Intelligent laser triangulation ranging method

Abstract

The invention relates to an intelligent laser triangulation ranging method, which comprises the following steps of: acquiring an imaging light spot image of a measured object, and preprocessing the imaging light spot image to form a data set of light spot image coordinates; an improved neural network model is constructed, and the nonlinear relation between the light spot position and the displacement of the measured object can be effectively captured; based on the coordinate difference data between the light spots in the training set, training the improved neural network model to obtain a trained improved neural network model; inputting coordinate difference data between light spots in a test set into the trained improved neural network model, and outputting a prediction result; and inputting normalized imaging light spot position information, and carrying out reverse normalization processing on a prediction result output by the trained improved neural network model to obtain an actual displacement distance of the measured object. According to the invention, accurate measurement of displacement parameters of objects with complex surface morphology and rapid movement is realized, and application requirements of product quality control, process monitoring and the like under high-precision requirements are met.

Description

Technical Field

[0001] The invention belongs to the field of measurement technology and instruments, and relates to an intelligent laser triangulation distance measurement method. Background Art

[0002] At present, laser triangulation systems still rely on traditional digital image processing algorithms (grayscale centroid method, Gaussian fitting method and ellipse fitting method) to locate the imaging spot in order to complete the ranging function. In actual measurements, traditional algorithms are sensitive to environmental noise and it is difficult to maintain stable measurement accuracy under complex lighting conditions, especially when there are interference signals or uneven reflective surfaces. The accuracy and reliability of the measurement results will be significantly affected. Secondly, traditional methods rely on manual adjustment of system structure parameters multiple times to optimize the measurement effect. The whole process is cumbersome and difficult to adapt to different industrial measurement scenarios, which reduces the versatility and operational efficiency of the system. The above defects make it difficult for traditional laser triangulation systems to meet the needs of modern industry for real-time, high-precision and intelligent measurement. Summary of the invention

[0003] In order to solve the above problems, the technical solution adopted by the present invention is: an intelligent laser triangulation ranging method, comprising the following steps:

[0004] Acquire an imaging spot image of the object under test;

[0005] Preprocessing the imaging spot image to form a data set of spot image coordinates;

[0006] An improved neural network model is constructed to effectively capture the nonlinear relationship between the light spot position and the displacement of the object being measured;

[0007] Based on the coordinate difference data between the light spots in the training set, the improved neural network model is trained to obtain a trained improved neural network model;

[0008] Input the coordinate difference data between the light spots in the test set into the trained improved neural network model and output the prediction results;

[0009] The normalized imaging spot position information is input, and then the prediction result output by the trained improved neural network model is denormalized to obtain the actual displacement distance of the measured object.

[0010] Furthermore, the imaging spot image is collected by a direct laser triangulation measurement system of the imaging spot image.

[0011] Furthermore, the imaging spot image is preprocessed by using the grayscale centroid method, Gaussian fitting method and ellipse fitting method to locate the imaging spot collected.

[0012] Furthermore, the improved neural network includes a structure of: input layer-fully connected layer-output layer, and a BP neural network using a stochastic gradient descent momentum method.

[0013] Furthermore, the training of the improved neural network model corresponds to the upward movement and downward movement of the object under test relative to the calibration reference plane. Specifically, the object under test is placed on the reference plane, and it is first moved up and the imaging spot is collected, and then it is moved down and the imaging spot is collected, and then the upward and downward movement training is performed in sequence.

[0014] Furthermore, the fully connected layer has 12 layers, each of the 1-11 layers has 550 nodes, the 12th layer has only one neuron, and the output layer has one node, corresponding to the displacement information of the object being measured.

[0015] An intelligent laser triangulation ranging device, comprising:

[0016] Acquisition module: used to acquire the imaging spot image of the object to be measured, pre-process the imaging spot image, and form a data set of spot image coordinates;

[0017] Building module: used to build an improved neural network model, which can effectively capture the nonlinear relationship between the spot position and the displacement of the object under test;

[0018] Training module: used to train the improved neural network model based on the coordinate difference data between the light spots in the training set to obtain a trained improved neural network model;

[0019] Prediction module: used to input the coordinate difference data between the light spots in the test set into the trained improved neural network model and output the prediction results;

[0020] Obtaining module: used to input the normalized imaging spot position information, and then perform denormalization processing on the prediction results output by the trained improved neural network model to obtain the actual displacement distance of the object being measured.

[0021] A computer device comprises: a processor and a memory, wherein the memory stores a program module, and wherein the program module runs on the processor to implement any of the methods described above.

[0022] A readable storage medium stores a program module, wherein the program module is executed in a processor to implement any of the methods described above.

[0023] An intelligent laser triangulation ranging system, comprising:

[0024] Linear translation stage: used to fix the object to be measured;

[0025] Image acquisition sensor: used to collect images and imaging spots; when the object to be measured moves, the position of the imaging spot on the CMOS photosensitive surface also changes accordingly; each position of the object to be measured within the measuring range corresponds to the position of the imaging spot on the CMOS;

[0026] Intelligent laser triangulation distance measuring device: used for measuring the actual displacement distance of the object to be measured based on the image collected by the image acquisition sensor and using the method described.

[0027] The present invention provides an intelligent laser triangulation ranging method for improving the measurement accuracy and calculation efficiency of the laser triangulation ranging system. A nonlinear programming genetic algorithm is used to determine the optimal structural parameters of the system, and a direct laser triangulation measurement system is built to achieve high-resolution distance measurement. The three digital image processing algorithms of grayscale centroid method, Gaussian fitting method and ellipse fitting method are used to respectively locate the imaging spot, and the BP neural network is further used to fuse the spot position to directly determine the displacement of the object being measured. The laser triangulation ranging system of the present invention does not need to use a ranging formula based on system parameters, thereby avoiding the measurement error and uncertainty caused by structural parameters, and has significant improvements in measurement accuracy, stability, robustness and intelligence, and is suitable for industrial scenarios that require high-precision, non-contact real-time measurement.

[0028] In industrial manufacturing scenarios, it can be used for high-precision, non-contact, real-time detection of metal and other material surfaces. By integrating deep learning and laser triangulation technology, the present invention can achieve accurate measurement of parameters such as complex surface topography and displacement of fast-moving objects, meeting the application requirements of product quality control and process monitoring under high-precision requirements, and is widely applicable to industrial scenarios that require precision measurement, such as semiconductor manufacturing, automobile production, and precision machining.

[0029] Beneficial effects of the present invention

[0030] 1. Reduction of measurement error and relative uncertainty: With the spot positioning data of grayscale centroid method, Gaussian fitting method and ellipse fitting method as feature input, the trained BP neural network can effectively learn the complex nonlinear relationship between the imaging spot position and the displacement of the measured object. The measurement system no longer needs to use the distance measurement formula based on structural parameters, thus avoiding the measurement error and uncertainty caused by these parameters.

[0031] 2. Enhanced system robustness and adaptability: Compared with the traditional spot positioning method, the present invention significantly enhances the measurement system's resistance to noise and environmental changes through the adaptive ability of the BP neural network. The spot is pre-processed by a variety of traditional algorithms and then input into the neural network model, which further improves the stability and robustness of the system, making the laser triangulation system applicable to various complex industrial measurement environments.

[0032] 3. High degree of intelligence and automation: The present invention integrates BP neural network and traditional spot positioning algorithm, which greatly improves the automation of the system measurement process. The system can automatically adjust and optimize the measurement strategy according to the input spot data, has self-learning ability, reduces dependence on manual intervention, and improves the level of intelligence in industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0034] Figure 1 A flowchart of the method;

[0035] Figure 2 It is the principle of laser triangulation distance measurement;

[0036] Figure 3 It is an elliptical speckle image;

[0037] Figure 4 It is the structure diagram of BP neural network model;

[0038] Figure 5 It is the training flow chart of BP neural network. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The invention is a method based on the principle of direct laser triangulation measurement, which introduces BP neural network to fuse three traditional imaging spot positioning algorithms to achieve non-contact, high-precision and real-time measurement of the displacement of the object to be measured.

[0042] Figure 1is a flow chart of the method;

[0043] An intelligent laser triangulation ranging method comprises the following steps:

[0044] S1: Acquire the imaging spot image of the object under test;

[0045] S2: Preprocess the imaging spot image to form a data set of spot image coordinates;

[0046] S3: Construct an improved neural network model that can effectively capture the nonlinear relationship between the spot position and the displacement of the object being measured;

[0047] S4: Based on the coordinate difference data between the light spots in the training set, the improved neural network model is trained to obtain a trained improved neural network model;

[0048] S5: input the coordinate difference data between the light spots in the test set into the trained improved neural network model, and output the prediction results;

[0049] S6: Input the normalized imaging spot position information, and then perform denormalization processing on the prediction result output by the trained improved neural network model to obtain the actual displacement distance of the object being measured.

[0050] The steps S1 / S2 / S3 / S4 / S5 are performed sequentially;

[0051] Figure 2 It is the principle of laser triangulation distance measurement;

[0052] Figure 3 It is an elliptical speckle image;

[0053] Build a measurement system for imaging spot images; under the constraints of Scheimpflug imaging law, imaging spot size, and mechanical structure geometry, use the system resolution as the objective function to determine the structural parameters of the laser triangulation system through a nonlinear programming genetic algorithm (E-NPGA). Build a direct laser triangulation system, where the laser is collimated by a plano-convex lens and vertically incident on the surface of the object to be measured for diffuse reflection, and the scattered light is focused on the CMOS photosensitive surface by an imaging lens.

[0054] The direct laser triangulation system has a resolution of Or system range As the objective function of N-GA;

[0055] Where c is the pixel size of the image sensor, The object distance when the diffuse reflected light of the object is focused by the imaging lens when the object is on the calibration reference plane. is the angle between diffuse reflected light and the image sensor, α is the angle between diffuse reflected light of the reference plane and the incident light, H is the height of the imaging lens from the reference plane, and f is the ideal focal length of the imaging lens. It is worth noting that α, H, and f are the basic variables of the entire laser triangulation system. As long as these three parameters are determined, the entire optical system can be determined.

[0056] The three constraints of N-GA are determined according to different measurement scenarios:

[0057] (1) To ensure that the light spot can always be clearly imaged on the image sensor during the movement of the object under test, the Scheimpflug imaging law L1*tanα=L′1*tanβ must be satisfied, where It is the image distance of the diffusely reflected light from the reference surface during the focusing process of the imaging lens.

[0058] (2) To ensure that the performance of the measurement system is not affected, it is necessary to limit the size of the elliptical imaging spot on the image sensor, where the size in the X direction on the image sensor is (vertical direction of light spot movement), the Y-direction size is (the direction of movement of the light spot), where r is the radius of the incident light spot on the reference plane. (3) Different measurement scenarios leave different spaces for the laser triangulation system, so it is very important to constrain the geometric dimensions of the system's mechanical structure.

[0059] Since the positions of the collimating lens, image sensor, and imaging lens are uncertain, the size formula of the system's mechanical structure is discussed in three cases. First, if the image sensor is lower than the collimating lens, the size formula of the measurement system is W = (L1 + L′1) cos α, R = VH. Secondly, if the imaging lens is higher than the collimating lens, the size formula of the measurement system is W = (L1 + L′1) cos α, R = (L1 + L′1) cos α-V. Finally, if the image sensor is higher than the collimating lens, and the imaging lens is lower than the collimating lens, the size formula is W = (L1 + L′1) cos α, R = (L1 + L′1) cos α-H. Among them, W is the lateral size of the system, R is the longitudinal size of the system, and V is the height of the collimating lens from the collimating surface;

[0060] The first generation of population is randomly generated under the premise of satisfying all constraints, and the fitness is evaluated according to the objective function and the best chromosome of the first generation is recorded. Entering the evolution stage, six evolutionary orders of selection, crossover and mutation are randomly determined to start evolution.

[0061] After each evolution is completed, the elitist strategy is executed once, and the best chromosome in the current population is replaced with the best chromosome in the previous generation. The best chromosome of this generation after evolution is compared with the current best chromosome and recorded. After a certain number of generations of evolution, the nonlinear programming function is called once to perform local optimization with the current best chromosome as the initial value. After optimization, the elitist strategy is executed again to improve the local search ability of the traditional genetic algorithm.

[0062] This cycle repeats itself, and after a certain number of generations of evolution, the initial structural parameters of the system that meet the system resolution or range requirements are finally found: the angle α between the diffuse reflected light and the incident light corresponding to the reference plane, the height H of the imaging lens from the reference plane, and the focal length f of the imaging lens. The evolutionary trends of the system resolution and system range during the N-GA evolution process are also demonstrated.

[0063] Optical simulation software is used to simulate the system structure parameters provided by N-GA. The accuracy of the initial structure parameters is verified by calculating the relative error between the RMS radius of the light spot on the image sensor and the theoretical value of the imaging spot size.

[0064] Based on the direct laser triangulation measurement system, the displacement formula of the measured object is determined as follows:

[0065]

[0066] Where S' is the interval value of the imaging spot on the CMOS photosensitive surface, and S is the displacement of the object to be measured. When the object to be measured moves upward along the optical axis relative to the reference plane, the denominator of the range formula is "+". Otherwise, the denominator of the range formula is "-".

[0067] The imaging spot image is preprocessed by using the grayscale centroid method, Gaussian fitting method and ellipse fitting method to locate the imaging spot collected.

[0068] The basic idea of ​​the grayscale centroid method is to regard the imaging spot as a two-dimensional grayscale distribution map, regard the grayscale value of each pixel as the mass of the point, and locate the position of the imaging spot by calculating the centroid of these grayscale values.

[0069] (2) Gaussian fitting method

[0070] Gaussian fitting method: Under ideal conditions, the light intensity of the imaging spot collected on the CMOS photosensitive surface satisfies the Gaussian distribution, that is, the intensity of the imaging spot decreases smoothly from the center of the spot to the outside. The Gaussian function is used to fit the spot, and the position of the intensity amplitude is the position of the center of the spot;

[0071] (3) Ellipse fitting method

[0072] Ellipse fitting is an image detection method based on the least squares method. It searches for the optimal ellipse parameters by minimizing the sum of the squares of the distances between the fitted ellipse and the edge points of the actual light spot, thereby accurately determining the center position of the imaging light spot.

[0073] Preprocessing process: (1) Read the imaging spots in sequence; (2) Use the grayscale centroid method, Gaussian fitting method, and ellipse fitting method to determine the spot position coordinates; (3) Each imaging spot has three sets of coordinates (x1, y1), (x2, y2), and (x3, y3), and calculate the difference between the three sets of coordinates and the reference spot coordinates; (4) The difference is then normalized and becomes the input data of the BP neural network.

[0074] The three traditional processing algorithms have their own advantages and disadvantages and are suitable for different industrial measurement scenarios. The combination of the three algorithms can maximize the ranging accuracy of the laser triangulation system. The imaging spot position data obtained by the three algorithms is used as a feature quantity and input into the subsequent neural network training model.

[0075] Figure 4 It is the structure diagram of BP neural network model;

[0076] Figure 5 It is the training flow chart of BP neural network.

[0077] The improved neural network includes a structure of: input layer-fully connected layer-output layer, and a BP neural network using the stochastic gradient descent momentum method (SGDM). The activation function uses ReLU (Rectified Linear Unit), and the loss function uses mean square error (MSE).

[0078] The imaging spot position data after Min-Max normalization is used as the input of the BP neural network for training. The training of the improved neural network model corresponds to the upward movement and downward movement of the measured object relative to the calibration reference plane.

[0079] This is a BP neural network training method specifically for laser triangulation; (1) The relative distance measured by laser triangulation is the movement of the object under test relative to the reference plane; (2) When the object under test is on the calibration reference plane, the corresponding imaging spot is located at the center of the CMOS photosensitive surface; (3) When collecting the training spot: ensure that the object under test is on the reference plane, first move it up and collect the imaging spot, then move it down and collect the imaging spot. Then perform the up-and-down training in sequence.

[0080] The fully connected layer has 12 layers, with 550 nodes in each layer from 1 to 11, only one neuron in the 12th layer, and one node in the output layer, corresponding to the displacement information of the object being measured.

[0081] Furthermore, the actual displacement data of the object being measured combined with the acquisition frame rate of the measurement system CMOS can realize real-time monitoring of the movement of the object being measured or restoration of the surface morphology.

[0082] During the continuous movement of the object under test, the CMOS will continuously collect imaging spots: (1) If the acquisition frame rate of the CMOS is known, the acquisition interval between the two spots can be determined. The moving distance of the object under test divided by the acquisition interval is the moving speed of the object under test; (2) Or during the lateral movement of the object under test with surface undulations, as long as its lateral movement speed is known, the lateral distance between the surface undulations can be determined to achieve surface morphology restoration.

[0083] An intelligent laser triangulation ranging device, comprising:

[0084] Acquisition module: used to acquire the imaging spot image of the object to be measured, pre-process the imaging spot image, and form a data set of spot image coordinates;

[0085] Building module: used to build an improved neural network model that can effectively capture the nonlinear relationship between the spot position and the displacement of the object being measured;

[0086] Training module: used to train the improved neural network model based on the coordinate difference data between the light spots in the training set to obtain a trained improved neural network model;

[0087] Prediction module: used to input the coordinate difference data between the light spots in the test set into the trained improved neural network model and output the prediction results;

[0088] Obtaining module: used to input the normalized imaging spot position information, and then perform denormalization processing on the prediction results output by the trained improved neural network model to obtain the actual displacement distance of the object being measured.

[0089] An intelligent laser triangulation ranging system, comprising:

[0090] Linear translation stage: used to fix the object to be measured; (1) through the precise forward and backward movement of the translation stage, the movement distance of the object to be measured relative to the calibration reference plane is determined; (2) it is also the source of label data in the BP neural network training process.

[0091] CMOS image acquisition sensor: used to collect images and imaging spots; when the object to be measured moves, the position of the imaging spot on the CMOS photosensitive surface also changes accordingly; each position of the object to be measured within the measuring range corresponds to the position of the imaging spot on the CMOS;

[0092] Intelligent laser triangulation distance measuring device: used for measuring the actual displacement distance of the object to be measured based on the image collected by the image acquisition sensor and using the method described.

[0093] A computer device comprises: a processor and a memory, wherein the memory stores a program module, and the program module runs on the processor to implement any one of the methods described above.

[0094] A readable storage medium stores a program module, wherein the program module is executed in a processor to implement any of the methods described above.

[0095] Embodiment 1: This embodiment is based on the principle of laser triangulation and deep neural network, and the method for measuring the distance of the object to be measured is: projecting the collimated Gaussian light beam vertically onto the surface of the object to be measured; using the reflected light image acquisition module to collect the light spot on the surface of the object to be measured, and obtaining the diffuse reflected light image corresponding to the object to be measured on each calibration surface; after pre-processing the imaging light spot, input it into the BP neural network for model training to achieve accurate measurement of the distance of the object to be measured.

[0096] 1. System Construction

[0097] 1. Build a laser triangulation measurement system with a resolution of 5um. The system structure parameters are as follows: the ideal focal length of the collimating lens is f=2mm, 2mm from the laser fiber port, 70mm from the calibration reference plane of the object to be measured, the angle between the diffuse reflected light of the reference plane and the incident light is α=31.14°, the height of the imaging lens from the reference plane is H=59.27mm, the ideal focal length of the imaging lens is f=25mm, the inclination angle of the CMOS photosensitive surface is β=46.92°, and it is 35.93mm away from the imaging lens.

[0098] 2. The metal object to be measured is fixed on a linear displacement stage (resolution of 0.018um) to precisely control the displacement of the object. The acquisition frequency of the CMOS image sensor is set to 100Hz to ensure that during the acquisition of the imaging spot, the position of the spot always corresponds to the displacement information of the object to be measured.

[0099] 2. BP neural network training process

[0100] 3. Build a BP neural network model, which adopts the structure of input layer-fully connected layer-output layer. The input layer contains 3 nodes, corresponding to the position information of the imaging spot; the fully connected layer has 12 layers, and each layer has 550 nodes from 1 to 11; the 12th layer has 1 neuron node, and the output layer has 1 node, corresponding to the displacement information of the object being measured. The activation function selects ReLU and Sigmoid, the loss function is the mean square error (MSE), and the optimization algorithm is the stochastic gradient descent momentum method (SGDM).

[0101] 4. Use the grayscale centroid method, Gaussian fitting method, and ellipse fitting method to pre-position the imaging spot collected by CMOS. The spot position data is then normalized by Min-Max to eliminate the influence of different dimensions, keep the eigenvalues ​​in the same numerical range, and improve the training speed and generalization ability of the BP neural network model.

[0102] 5. The normalized spot positioning data is input into the BP neural network for training. The training process is divided into two parts, corresponding to the upward and downward movements of the object under test based on the calibration reference plane. The training parameters for upward movement are: the learning rate is set to 0.003, and the training lasts for 80 rounds; then incremental learning is performed, the learning rate is reduced to 0.002, the learning rate attenuation cycle is 10, the attenuation factor is 0.5, and the training lasts for 100 rounds. The training parameters for downward movement are: the learning rate is set to 0.002, and the training lasts for 100 rounds; then incremental learning is performed, the learning rate is further reduced to 0.0015, and the training lasts for 200 rounds. After each round of training, the data is randomly rearranged.

[0103] 6. The BP neural network model was tested using the training set and validation set to evaluate its performance in the actual measurement process. The results are as follows:

[0104]

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent laser triangulation ranging method, characterized in that: The following steps are involved: Acquire an imaging spot image of the object under test; Preprocessing the imaging spot image to form a data set of spot image coordinates; An improved neural network model is constructed to effectively capture the nonlinear relationship between the light spot position and the displacement of the object being measured; Based on the coordinate difference data between the light spots in the training set, the improved neural network model is trained to obtain a trained improved neural network model; Input the coordinate difference data between the light spots in the test set into the trained improved neural network model and output the prediction results; The normalized imaging spot position information is input, and then the prediction result output by the trained improved neural network model is denormalized to obtain the actual displacement distance of the measured object.

2. The intelligent laser triangulation ranging method according to claim 1, characterized in that: The imaging spot image is collected by a direct laser triangulation measurement system of the imaging spot image.

3. The intelligent laser triangulation ranging method according to claim 1, characterized in that: The imaging spot image is preprocessed by using the grayscale centroid method, Gaussian fitting method and ellipse fitting method to locate the imaging spot collected.

4. The intelligent laser triangulation ranging method according to claim 1, characterized in that: The improved neural network includes a structure of: input layer-fully connected layer-output layer, and a BP neural network using a stochastic gradient descent momentum method.

5. The intelligent laser triangulation ranging method according to claim 1, characterized in that: The training of the improved neural network model corresponds to the upward movement and downward movement of the object under test relative to the calibration reference plane. Specifically, the object under test is placed on the reference plane, and is first moved upward and an imaging spot is collected, and then moved downward and an imaging spot is collected, and then upward and downward training is performed in sequence.

6. The intelligent laser triangulation ranging method according to claim 3, characterized in that: The fully connected layer has 12 layers, with 550 nodes in each layer from 1 to 11, only one neuron in the 12th layer, and one node in the output layer, corresponding to the displacement information of the object being measured.

7. An intelligent laser triangulation ranging device, characterized in that: include: Acquisition module: used to acquire the imaging spot image of the object to be measured, pre-process the imaging spot image, and form a data set of spot image coordinates; Building module: used to build an improved neural network model, which can effectively capture the nonlinear relationship between the spot position and the displacement of the object under test; Training module: used to train the improved neural network model based on the coordinate difference data between the light spots in the training set to obtain a trained improved neural network model; Prediction module: used to input the coordinate difference data between the light spots in the test set into the trained improved neural network model and output the prediction results; Obtaining module: used to input the normalized imaging spot position information, and then perform denormalization processing on the prediction results output by the trained improved neural network model to obtain the actual displacement distance of the object being measured.

8. A computer device comprising: A processor and a memory, wherein the memory stores a program module, wherein the program module runs on the processor to implement the method according to any one of claims 1 to 6.

9. A readable storage medium storing a program module, characterized in that: The program module is executed in a processor to implement the method according to any one of claims 1 to 6.

10. An intelligent laser triangulation ranging system, characterized in that: include: Linear translation stage: used to fix the object to be measured; Image acquisition sensor: used to collect images and imaging spots; when the object to be measured moves, the position of the imaging spot on the CMOS photosensitive surface also changes accordingly; each position of the object to be measured within the measuring range corresponds to the position of the imaging spot on the CMOS; Intelligent laser triangulation distance measuring device: used for measuring the actual displacement distance of the object to be measured based on the image collected by the image acquisition sensor and using the method described in claims 1-6.

Images