] where Q Tis the transmission influence factor, w1, w2, w3 are the transmission influence factor weight coefficients, Δf is the deviation between the actual frequency of the emitted laser and the preset frequency, f0 is the preset frequency of the emitted laser, ΔT is the temperature drift of the laser, k is the temperature wavelength influence coefficient, Δλ max is the deviation between the end value of the laser spectrum and the atmospheric spectrum, Δλ min is the deviation between the laser spectrum and the starting value of the atmospheric spectrum, Δλ is the range of the atmospheric spectrum;

[0021] The second laser measurement length is determined according to the transmission influence factor and the first laser measurement length, and the expression is:

[0022]

[0023] Wherein L2 is the second laser measurement length, L1 is the first laser measurement length, and w4 is the transmission influence factor weight.

[0024] Furthermore, the method for obtaining the pixel edge coordinates includes:

[0025] The OpenCV image processing library is used to read the camera image to be processed, the camera image is grayed to obtain a grayscale image, and the edge detection of the grayscale image is performed to identify the edge position of the grayscale image of the test piece;

[0026] The edge detector uses a Gaussian Laplace filter to perform Gaussian filtering on the grayscale image and perform Laplace transform to obtain the image edge information. The expression is:

[0027]

[0028] in is the Laplace Gaussian operator, is the Laplace operator, G(u,v) is a two-dimensional Gaussian function, σ is the standard deviation of the Gaussian function, g(u,v) is the image after Gaussian Laplace filtering, and f(u,v) is the input grayscale image;

[0029] According to the edge information, a zero-crossing test is performed to obtain the specific edge position. According to the edge position, the eight-connected region is used to trace the image contour in a counterclockwise direction. The expression of the eight-connected rule is:

[0030] N8(p)=H4(p)∪(u+1,v+1),(u+1,v-1),(u-1,v+1),(u-1,v-1)

[0031] Where N8(p) represents the eight-connected region of the starting point p of the boundary tracking, including point p and its corresponding upper, lower, left, right and four diagonal neighboring points; N4(p) represents the four-connected region of point p, including point p and its corresponding upper, lower, left and right neighboring points;

[0032] Filter the contour points by dividing them into five distance intervals with different distances according to the grayscale calculation value. The filtering principle is as follows:

[0033] M1>1-T, M2<T, M4<T, M5>1-T

[0034] Where M1, M2, M4, and M5 are the average grayscale values of regions 1, 2, 4, and 5, and T is the grayscale threshold;

[0035] Determine the filtered contour points, calculate the geometric center using the moment function, and extract the pixel edge coordinates according to the contour data.

[0036] Furthermore, the method for converting the pixel edge coordinates into world edge coordinates includes:

[0037] The second state data is the camera focal length, camera pitch angle, and translation amount;

[0038] Construct the camera internal parameter matrix and external parameter matrix according to the second state data, and convert the pixel edge coordinates (u, v) into camera edge coordinates (X c , Y c , Z c ). Then convert the camera edge coordinates (X c , Y c , Z c ) into world edge coordinates (X w , Y w , Z w ).

[0039] Furthermore, the method for obtaining the first edge coordinates includes:

[0040] Calculate the comprehensive similarity between the world edge coordinates and the laser edge coordinates. The expression is:

[0041]

[0042] Where S w,l is the comprehensive similarity between the world edge coordinates and the laser edge coordinates, α and β are weight parameters, is the Euclidean distance between the i-th world edge coordinate point A i and the j-th laser edge coordinate point B j , n is the number of coordinate points, r is the buffer radius, d A is the distance from the world edge coordinate point to the origin, d B is the distance from the laser edge coordinate point to the origin, is the mean value of the corresponding vectors of the world edge coordinate points, is the mean value of the corresponding vectors of the laser edge coordinate points;

[0043] Match the edge coordinate points according to the comprehensive similarity between the world edge coordinates and the laser edge coordinates, and determine the first edge coordinates according to the mean, variance, and skewness of the two edge coordinate points.

[0044] Further, the method for obtaining the length of the laser image includes:

[0045] Calculate the distance and angle between any two first edge coordinate points, and define two first edge coordinate points and the corresponding distance and angle as a set of edge coordinate data;

[0046] Perform feature fusion on the edge coordinate data and the second laser measurement length to obtain the laser image length;

[0047] Adopt the interactive multi-feature fusion method RCI and the stacked ensemble classifier for feature fusion;

[0048] The interactive multi-feature fusion method RCI includes: calculating the correlation degree, calculating the connection entropy, calculating the interaction factor, and determining the evaluation criterion;

[0049] Determine the evaluation criterion for candidate features, and the expression is:

[0050]

[0051] where J RCI (L1) is the evaluation criterion for candidate features, F is all features, L2 is the correlation degree between the second laser measurement length and the edge coordinate data feature c Edge ; is the redundancy, s is the feature in the selected feature subset Ω s ; is the interaction factor, H c (Ω s ,L2,c Edge ) is the joint entropy of the selected feature subset Ω s and the edge coordinate data feature c Edge when considering the second laser measurement length L2, H c (Ω s ,c Edge ) is the joint entropy of only considering the selected feature subset Ω s and the edge coordinate data feature c;

[0052] Obtain the optimal feature subset according to the evaluation criterion of candidate features, and input the optimal feature subset into the stacked ensemble classifier to obtain the laser image length;

[0053] The stacked ensemble classifier includes an input layer, a first classification layer, a second classification layer, and an output layer;

[0054] The first classification layer consists of a deep neural network (DNN) and a lightweight gradient boosting machine (LightGBM), which perform basic classification on the input data to output probability results; the second classification layer consists of a support vector machine (SVM), which performs meta-classification on the input probability results to determine the fusion result of the feature data, and the comprehensive measurement length is output by the output layer; the performance of the model is evaluated using the F1 score, G-mean, mean squared error (MSE), and cross-validation.

[0055] Further, the method for obtaining the measurement length includes:

[0056] Adjust the laser image length of different spatial position scanning groups according to the ranging scale to obtain the comprehensive measurement length, and form a comprehensive measurement set with the transmission influence factor, ranging scale, and first edge coordinate. Divide the comprehensive measurement set into a training set and a test set;

[0057] Construct a high-precision length measurement model, which includes an input layer, a base model layer, a strategy layer, and an output layer;

[0058] The base model layer is composed of two multi-layer perceptron models and two gradient boosting machine models in parallel;

[0059] In the multi-layer perceptron model: Design two hidden layers, use the ReLU function as the activation function to accelerate training and reduce gradient disappearance, select the Cross-Entropy Loss function to solve the deviation of the cross-entropy loss function from the evaluation index and overconfidence, and select Adam to optimize the model weight parameters;

[0060] In the gradient boosting machine model: Construct a mathematical model through an iterative method, use the logarithmic loss function to measure the difference between the predicted probability distribution and the true distribution, and adjust the learning rate to control the optimization process;

[0061] The strategy layer uses the stacking method to input the prediction structure of the base model into a new model for prediction to obtain the prediction result, and uses the test set to evaluate the high-precision length measurement model;

[0062] Input the working data and status data of the machine to be measured into the high-precision length measurement model to obtain the measurement length.

[0063] In a second aspect, a high-precision length measurement system based on laser technology includes:

[0064] A data acquisition module: including a laser transmitter, a laser receiver, and a high-resolution camera; used to collect the working data and status data of the machine, and preprocess the working data and the status data;

[0065] Data processing module: used to calculate laser edge coordinates, the first laser measurement length, and the ranging scale based on the laser scanning data, used to obtain the transmission influence factor, used to determine the second laser measurement length, and used to obtain the comprehensive measurement length based on the ranging scale;

[0066] Fusion model module: used to perform feature fusion on the edge coordinate data and the first laser measurement length by using the interactive multi-feature fusion method RCI and the stacked ensemble classifier;

[0067] Measurement model module: used to construct a high-precision length measurement model based on the comprehensive measurement length, the ranging scale, the transmission influence factor, and the first edge coordinate, and input the work data and status data to be measured into the high-precision length measurement model to obtain the measurement length;

[0068] Intelligent supervision module: used to store, view, and manage the work data, the status data, and the measurement length.

[0069] The beneficial effects of the present invention are:

[0070] The present invention is a high-precision length measurement method and system based on laser technology. Compared with the prior art, the present invention has the following technical effects:

[0071] Through steps such as processing laser scanning data, constructing a transmission influence factor, image recognition, coordinate transformation, data fusion, and model construction, the present invention can improve the accuracy of laser technology length measurement, thereby improving the efficiency and accuracy of high-precision length measurement based on laser technology. The intelligentization of laser technology length measurement can greatly save resources and improve the length measurement efficiency. It can achieve high-precision length measurement of the object to be measured, quickly perform high-precision length measurement on the object to be measured, provide strong technical support for fields such as modern industrial manufacturing, precision measurement, and automation control, and is of great significance for high-precision length measurement based on laser technology. It can adapt to the measurement needs of different measurement systems and different users for high-precision length based on laser technology, and has a certain universality. Description of the Drawings

[0072] Figure 1 It is a flowchart of the steps of the high-precision length measurement method based on laser technology of the present invention. Detailed Embodiments

[0073] The present invention will be further described below through specific embodiments. The illustrative embodiments and explanations of the present invention are used to explain the present invention, but do not limit the present invention.

[0074] The high-precision length measurement method and system based on laser technology of the present invention include the following steps:

[0075] As Figure 1 shown, in this embodiment, the following steps are included:

[0076] Obtain the working data and status data measured by laser, and preprocess the working data and the status data; the working data includes laser scan data and camera images; the status data includes first status data and second status data;

[0077] Calculate the laser edge coordinates, the first laser measurement length, and the ranging scale according to the laser scan data, input the first status data into the transmission influence function to obtain the transmission influence factor, and determine the second laser measurement length according to the transmission influence factor and the first laser measurement length;

[0078] Process the camera image to obtain pixel edge coordinates, convert the pixel edge coordinates into world edge coordinates according to the second status data, and match the world edge coordinates with the laser edge coordinates to obtain the first edge coordinates;

[0079] Fuse the first edge coordinates and the second laser measurement length to obtain the laser image length, and adjust the laser image length by using the ranging scale to obtain the comprehensive measurement length;

[0080] Construct a high-precision length measurement model according to the comprehensive measurement length, the ranging scale, the transmission influence factor, and the first edge coordinates, and input the working data and status data to be measured into the high-precision length measurement model to obtain the measurement length.

[0081] In this embodiment, the method for calculating the laser edge coordinates, the first laser measurement length, and the ranging scale includes:

[0082] Take the laser scanner and the camera as a pair of scanning groups, set more than three pairs of scanning groups at different spatial positions, and obtain the machine working data and status data;

[0083] Determine the reference center and coordinate system of the surface of the workpiece to be measured. When laser scanning, project the spectral identifier as an identifier on the reference center and the edge of the surface of the workpiece to be measured, calculate the relative coordinates of the edge point and the reference center as the relative laser edge coordinates, and determine the laser edge coordinates according to the mean value, variance, and skewness of the relative laser edge coordinates of multiple scanning groups;

[0084] Sort the laser scan data in time series, set the time point when the phase of the laser scan data changes as the edge time point, calculate the interval of the edge time points to obtain the scanning time, determine the rotation angle of the laser scanner through the scanning time and the rotation speed of the laser scanner, calculate the edge distance from the scanning point to the edge point according to the round-trip time of the laser beam corresponding to the edge time point, and calculate the first laser measurement length according to the edge distance and the rotation angle of the laser scanner;

[0085] Calculate the ranging scale from the scanning point to the centroid of the reference object surface based on the round-trip time of the laser beam corresponding to the time point when the laser scans to the centroid mark on the reference object surface;

[0086] In actual evaluation, for high-precision length measurement of a drive shaft produced by a certain factory based on laser technology, three pairs of scanning groups (left, middle, right) are set to obtain machine working data and corresponding status data. When the laser scans, the spectral mark is projected as a mark on the center and side lines of the drive shaft axis to obtain laser edge coordinates (unit: mm): Edge point 1 (250.61, 250.19, 0.51), Edge point 2 (25.91, 249.87, 1500.49); Edge point 1 (250.33, 250.47, 0.02), Edge point 2 (249.99, 250.15, 1500.13); Edge point 1 (250.18, 250.22, -0.08), Edge point 2 (250.03, 250.05, 1499.97). Thus, the laser edge coordinates obtained are: Edge point 1 (250.33, 250.13, -0.10), Edge point 2 (250.14, 250.02, 1500.14);

[0087] Calculate the scanning rotation angle and the distance from the laser scanner to the edge point based on the laser scanning data respectively as: Left-side scanning point 45.01°, 1500.12mm, 2121.45mm; Middle scanning point 20°, 4319.38mm, 4319.19mm; Right-side scanning point 30°, 3000.33mm, 2597.84mm; The corresponding first laser measurement length is: Left-side scanning point 1500.83mm, Middle scanning point 1499.75mm, Right-side scanning point 1501.07mm;

[0088] The ranging scales from the three laser scanners to the center point of the drive shaft are: Left-side scanning point 1677.24mm, Middle scanning point 4253.79mm, Right-side scanning point 2704.08mm.

[0089] In this embodiment, the method for obtaining the transmission influence factor includes:

[0090] Input the first status data into the transmission influence function to obtain the transmission influence factor, and the expression is:

[0091]

[0092] Where Q T is the transmission influence factor, w1, w2, w3 are transmission influence factor weight coefficients, Δf is the deviation between the actual frequency and the preset frequency of the emitted laser, f0 is the preset frequency of the emitted laser, ΔT is the temperature drift of the laser, k is the temperature wavelength influence coefficient, Δλmax is the deviation between the end value of the laser spectrum and the atmospheric spectrum, Δλ min is the deviation between the laser spectrum and the starting value of the atmospheric spectrum, Δλ is the range of the atmospheric spectrum;

[0093] The second laser measurement length is determined according to the transmission influence factor and the first laser measurement length, and the expression is:

[0094]

[0095] Where L2 is the second laser measurement length, L1 is the first laser measurement length, and w4 is the transmission influence factor weight;

[0096] In the actual evaluation, when high-precision length measurement of the drive shaft produced in a factory was performed based on laser technology, the rated frequency of the laser scanner was 20kHz, the actual frequency was 18kHz, the scanning wavelength was 790-910nm, the atmospheric visible spectrum was 380-780nm, the temperature drift of the laser was 0.5℃, and the temperature-wavelength influence coefficient was 0.1nm / ℃. The transmission influence factor was 0.99944, and the corresponding second laser measurement lengths were 1500.93mm, 1499.84mm, and 1501.16mm.

[0097] In this embodiment, the method for obtaining the pixel edge coordinates includes:

[0098] The OpenCV image processing library is used to read the camera image to be processed, the camera image is grayed to obtain a grayscale image, and the edge detection of the grayscale image is performed to identify the edge position of the grayscale image of the test piece;

[0099] The edge detector uses a Gaussian Laplace filter to perform Gaussian filtering on the grayscale image and perform Laplace transform to obtain the image edge information. The expression is:

[0100]

[0101] in is the Laplace Gaussian operator, is the Laplace operator, G(u,v) is a two-dimensional Gaussian function, σ is the standard deviation of the Gaussian function, g(u,v) is the image after Gaussian Laplace filtering, and f(u,v) is the input grayscale image;

[0102] A zero-crossing test is performed based on the edge information to obtain the specific edge position, a grayscale threshold is set, the absolute value of the difference between the grayscale values of adjacent pixels on the four side lines and the two diagonal lines is calculated, and the pixels whose absolute value is greater than the grayscale threshold are determined to be located on the edge;

[0103] Perform contour tracing on the image in a counterclockwise direction for the eight-connected region according to the edge position. The expression for eight-connectivity is:

[0104] N8(p) = N4(p) ∪ (u + 1, v + 1), (u + 1, v - 1), (u - 1, v + 1), (u - 1, v - 1)

[0105] Where N8(p) represents the eight-connected region of the starting point p of the boundary tracking, including point p and its corresponding upper, lower, left, right, and four diagonal neighboring points, and N4(p) represents the four-connected region of point p, including point p and its corresponding upper, lower, left, and right neighboring points;

[0106] Screen the contour points by dividing them into five distance intervals with different distances according to the grayscale calculation value. The screening principle is:

[0107] M1 > 1 - T, M2 < T, M4 < T, M5 > 1 - T

[0108] Where M1, M2, M4, and M5 are the average grayscales of regions 1, 2, 4, and 5, and T is the grayscale threshold;

[0109] Determine the screened contour points, calculate the geometric center using the moment function, and extract the pixel edge coordinates according to the contour data;

[0110] In the actual evaluation, the camera pixels are 1920 * 1080, the camera focal length is 50mm, and the camera pitch angles are -10°, 0°, and 10°. After grayscale conversion, edge detection, and contour screening for three groups of images, three groups of pixel edge coordinates are obtained: left-side scanning edge point 1(650.12, 539.23), edge point 2(800.12, 689.23); middle-side scanning edge point 1(650.09, 539.25), edge point 2(800.16, 689.19); right-side scanning edge point 1(650.15, 539.19), edge point 2(800.08, 689.24).

[0111] In this embodiment, the method for converting the pixel edge coordinates into world edge coordinates includes:

[0112] The second state data is the camera focal length, camera pitch angle, and translation amount;

[0113] Convert the pixel edge coordinates (u, v) to camera edge coordinates (X c , Y c , Z c ), and the expression is:

[0114]

[0115] Where (u0, v0) is the coordinate of the image coordinate origin in the pixel coordinate system, is the scaling factor along the axis of the image coordinate system, f is the camera focal length, and x, y are the image coordinates;

[0116] Convert the camera edge coordinates (X c , Y c , Z c ) into world edge coordinates (X w , Y w , Z w ), and the expression is:

[0117]

[0118] where R is the rotation matrix, θ is the camera pitch angle, and T is the translation matrix;

[0119] In the actual evaluation, the world edge coordinates (unit: mm) are calculated based on the internal and external parameters of the camera: Edge point 1 (-155.88, -3.8, 1810.79), Edge point 2 (1344.12, -3.8, 1810.79); Edge point 1 (-155.90, -3.8, 1809.55), Edge point 2 (1344.17, -3.8, 1809.55); Edge point 1 (-155.76, -3.8, 1810.11), Edge point 2 (1344.06, -3.8, 1810.11).

[0120] In this embodiment, the method for obtaining the first edge coordinates includes:

[0121] Calculate the comprehensive similarity between the world edge coordinates and the laser edge coordinates, and the expression is:

[0122]

[0123] where S w,l is the comprehensive similarity between the world edge coordinates and the laser edge coordinates, α, β are weight parameters, is the Euclidean distance between the i-th world edge coordinate point A i and the j-th laser edge coordinate point B j , n is the number of coordinate points, r is the buffer radius, d A is the distance from the world edge coordinate point to the origin, d B is the distance from the laser edge coordinate point to the origin, is the mean value of the corresponding vectors of the world edge coordinate points, is the mean value of the corresponding vectors of the laser edge coordinate points;

[0124] Match the edge coordinate points according to the comprehensive similarity between the world edge coordinates and the laser edge coordinates, and determine the first edge coordinates according to the mean value, variance, and skewness of the two edge coordinate points.

[0125] In the actual evaluation, the X coordinate of the world edge coordinate corresponds to the Z coordinate of the laser edge coordinate, and the Z coordinate of the world edge coordinate corresponds to the X coordinate of the laser edge coordinate. According to the similarity matching, the first edge coordinates are determined as: Edge point 1 (1030.24, 123.17, -77.98), Edge point 2 (1030.15, 123.11, 142.13).

[0126] In this embodiment, the method for obtaining the length of the laser image includes:

[0127] Calculate the distance and angle between any two first edge coordinate points, and define two first edge coordinate points and the corresponding distance and angle as a set of edge coordinate data;

[0128] Perform feature fusion on the edge coordinate data and the second laser measurement length to obtain the length of the laser image;

[0129] Adopt the interactive multi-feature fusion method RCI and the stacked ensemble classifier for feature fusion;

[0130] The interactive multi-feature fusion method RCI includes: calculating the correlation degree, calculating the connection entropy, calculating the interaction factor, and determining the evaluation criterion;

[0131] Analyze the correlation degree between the edge coordinate data and the second laser measurement length, and select the corresponding edge coordinate data, correlation degree, and second laser measurement length according to the correlation degree to form a comprehensive measurement feature;

[0132] Calculate the connection entropy between the comprehensive measurement features to measure the non-linear relationship between variables. The expression is:

[0133]

[0134] Where is the connection entropy between the comprehensive measurement features, c(·) is the density function, is a random variable, and its marginal function is F i is the candidate feature, and n is the number of features;

[0135] Calculate the interaction factor to measure the interaction between the edge coordinate data feature and the second laser measurement length, and select the variable with the largest interaction factor. The expression is:

[0136]

[0137] Where is the interaction factor, L2 is the second laser measurement length, H c (Ω s ,L2,c EdgeWhen considering the second laser measurement length L2, the selected feature subset Ω s and the edge coordinate data feature c Edge of the joint entropy, H c (Ω s , c Edge ) is to only consider the selected feature subset Ω s and the edge coordinate data feature c Edge of the joint entropy;

[0138] Determine the evaluation criteria for candidate features, and the expression is:

[0139]

[0140] where J RCI (L2) is the evaluation criteria for candidate features, F is all features, is the correlation between L2 as the second laser measurement length and the edge coordinate data feature c Edge ; is the redundancy, s is the feature in the selected feature subset Ω s ;

[0141] Obtain the optimal feature subset according to the evaluation criteria of candidate features, and input the optimal feature subset into the stacked ensemble classifier to obtain the laser image length;

[0142] The stacked ensemble classifier includes an input layer, a first classification layer, a second classification layer, and an output layer;

[0143] The first classification layer is composed of a deep neural network DNN and a lightweight gradient boosting machine LightGBM, which performs basic classification on the input data and outputs a probability result; the second classification layer is composed of a support vector machine SVM, which inputs the probability result for meta-classification to determine the fusion result of feature data, and the output layer outputs the comprehensive measurement length; the performance of the model is evaluated using the F1 score, G-mean, mean square error MSE, and cross-validation;

[0144] In actual evaluation, data fusion is performed on the edge coordinates and the second laser measurement length to obtain the laser image length: 1500.52 mm on the left, 1499.97 mm in the middle, and 1500.63 mm.

[0145] In this embodiment, the method for obtaining the measurement length includes:

[0146] Adjust the laser image length of different spatial position scanning groups according to the ranging scale to obtain the comprehensive measurement length, and form a comprehensive measurement set with the transmission influence factor, ranging scale, and first edge coordinates, and divide the comprehensive measurement set into a training set and a test set;

[0147] Build a high-precision length measurement model, which includes an input layer, a base model layer, a policy layer, and an output layer;

[0148] The base model layer is composed of two multi-layer perceptron models and two gradient boosting machine models in parallel;

[0149] In the multi-layer perceptron model: Design two hidden layers, use the ReLU function as the activation function to accelerate training and reduce gradient disappearance, select the Cross-Entropy Loss function to solve the problem that the cross-entropy loss function deviates from the evaluation index and overconfidence, and select Adam to optimize the model weight parameters;

[0150] In the gradient boosting machine model: Build a mathematical model through an iterative method, use the logarithmic loss function to measure the difference between the predicted probability distribution and the true distribution, and adjust the learning rate to control the optimization process;

[0151] The policy layer uses the stacking method to input the prediction structure of the base model into a new model for prediction to obtain the prediction result, and uses the test set to evaluate the high-precision length measurement model;

[0152] Input the working data and status data of the machine to be measured into the high-precision length measurement model to obtain the measured length;

[0153] In the actual evaluation, adjust the length of the laser image according to the ranging scale (1677.24mm on the left, 4253.79mm in the middle, 2704.08mm on the right) to obtain the comprehensive measured length: (1500.13mm on the left, 1500.06mm in the middle, 1500.38mm on the right);

[0154] Input the working data and status data of the group to be measured into the high-precision length measurement model, and the measured length is 1500.16mm. The nominal length of the drive shaft is 1500.15mm. The measurement accuracy of the high-precision length measurement method based on laser technology reaches 0.01mm, and the length measurement deviation is 0.00066%.

[0155] In the second aspect, a high-precision length measurement system based on laser technology includes:

[0156] Data acquisition module: It includes a laser transmitter, a laser receiver, and a high-resolution camera; it is used to collect the machine working data and status data, and preprocess the working data and the status data;

[0157] Data processing module: It is used to calculate the laser edge coordinates, the first laser measurement length, and the ranging scale according to the laser scan data, to obtain the transmission influence factor, to determine the second laser measurement length, and to obtain the comprehensive measurement length according to the ranging scale;

[0158] Fusion model module: It is used to perform feature fusion on the edge coordinate data and the first laser measurement length by using the interactive multi-feature fusion method RCI and the stacked ensemble classifier;

[0159] Measurement model module: It is used to construct a high-precision length measurement model based on the comprehensive measurement length, the ranging scale, the transmission influence factor, and the first edge coordinate, and input the work data and status data to be measured into the high-precision length measurement model to obtain the measurement length;

[0160] Intelligent supervision module: It is used to store, view, and manage the work data, the status data, and the measurement length.

[0161] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A high-precision length measurement method based on laser technology, characterized in that, Including the following steps: S1. Obtain the working data and status data measured by laser, and preprocess the working data and the status data; the working data includes laser scanning data and camera images; the status data includes first status data and second status data; S2. Calculate the laser edge coordinates, the first laser measurement length and the ranging scale according to the laser scanning data, input the first status data into the transmission influence function to obtain the transmission influence factor, and determine the second laser measurement length according to the transmission influence factor and the first laser measurement length; S3. Process the camera image to obtain the pixel edge coordinates, convert the pixel edge coordinates into world edge coordinates according to the second status data, and match the world edge coordinates with the laser edge coordinates to obtain the first edge coordinates; S4. Fuse the first edge coordinates and the second laser measurement length to obtain the laser image length, and adjust the laser image length by using the ranging scale to obtain the comprehensive measurement length; S5. Construct a high-precision length measurement model according to the comprehensive measurement length, the ranging scale, the transmission influence factor and the first edge coordinates, and input the working data and status data to be measured into the high-precision length measurement model to obtain the measurement length; The second status data is the camera focal length, the camera pitch angle and the translation amount.

2. The high-precision length measurement method based on laser technology according to claim 1, characterized in that, The method for calculating the laser edge coordinates, the first laser measurement length and the ranging scale includes: Taking the laser scanner and the camera as a pair of scanning groups, setting more than three pairs of scanning groups at different spatial positions, and obtaining the machine working data and status data; Determine the reference center and coordinate system of the surface of the workpiece to be measured. When laser scanning, project the spectral identifier as an identifier on the reference center and edge of the surface of the workpiece to be measured, calculate the relative coordinates of the edge point and the reference center as the relative laser edge coordinates, and determine the laser edge coordinates according to the mean value, variance and skewness of the relative laser edge coordinates of multiple scanning groups; Sort the laser scanning data in time series, set the time point when the phase of the laser scanning data changes as the edge time point, calculate the interval of the edge time points to obtain the scanning time, determine the rotation angle of the laser scanner according to the scanning time and the rotation speed of the laser scanner, calculate the edge distance from the scanning point to the edge point according to the round-trip time of the laser beam corresponding to the edge time point, and calculate the first laser measurement length according to the edge distance and the rotation angle of the laser scanner; Calculate the ranging scale from the scanning point to the centroid of the reference object surface according to the round-trip time of the laser beam corresponding to the time point when the laser scans to the centroid identifier of the reference object surface.

3. The high-precision length measurement method based on laser technology according to claim 1, wherein The method for obtaining the transmission influence factor includes: Input the first status data into the transmission influence function to obtain the transmission influence factor, and the expression is: Among them is the transmission influence factor, , , are the weight coefficients of the transmission influence factor, is the deviation between the actual frequency and the preset frequency of the emitted laser, is the preset frequency of the emitted laser, is the temperature drift of the laser, is the temperature-wavelength influence coefficient, is the deviation between the end value of the laser spectrum and the atmospheric spectrum, is the deviation between the start value of the laser spectrum and the atmospheric spectrum, is the range size of the atmospheric spectrum; Determine the second laser measurement length according to the transmission influence factor and the first laser measurement length, and the expression is: Among them is the second laser measurement length is the first laser measurement length is the transmission influence factor weight 4. The high-precision length measurement method based on laser technology according to claim 1, characterized in that The method for obtaining the pixel edge coordinates includes: Use the OpenCV image processing library to read the camera image to be processed, perform grayscale processing on the camera image to obtain a grayscale image, and perform edge detection on the grayscale image to identify the edge position of the grayscale image of the workpiece to be measured; The edge detector selects a Laplacian of Gaussian filter to perform Gaussian filtering on the grayscale image and perform Laplacian transformation to obtain image edge information. The expression is as follows: where is the Laplacian of Gaussian operator, is the Laplacian operator, is the two-dimensional Gaussian function, is the standard deviation of the Gaussian function, is the image after Laplacian of Gaussian filtering processing, is the input grayscale image; Zero-crossing detection is performed based on the edge information to obtain the specific edge positions. According to the edge positions, the image is contour-traced in the counterclockwise direction using an eight-connected region. The expression for the eight-connected region is as follows: Among them represents the starting point of boundary tracking of the eight-connected region, including the point and the corresponding upper, lower, left, right, and four diagonal neighboring points represents the four-connected region of the point including the point and the corresponding upper, lower, left, and right neighboring points; Five distance intervals with different distances are divided according to the calculated grayscale values to screen the contour points. The screening principle is as follows: Among them , , , are the average gray values of regions 1, 2, 4, and 5, is the gray threshold; The determined screened contour points are used to calculate the geometric center using the moment function, and the pixel edge coordinates are extracted according to the contour data.

5. The high-precision length measurement method based on laser technology according to claim 1, wherein The method for converting the pixel edge coordinates into world edge coordinates includes: The second state data is the camera focal length, camera pitch angle, and translation amount; Construct the camera intrinsic matrix and extrinsic matrix based on the second state data, and convert the raw edge coordinates into camera edge coordinates , and convert the camera edge coordinates into world edge coordinates .

6. The high-precision length measurement method based on laser technology according to claim 1, characterized in that The method for obtaining the first edge coordinates includes: Calculate the comprehensive similarity between the world edge coordinates and the laser edge coordinates. The expression is as follows: Among them is the comprehensive similarity between the world edge coordinates and the laser edge coordinates, and are weight parameters, is the th world edge coordinate point and the th laser edge coordinate point 's Euclidean distance, is the number of coordinate points, is the buffer radius, is the distance from the world edge coordinate point to the origin, is the distance from the laser edge coordinate point to the origin, is the mean value of the corresponding vectors of the world edge coordinate points, is the mean value of the corresponding vectors of the laser edge coordinate points; Match the edge coordinate points according to the comprehensive similarity between the world edge coordinates and the laser edge coordinates, and determine the first edge coordinates according to the mean, variance, and skewness of the two edge coordinate points.

7. The high-precision length measurement method based on laser technology according to claim 1, characterized in that, The method for obtaining the length of the laser image includes: Calculate the distance and angle between any two first edge coordinate points, and define two first edge coordinate points and the corresponding distance and angle as a set of edge coordinate data; Perform feature fusion on the edge coordinate data and the second laser measurement length to obtain the length of the laser image; Adopt an interactive multi-feature fusion method and a stacked ensemble classifier for feature fusion; Interactive multi-feature fusion method It includes: calculating the correlation degree, calculating the connection entropy, calculating the interaction factor, and determining the evaluation criteria; Determine the evaluation criteria for candidate features. The expression is as follows: Among them is the evaluation criterion for candidate features, is all features, is is the correlation degree between the second laser measurement length and the edge coordinate data feature is the redundancy, is the redundancy, is the selected feature subset is the feature in is the interaction factor, is considering the second laser measurement length when, the selected feature subset and the edge coordinate data feature is the joint entropy of is only considering the selected feature subset and the edge coordinate data feature is the joint entropy; Obtain the optimal feature subset according to the evaluation criteria for candidate features, and input the optimal feature subset into the stacked ensemble classifier to obtain the length of the laser image; The stacked ensemble classifier includes an input layer, a first classification layer, a second classification layer, and an output layer; The first classification layer is composed of a deep neural network DNN and a lightweight gradient boosting machine LightGBM, which perform basic classification on the input data and output probability results; the second classification layer is composed of a support vector machine SVM, which performs meta-classification on the input probability results to determine the fusion result of the feature data, and the output layer outputs the comprehensive measurement length; the performance of the model is evaluated using the F1 score, G-mean, mean squared error MSE, and cross-validation.

8. The high-precision length measurement method based on laser technology according to claim 1, wherein The method for obtaining the measurement length includes: Adjust the length of the laser image of different spatial position scanning groups according to the ranging scale to obtain the comprehensive measurement length, and form a comprehensive measurement set with the transmission influence factor, ranging scale, and first edge coordinates. The comprehensive measurement set is divided into a training set and a test set; Construct a high-precision length measurement model. The high-precision length measurement model includes an input layer, a base model layer, a strategy layer, and an output layer; The base model layer is composed of two multi-layer perceptron models and two gradient boosting machine models in parallel; In the multi-layer perceptron model: Design two hidden layers, use the ReLU function as the activation function to accelerate training and reduce gradient disappearance, select the Cross-Entropy Loss function to solve the problem that the cross-entropy loss function deviates from the evaluation index and overconfidence, and select Adam to optimize the model weight parameters; In the gradient boosting machine model: Build a mathematical model through an iterative method, use the logarithmic loss function to measure the difference between the predicted probability distribution and the true distribution, and adjust the learning rate to control the optimization process; The strategy layer uses the stacking method to input the prediction structure of the base model into a new model for prediction to obtain the prediction result, and uses the test set to evaluate the high-precision length measurement model; Input the working data and status data of the machine to be measured into the high-precision length measurement model to obtain the measured length.

9. A high-precision length measurement system based on laser technology for performing the method according to any one of claims 1-8, characterized in that, It includes: Data acquisition module: including a laser emitter, a laser receiver and a high-resolution camera; It is used to collect the working data and status data of the machine and preprocess the working data and the status data; Data processing module: used to calculate the laser edge coordinates, the first laser measurement length and the ranging scale according to the laser scan data, used to obtain the transmission influence factor, used to determine the second laser measurement length, and used to obtain the comprehensive measurement length according to the ranging scale; Fusion model module: used to perform feature fusion on the edge coordinate data and the first laser measurement length by using an interactive multi-feature fusion method and a stacked ensemble classifier; Measurement model module: used to construct a high-precision length measurement model according to the comprehensive measurement length, the ranging scale, the transmission influence factor and the first edge coordinates, and input the working data and status data to be measured into the high-precision length measurement model to obtain the measured length; Intelligent supervision module: used to store, view and manage the working data, the status data and the measured length.

Images