Industrial machine vision measurement system calibration method and device

Abstract

The invention relates to the technical field of machine vision, in particular to a calibration method and device for an industrial machine vision measurement system, and the method comprises the following steps: collecting error data through a glass linear scale and a circular target, constructing a standardized calibration data set, obtaining a global error gradient field, a residual error trend map and a residual error weight coefficient table, and obtaining a standard calibration data set; and fusing the size compensation value and the directivity correction value, constructing a calibration matrix, and obtaining a system calibration result. According to the method, through fusion modeling of multi-source error data, the recognition precision of the system on error sources is improved, gradient field construction and trend graph extraction are combined, the directionality and spatial continuity of error compensation are enhanced, and a weight coefficient distribution mechanism is utilized, so that differential control on dynamic and static error influence is realized; and a calibration matrix with multi-dimensional parameter fusion is adopted to optimize the consistency and adaptability of the overall measurement precision, so that the system keeps stable error correction capability and calibration accuracy in a complex environment.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision, and in particular to a calibration method and device for an industrial machine vision measurement system. Background Art

[0002] The field of machine vision technology includes image processing, pattern recognition, automatic control and other technologies, which are widely used in electronics, logistics, chemical industry and other industries for automated detection, measurement and control. The industrial machine vision system is composed of industrial cameras, lenses, light sources, computers and related software. It focuses on simulating human visual functions and quickly extracting target features such as size, position and angle for precise measurement and detection. The core is to analyze and calculate the collected images through image processing technology, judge the state or characteristics of the target object, provide decision-making basis and improve production efficiency and product quality.

[0003] Among them, the industrial machine vision measurement system calibration method refers to the effective calibration of the measurement errors existing in the existing machine vision measurement system through a variety of technical means to ensure the measurement accuracy of the system. The patent subject mainly involves a method for systematically calibrating the industrial machine vision system to solve the measurement error problem caused by environmental changes, equipment aging and other factors during the use of the existing vision system. The method designs a special calibration device and calibration procedure, implements regular or on-demand calibration for specific equipment performance and working environment, ensures the accuracy of the measurement results output by the system under various operating conditions, and improves the stability and reliability of the system.

[0004] Traditional industrial machine vision measurement system calibration technology relies on the error mean comparison and correction method under a fixed path, lacks the identification of residual structure and directional error, and lacks clear distinction and processing rules for various error sources during the calibration process, resulting in compensation offset when the scene switches or the device status changes. Since the calibration data structure is relatively flat, the collection points are sparse and there is no cross-scene comparison mechanism, it is difficult to identify the directional error changes under dynamic interference. In the presence of equipment vibration, light fluctuations and other conditions, the stability of the measurement results decreases, including the failure to introduce directional correction factors when measuring in edge areas. Errors often show clustered offsets, affecting the overall calibration accuracy. The error weight processing has not formed a differentiated judgment mechanism, and the data of static and dynamic measurement points have the same weight in the compensation processing, resulting in the system's fuzzy judgment on different error sensitivities, reducing the pertinence and effectiveness of error correction. Summary of the invention

[0005] In order to solve the technical problems existing in the prior art, the embodiment of the present invention provides an industrial machine vision measurement system calibration method and device. The technical solution is as follows: In order to achieve the above object, the present invention adopts the following technical solution, a method for calibrating an industrial machine vision measurement system, comprising the following steps: S1: Using a glass line ruler, collect multiple dimensional error measurement values ​​at multiple calibration positions and calculate the average value. Using a circular target image, synchronously collect the residual extreme values ​​of multiple measurement points in dynamic scenes and the residual mean value in static scenes to obtain a standardized calibration data set. S2: calling the standardized calibration data set, using the size error data set, calculating the average error value of each calibration position, analyzing the gradient change and calculating the smoothing correction parameter, using discrete points to construct a continuous error field of the entire workbench, and generating a global error gradient field; S3: calling the standardized calibration data set, using the dynamic residual data set, taking the fitting circle center as the origin, dividing the residual distribution area into multiple sector units according to a fixed angle, aggregating the residual values ​​in each sector, calculating the sector residual range, analyzing the radial distance change trend of the measuring point, and generating a residual trend map; S4: calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, calculating the weight coefficients of the dynamic residual and the static residual by analyzing the influence of the dynamic residual fluctuation data and the static residual stability on the system error, and generating a residual weight coefficient table.

[0006] As a further solution of the present invention, the standardized calibration data set includes a size error mean set, a dynamic residual range set, and a static residual mean set; the global error gradient field is specifically an error change direction layer, a smoothing correction parameter layer, and a path continuity fitting layer; the residual trend map specifically refers to a radial expansion trend layer, a sector range distribution layer, and an angle mapping layer; the residual weight coefficient table includes a dynamic error response weight, a static error stability weight, and a directional sensitivity coefficient.

[0007] As a further solution of the present invention, a glass linear ruler is used to collect multiple dimensional error measurement values ​​at multiple calibration positions and calculate the average value. A circular target image is used to synchronously collect the residual extreme values ​​of multiple measurement points in a dynamic scene and the residual mean value in a static scene. The specific steps for obtaining a standardized calibration data set are: S101: using a glass line ruler, perform multiple dimension measurements at each calibration position, call multiple groups of measurement values ​​corresponding to each position, calculate the average difference between each group of measurement values, and establish a set of dimension error average values ​​in combination with the position number; S102: calling the size error average value set, using the circular target image, collecting the numerical extreme difference of each measuring point in the dynamic scene, obtaining the numerical mean of each group of measuring points in the static scene, and generating a residual data group of multi-scene measuring points; S103: According to the multi-scenario measurement point residual data group, combined with the measurement position information, the dimensional error value, the dynamic residual value, and the static residual value are standardized, and measurement data sets are respectively constructed, including a dimensional error data set, a dynamic residual data set, and a static residual data set, to obtain a standardized calibration data set.

[0008] As a further solution of the present invention, the standardized calibration data set is called, the size error data set is used to calculate the average error value of each calibration position, the gradient change is analyzed and the smoothing correction parameter is calculated, and the continuous error field of the entire workbench is constructed using discrete points. The steps of generating the global error gradient field are specifically as follows: S201: calling the standardized calibration data set, using the dimension error data set, extracting each set of measured coordinate values ​​and error values, calculating the average error value of each set of coordinate points, establishing a corresponding relationship between the error value and the coordinate number in combination with the position information, and generating a coordinate error mean value list; S202: According to the coordinate error mean value list and the error change trend of adjacent coordinate points, the gradient change rate of the error data at each position is obtained, and a smoothing correction parameter is calculated; S203: calling the smoothing correction parameter, combining with spatial interpolation, constructing a continuous error field of the entire workbench, and obtaining a global error gradient field.

[0009] As a further solution of the present invention, the specific formula for constructing the continuous error field of the entire workbench is: ; Calculate the coordinate interpolation prediction value; in, Represents coordinate point The error interpolation prediction value of Representative Smoothing correction parameters for each measuring point, Representative The residual weight factor of each measurement point is Representative Measuring points and target points The Euclidean distance of represents the smoothing control factor, Represents the total number of measurement points involved in the interpolation calculation. Represents the measurement point number, Represents the X-axis coordinate value of the target interpolation point, Represents the Y-axis coordinate value of the target interpolation point.

[0010] As a further solution of the present invention, the standardized calibration data set is called, and the dynamic residual data set is used. The residual distribution area is divided into a plurality of sector units according to a fixed angle with the center of the fitting circle as the origin. The residual values ​​in each sector are aggregated, the sector residual range is calculated, and the radial distance variation trend of the measuring point is analyzed. The steps of generating the residual trend map are specifically as follows: S301: calling the standardized calibration data set, using the dynamic residual data set, extracting the coordinate position data of each measuring point relative to the center of the fitting circle, dividing the residual distribution area into a plurality of sector units in turn according to the coordinate angle of each measuring point, and generating a sector division coordinate group; S302: Based on the sector division coordinate group, aggregate all residual measurement point values ​​in each sector, record the residual extreme value corresponding to each sector, analyze and mark the residual fluctuation direction, and obtain the sector residual fluctuation parameter set; S303: According to the sector residual fluctuation parameter set, using the residual extreme value and fluctuation direction of each sector, combined with the radial distance data from each measuring point to the center of the circle, extract the radial distance change of the measuring point in each sector, analyze the radial distance change trend of the measuring point, and establish a residual trend map.

[0011] As a further solution of the present invention, the standardized calibration data set is called, the dynamic residual data set and the static residual data set are called, and the weight coefficients of the dynamic residual and the static residual are calculated by analyzing the influence of the dynamic residual fluctuation data and the static residual stability on the system error. The steps of generating the residual weight coefficient table are specifically as follows: S401: calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, analyzing the dynamic residual fluctuation intensity by calculating the extreme values ​​of multiple measuring points in the dynamic class, identifying the static residual stability by calculating the mean square error values ​​of multiple measuring points in the static class, and generating a residual characteristic data set; S402: Based on the residual characteristic data group, normalize the fluctuation intensity index and the stability index, and establish a residual influence analysis result by analyzing the influence of the fluctuation intensity index and the stability index on the visual measurement system error; S403: According to the residual impact analysis result, a dynamic residual coefficient and a static residual coefficient are set according to the impact degree, and a residual weight coefficient table is obtained.

[0012] As a further embodiment of the present invention, the method further comprises: S5: calling the global error gradient field, residual trend map and residual weight coefficient table, using the gradient field as the size error compensation reference value, using the trend map as the directional error correction value, combining the residual weight coefficient, dynamically adjusting the application intensity of the directional correction value, fusing the adjusted correction value and the size error compensation value according to the coordinate data, constructing a calibration matrix to calibrate the visual measurement system, and obtaining the system calibration result; The system calibration result specifically includes a coordinate mapping compensation value, a directional error adjustment coefficient, and a fusion correction output value.

[0013] As a further solution of the present invention, the global error gradient field, residual trend map and residual weight coefficient table are called, the gradient field is used as the size error compensation reference value, the trend map is used as the directional error correction value, and the application intensity of the directional correction value is dynamically adjusted in combination with the residual weight coefficient. According to the coordinate data, the adjusted correction value and the size error compensation value are fused, and a calibration matrix is ​​constructed to calibrate the visual measurement system. The steps of obtaining the system calibration result are specifically as follows: S501: calling the global error gradient field, residual trend map and residual weight coefficient table, calculating the basic compensation values ​​and directional correction values ​​of multiple coordinate points according to the size error compensation value and the directional error correction value, and generating a coordinate mapping error value group; S502: According to the coordinate mapping error value group, according to the dynamic and static residual weight values ​​corresponding to each coordinate point, the directional correction value is adjusted to obtain a correction adjustment application value set; S503: Based on the correction adjustment application value set, according to the coordinate data, the adjusted correction value and the size error compensation value are integrated, the error correction parameters of multiple positions are calculated, the calibration matrix is ​​constructed, and the visual measurement system is calibrated to obtain the system calibration result; The specific formula for calculating the error correction parameters for multiple positions is: ; Calculate the standardized error correction parameters after fusion; in, Representative The fusion standardized error correction parameter value of the coordinate position, represents the fusion weight coefficient, Representative Directional correction adjustment value for the coordinate position, Representative The size error compensation value of the coordinate position, is the total number of coordinates of the calibration points to be fused, The coordinate point number currently being calculated.

[0014] On the other hand, an industrial machine vision measurement system calibration device is provided, which is applied to an industrial machine vision measurement system calibration method, and the device includes: The data acquisition module is based on a glass line scale. It collects multiple dimensional error measurements at multiple calibration positions and calculates the average value. It uses a circular target image to synchronously obtain residual measurement data of multiple measurement points in dynamic and static scenes, extracts the residual range of dynamic measurement points and the residual variance of static measurement points, and generates a standardized calibration data set. The error field building module is based on the standardized calibration data set and uses the size error data set to extract the coordinates of multiple calibration points and the corresponding error averages, and performs gradient analysis according to the error variation amplitude and spatial spacing between adjacent points, calculates smoothing correction parameters, and fills the error data in the region in combination with spatial interpolation to establish a global error gradient field; The residual analysis module is based on the standardized calibration data set, uses the dynamic residual data set, calls the fitting circle center position as the angle division origin, divides the residual distribution area into multiple equal-angle sector units, aggregates the residual measurement points in multiple sectors and calculates the extreme difference, extracts the change trend in multiple sectors in combination with the radial distance information, and establishes a residual trend map; The weight calculation module is based on the standardized calibration data set, uses the dynamic residual data set and the static residual mean data set to normalize the two sets of residual data, analyzes the influence of the dynamic residual fluctuation data on the system error, analyzes the influence of the static residual stability on the system error, obtains the weight coefficients of the dynamic residual and the static residual, and obtains the residual weight coefficient table; The compensation fusion module calls the global error gradient field, residual trend map and residual weight coefficient table, extracts the error compensation value, direction correction value and application intensity coefficient of each coordinate point, fuses the direction correction value and size compensation value, constructs the calibration matrix parameters, calibrates the visual measurement system, and obtains the system calibration result.

[0015] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: Through the fusion modeling of multi-source error data, the system's identification accuracy of error sources is improved. Combined with gradient field construction and trend map extraction, the directionality and spatial continuity of error compensation are enhanced. The weight coefficient allocation mechanism is used to achieve differentiated control of the influence of dynamic and static errors. The calibration matrix with multi-dimensional parameter fusion is used to optimize the consistency and adaptability of the overall measurement accuracy, so that the system can maintain stable error correction capability and calibration accuracy in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2 It is a flow chart of the device of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0019] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0020] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0021] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0022] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0023] See also Figure 1 The present invention provides a technical solution, a method for calibrating an industrial machine vision measurement system, comprising the following steps: S1: Using a glass line ruler, collect multiple dimensional error measurement values ​​at multiple calibration positions and calculate the average value. Using a circular target image, synchronously collect the residual extreme values ​​of multiple measurement points in dynamic scenes and the residual mean value in static scenes to obtain a standardized calibration data set. S2: Call the standardized calibration data set, use the size error data set to calculate the average error value of each calibration position, analyze the gradient change and calculate the smoothing correction parameter, use the discrete points to construct the continuous error field of the entire workbench, and generate the global error gradient field; S3: Call the standardized calibration data set, use the dynamic residual data set, take the fitting circle center as the origin, divide the residual distribution area into multiple sector units according to the fixed angle, aggregate the residual values ​​in each sector, calculate the sector residual range, analyze the radial distance change trend of the measuring point, and generate the residual trend map; S4: calling the standardized calibration data set, calling the dynamic residual data set, and the static residual data set, calculating the weight coefficients of the dynamic residual and the static residual by analyzing the influence of the dynamic residual fluctuation data and the static residual stability on the system error, and generating a residual weight coefficient table; S5: Call the global error gradient field, residual trend map and residual weight coefficient table, use the gradient field as the size error compensation reference value, use the trend map as the directional error correction value, and dynamically adjust the application intensity of the directional correction value in combination with the residual weight coefficient. According to the coordinate data, the adjusted correction value and the size error compensation value are integrated to construct a calibration matrix to calibrate the visual measurement system and obtain the system calibration result.

[0024] The standardized calibration data set includes the size error mean set, the dynamic residual range set, and the static residual mean set. The global error gradient field specifically includes the error change direction layer, the smoothing correction parameter layer, and the path continuity fitting layer. The residual trend map specifically refers to the radial expansion trend layer, the sector range distribution layer, and the angle mapping layer. The residual weight coefficient table includes the dynamic error response weight, the static error stability weight, and the directional sensitivity coefficient. The system calibration results specifically include the coordinate mapping compensation value, the directional error adjustment coefficient, and the fusion correction output value.

[0025] Using a glass linear ruler, multiple dimensional error measurements are collected at multiple calibration positions and the average value is calculated. Using a circular target image, the residual extreme values ​​of multiple measurement points in dynamic scenes and the residual mean value in static scenes are synchronously collected. The specific steps for obtaining a standardized calibration data set are as follows: S101: using a glass line ruler, perform multiple dimension measurements at each calibration position, call multiple groups of measurement values ​​corresponding to each position, calculate the average difference between each group of measurement values, and establish a set of dimension error average values ​​in combination with the position number; Select 4 calibration positions on the system work surface, including two diagonals along the measurement range ( and direction) and parallel to Axis and At each calibration position, use a standard glass line scale as the calibration reference, with the minimum interval not exceeding 10% of the measurement stroke and the maximum interval not less than 90% of the measurement stroke ( The dimensional measurement is performed no less than 3 times according to the requirement of 66% (axis direction) or 66% (diagonal direction). Then, the data of three repeated measurements are collected for each set of measurement values, and the difference between the average value and the actual value of the glass line scale is calculated to obtain the dimensional measurement error of each position. The error average value is classified and stored according to the calibration position according to the position number to form a dimensional error average value set.

[0026] S102: calling the size error average value set, using the circular target image, collecting the numerical extreme difference of each measuring point in the dynamic scene, obtaining the numerical mean of each group of measuring points in the static scene, and generating a multi-scene measuring point residual data group; Based on the set of average size errors, multi-scenario data collection is further performed through circular target images. In dynamic scenes, the circular target is fixed horizontally on the workbench, and 25 measuring points are evenly selected within the field of view by moving the workbench. Data are collected one by one using single-point measurement. The least squares circle is fitted using all the measuring point data and the center of the circle is calculated, thereby obtaining the extreme value of the radius of each measuring point ( ) as the dynamic residual. In a static scene, a circular target with an image diameter of 10% to 30% of the field of view is selected, the workbench is kept stationary, and multiple points of the target circumference are measured from different positions in the field of view to calculate the residual mean of each set of measurement point data. Finally, the dynamic residual range is integrated with the static residual mean to generate a multi-scene measurement point residual data group containing the dynamic range and static mean.

[0027] S103: According to the residual data group of the multi-scenario measurement points, combined with the measurement position information, the dimensional error value, the dynamic residual value, and the static residual value are standardized, and the measurement data sets are respectively constructed, including the dimensional error data set, the dynamic residual data set, and the static residual data set, to obtain a standardized calibration data set; Combined with the spatial distribution information of the measurement position, the three types of data, namely the mean value of dimensional error, the dynamic residual range and the mean value of static residual, are standardized. This includes normalizing the dimensional error value to eliminate the dimension difference, converting the dynamic residual range into a standardized difference related to the measurement stroke, and scaling the static residual mean according to the field of view ratio. Subsequently, the dimensional error data set, dynamic residual data set and static residual data set corresponding to the calibration position are constructed respectively, and the data association is realized through the position number. Finally, the three types of standardized data sets are integrated to form a comprehensive calibration data set containing spatial position information, which provides standardized input for system error analysis and parameter optimization.

[0028] Call the standardized calibration data set, use the size error data set, calculate the average error value of each calibration position, analyze the gradient change and calculate the smoothing correction parameter, use discrete points to build a continuous error field of the entire workbench, and generate the global error gradient field in the following steps: S201: calling the standardized calibration data set, using the dimension error data set, extracting each set of measured coordinate values ​​and error values, calculating the average error value of each set of coordinate points, establishing a corresponding relationship between the error value and the coordinate number in combination with the position information, and generating a coordinate error mean value list; To call the standardized calibration data set, it is necessary to extract the spatial coordinate information of the measuring point and its corresponding error value from the original dimensional error data set. Each set of data is indexed by the point number P, recording the actual position of the point in the workbench coordinate system. And the corresponding measurement error value , the average calculation of multiple error measurement values ​​at the same numbered point is used to filter out the error offset caused by environmental fluctuations and image noise. Take the coordinate point P1 as an example, the position is (100,200), and its three error values ​​are 0.08mm, 0.10mm, and 0.07mm. The average error calculation is performed and the following formula is executed: ; in, For the The average error of the coordinate points is For the The error value of this point in the next measurement is is the sampling times of this point. , , , , substitute into the calculation: ; The calculation result is the average measurement error of point P1 in the current calibration cycle. This value is bound to the coordinate number of the point and recorded as (P1, 100, 200, 0.0833). The same process is performed on all coordinate points in sequence to build a table of correspondence between error values ​​and spatial coordinates, forming a structured error list with point number as index, coordinate position as spatial reference, and error mean as data field, which is used for subsequent spatial gradient modeling operations and finally generates a list of coordinate error mean values.

[0029] S202: According to the coordinate error mean list and the error change trend of adjacent coordinate points, the gradient change rate of the error data at each position is obtained, and a smoothing correction parameter is calculated; According to the coordinate error mean list, the distribution relationship of all points in space is grid mapped, the coordinate points are arranged in rows and columns according to the spatial sequence, the error value difference between each pair of adjacent coordinate points is extracted, and the error change rate per unit distance is calculated as the error gradient rate parameter in that direction. In actual processing, taking two points P1 (100, 200) and P2 (120, 200) as examples, the error values ​​are recorded as 0.08mm and 0.14mm respectively. The two points are 20mm apart in the X-axis direction, and the error change rate per unit spacing is (0.14-0.08) / 20=0.003mm / mm. The error gradient change rate is calculated using the following formula: ; in, is the error gradient change rate between points i and j, is the mean error between two points, is the spatial distance between two points. Set , , , substitute into the calculation: ; The calculated gradient value is compared with the set error fluctuation reference value, and the reference value is set to 0.002. If the gradient value is greater than the reference value, it is recorded as an abnormal gradient segment, and smoothing parameters need to be added in the interpolation modeling for correction. The correction parameter can be dynamically assigned according to the gradient ratio of the abnormal gradient area. In the above example, because the error change rate is 0.003, it exceeds the reference value and is marked as a segment that needs smoothing adjustment. After the gradient rate calculation is completed for all points, the results are stored by point coordinates, and the directionality (X or Y) is marked. It is subsequently used to generate smoothing weights in interpolation fitting, and finally the smoothing correction parameters are calculated.

[0030] S203: calling smoothing correction parameters, combining spatial interpolation, constructing a continuous error field of the entire workbench, and obtaining a global error gradient field; The specific formula for constructing the continuous error field of the entire workbench is: ; Calculate the coordinate interpolation prediction value; in, Represents coordinate point The error interpolation prediction value of Representative Smoothing correction parameters for each measuring point, Representative The residual weight factor of each measurement point is Representative Measuring points and target points The Euclidean distance of represents the smoothing control factor, Represents the total number of measurement points involved in the interpolation calculation. Represents the measurement point number, Represents the X-axis coordinate value of the target interpolation point, Represents the Y-axis coordinate value of the target interpolation point.

[0031] formula: ; Detailed explanation of the formula and the process of formula calculation and derivation: The formula is used to calculate the error interpolation prediction value of each coordinate point, and the result is used to construct a continuous error field in the workbench area and generate a global error gradient field; Parameter meaning and setting value: For the The smoothing correction parameter of each measuring point is set as ; For the The residual weight factor of the measurement points is set as ; For the Test points and target interpolation points The Euclidean distance of the measuring points is set as , the target point is , the corresponding distances are, , , ; is the smoothing control factor, set to ; is the number of measurement points involved in interpolation, set to 3; is the target interpolation coordinate, set .

[0032] Substitute the parameters into the formula for calculation: ; ; ; The result 0.2536 indicates the error interpolation prediction value of the target point, which is used to subsequently generate a continuous error field as the basic interpolation data source for the global error gradient field. The calculation results can be extended to other coordinate points in the same way to construct the error field distribution of the complete grid area.

[0033] Call the standardized calibration data set, use the dynamic residual data set, take the fitting circle center as the origin, divide the residual distribution area into multiple sector units according to the fixed angle, aggregate the residual values ​​in each sector, calculate the sector residual range, analyze the radial distance change trend of the measuring point, and generate the residual trend map in the following steps: S301: calling the standardized calibration data set, using the dynamic residual data set, extracting the coordinate position data of each measuring point relative to the center of the fitting circle, dividing the residual distribution area into a plurality of sector units in turn according to the coordinate angle of each measuring point, and generating a sector division coordinate group; Call the standardized calibration data set, use the dynamic residual data set, first fit the contour boundary of the circular target through image processing, identify the coordinate position of each measuring point, and use the fitted center coordinates as the origin to convert the coordinates of all measuring points into polar coordinates with the center of the circle as the origin, where the angle value Indicates the polar angle position of the measuring point relative to the center of the circle. The measuring point numbers are sorted by polar angle size to construct a continuous angle sequence from 0° to 360°. In order to classify the measuring points according to angle, the 360° circle is divided into several sector units with fixed angle widths. For example, if each sector is set to 15°, there are 24 sectors in total. Each sector contains all the measuring points within the polar angle range. The coordinate data of each measuring point is extracted. , and combined with its polar angle relative to the center of the circle , and divide it into the corresponding sector. With the center of the circle at (200,200) and the measuring point A at (210,210), we can calculate , so the measuring point A belongs to the 4th sector. Similarly, all measuring points are partitioned and classified, and the storage format is set to the "sector number-measuring point number-coordinate value-angle value" four-tuple structure. The results are sorted and output as sector-divided coordinate groups.

[0034] S302: Divide the coordinate groups based on sectors, aggregate all residual measurement point values ​​in each sector, record the residual extreme value corresponding to each sector, analyze and mark the residual fluctuation direction, and obtain the sector residual fluctuation parameter set; Based on the sector division coordinate group, the system aggregates all residual values ​​in each sector and extracts the residual value of each measuring point in the dynamic scene. The residual value is defined as the absolute value of the difference between the current frame radius of the measuring point and the average radius of the fitting circle. Perform a range calculation, which is the difference between the largest and smallest residuals. The calculation uses the following formula: ; in, is the residual range of the sth sector, is the residual value of the jth measuring point in the sector. Assuming the residual values ​​of the measuring points in the 5th sector are 0.04, 0.06, 0.09, and 0.03 mm, then: ; When the residual range of a sector exceeds the set fluctuation reference value, the sector is marked as having a fluctuation direction. According to the distribution direction of the measuring points and the polar angle position, the radial direction of the fluctuation direction is defined. The residual fluctuation direction is determined based on the sign of the product of the residual difference between the two measuring points and their polar angle direction. If the product is greater than zero, it is defined as an expansion trend, and if it is less than zero, it is a contraction trend. The system processes each sector in sequence according to the angle sequence number, records the residual range and fluctuation direction, and integrates all the results into a three-field data table with "sector number-range value-fluctuation trend", and finally obtains the sector residual fluctuation parameter set.

[0035] S303: according to the sector residual fluctuation parameter set, using the residual extreme value and fluctuation direction of each sector, combined with the radial distance data from each measuring point to the center of the circle, extracting the radial distance variation of the measuring points in each sector, analyzing the radial distance variation trend of the measuring points, and establishing a residual trend map; According to the sector residual fluctuation parameter set, the system further calls the actual coordinate data of each measuring point and the center coordinate to calculate its radial distance, which is recorded as , all the measuring points in each sector constitute a radial distance set. By calculating the difference between the maximum radial distance and the minimum radial distance, it is determined whether there is a non-uniform distribution of the radial structure in the sector. Combined with the residual extreme value and directional annotation obtained in the previous stage, the radial distance in each sector is sorted according to the polar angle sequence of the measuring points, and the trend of the radial change curve is observed. If it shows a monotonically increasing trend and the residual direction is marked as expansion, it is confirmed that the sector has an error expansion behavior, otherwise it is an inward contraction behavior. If the change is irregular, it is marked as no significant direction. The analysis logic is organized into a directional trend data structure, and the trend value is calculated using the following formula: ; in, is the radial variation trend value of the sth sector, is the radial distance of the jth measuring point in ascending order of polar angle, is the number of measurement points in the sth sector. Assume that the radial distances of the five points in the 6th sector are 8.0, 8.3, 8.5, 8.6, and 8.7 mm, then: ; The value is positive and continues to grow, which is marked as expansion. The trend results of all sectors are summarized into a residual trend curve mapping diagram, which reflects the spatial structural evolution state of the error in polar coordinates, and finally forms a residual trend map.

[0036] Call the standardized calibration data set, call the dynamic residual data set, and the static residual data set, analyze the influence of the dynamic residual fluctuation data and the static residual stability on the system error, calculate the weight coefficients of the dynamic residual and the static residual, and generate the residual weight coefficient table in the following steps: S401: calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, analyzing the dynamic residual fluctuation intensity by calculating the extreme values ​​of multiple measuring points in the dynamic class, identifying the static residual stability by calculating the mean square error values ​​of multiple measuring points in the static class, and generating a residual characteristic data set; Call the standardized calibration data set, call the dynamic residual data set and the static residual data set. First, for the dynamic residual part, extract the maximum and minimum values ​​of each measuring point in the continuous acquisition frame, and record its range. The range calculation uses the formula: ; in, is the dynamic residual extreme value of a certain measuring point, is the idth measured radius value of the measuring point in the dynamic acquisition process. Assuming that the continuous measured radius of measuring point A is 8.22, 8.19, 8.31, and 8.25 mm, then: ; The system performs the same operation on all dynamic measuring points to obtain the dynamic residual range set of all measuring points. At the same time, for the measuring point data in the static scene, the system collects the residual value sequence of each measuring point in different static frames and calculates the mean square error using the formula: ; in, is the static residual mean square error, is the jtth measurement value, is the average value of the residual value of all frames at the measuring point. Assuming that the measured values ​​of measuring point B in a static scene are 0.02, 0.01, 0.03, and 0.01 mm, the average value is: ; ; The system summarizes the dynamic extreme values ​​and static mean square errors of all measuring points into dynamic residual fluctuation intensity data columns and static residual stability data columns, and integrates them into residual characteristic data groups according to the measuring point numbers.

[0037] S402: Based on the residual characteristic data group, the fluctuation intensity index and the stability index are normalized, and the residual influence analysis result is established by analyzing the influence of the fluctuation intensity index and the stability index on the visual measurement system error; Based on the residual characteristic data set, the system performs normalization processing on the dynamic residual extreme value and the static residual mean square error value respectively, and sets the normalization benchmark as the maximum value of the residual values ​​of all measuring points in the current scene. The normalization processing formula is: ; in, is the residual index value of the icth measurement point after normalization, is the original residual value, is the maximum residual reference value. If the maximum dynamic residual value is 0.14 mm and the measuring point A is 0.12 mm, then: ; Similarly, the static residual is normalized in the same way. After normalization, the system calls the normalized dynamic and static residual indicators to construct an error impact matrix. The dynamic normalized value and static normalized value of each measuring point are used as two dimensions, and the corresponding coordinate points are the error impact distribution points. The offset angle and impact amplitude of each measuring point in the error map are calculated based on the two-dimensional distribution. The dynamic weight value is subtracted from the static weight value to preliminarily determine the main source direction of the system error. The system extracts the high-frequency band as the main direction of residual impact based on the distribution density of the offset angle concentration area, and establishes the residual impact analysis results.

[0038] S403: According to the residual impact analysis result, a dynamic residual coefficient and a static residual coefficient are set according to the impact degree, and a residual weight coefficient table is obtained; According to the residual impact analysis results, the system sets the residual impact ratio coefficient, establishes a correlation between the normalized values ​​of dynamic and static residuals and the degree of impact, and the dynamic residual weight coefficient and the static residual weight coefficient Assume that the following constraints are met: ; According to the ratio of normalized values, the weight coefficient is calculated by the proportional method. If the normalized dynamic residual of a measuring point is 0.8 and the static residual is 0.2, then: ; ; The system stores the dynamic and static residual weight coefficients corresponding to each measuring point in a tabular form, forming a three-field structure of "measuring point number-dynamic weight-static weight". After cyclic processing of all measuring points, the integrated output is a residual weight coefficient table.

[0039] The global error gradient field, residual trend map and residual weight coefficient table are called, the gradient field is used as the reference value for size error compensation, the trend map is used as the directional error correction value, and the application intensity of the directional correction value is dynamically adjusted in combination with the residual weight coefficient. According to the coordinate data, the adjusted correction value and the size error compensation value are integrated to construct a calibration matrix to calibrate the visual measurement system. The specific steps for obtaining the system calibration results are as follows: S501: calling the global error gradient field, the residual trend map and the residual weight coefficient table, calculating the basic compensation values ​​and the directional correction values ​​of multiple coordinate points according to the size error compensation value and the directional error correction value, and generating a coordinate mapping error value group; The coordinate error calculation submodule extracts the numbers, horizontal and vertical axis indexes and corresponding spatial position coordinates of all measured coordinate points one by one according to the established size error compensation values ​​in the global error gradient field and the directional error correction values ​​marked in the residual trend map, and performs a double-value reading operation on each coordinate point, that is, respectively obtains the basic compensation value E1 of the point in the error gradient field and the directional correction value E2 in the trend map, and performs an average merging operation on these two error values ​​for subsequent error fusion processing. In the process, the system coordinate index matching method is required to ensure the accuracy of the value correspondence. In addition, the corresponding compensation source tag needs to be written into the mapping table to facilitate subsequent weight call and adjustment processing. This process binds each coordinate number to its corresponding merged error value, and finally generates a complete coordinate mapping error value group for weighted correction in the next stage; the formula is used: ; Calculate the coordinate merging error value, where, The basic combined compensation value for the coordinate point, is the size error compensation value, is the directional correction value. is 10, is 6, substitute it into the calculation: ; The calculation results show that the combined error value is 8 in space. Then the error values ​​of all points are uniformly transferred into the coordinate mapping table, and stored in sequence with the combination of the number and the actual position number to complete the data operation of the error distribution structure. This data set will be used as the data input port for subsequent weight superposition correction to establish a coordinate mapping error value group.

[0040] S502: According to the coordinate mapping error value group, according to the dynamic and static residual weight values ​​corresponding to each coordinate point, the directional correction value is adjusted to obtain a correction adjustment application value set; The direction correction weighted submodule receives the directional correction values ​​of all position points in the coordinate mapping error value group, and performs weighted adjustment processing based on the dynamic residual coefficient α and the static residual coefficient β. This processing link first extracts the α and β values ​​corresponding to each coordinate number from the residual weight coefficient table, and weights the direction correction values, and introduces a normalization coefficient λ to balance the influence of extreme disturbance measurement points on the overall correction result. In specific operations, weighted multiplication processing is performed on each group of measurement points and written into the structure data set. Each item in the data set contains the coordinate number, the original value of the direction correction, the weighted adjustment value and the corrected error value; the formula is used: ; Calculate the direction correction weighted value, where is the weighted adjustment application value of the directional error correction, is the dynamic residual weight coefficient, is the static residual weight coefficient, Correct the original value for directionality, is the normalization coefficient. Set is 0.3, is 0.4, is 0.85, is 6, substitute it into the calculation: ; After the calculation is completed, each weighted adjustment value is remapped to the corresponding coordinate number, and the original direction correction item in the mapping value table is overwritten by the coordinate index method, updated to the weight correction result, and the correction adjustment application value set is obtained.

[0041] S503: Based on the correction adjustment application value set and according to the coordinate data, the adjusted correction value and the size error compensation value are integrated, the error correction parameters of multiple positions are calculated, the calibration matrix is ​​constructed, and the visual measurement system is calibrated to obtain the system calibration result; The specific formula for calculating the error correction parameters for multiple positions is: ; Calculate the standardized error correction parameters after fusion; in, Representative The fusion standardized error correction parameter value of the coordinate position, represents the fusion weight coefficient, Representative Directional correction adjustment value for the coordinate position, Representative The size error compensation value of the coordinate position, is the total number of coordinates of the calibration points to be fused, The coordinate point number currently being calculated.

[0042] formula: ; Detailed explanation of the formula and the process of formula calculation and derivation: The formula is used to calculate the fusion standardized error correction parameters of each coordinate point, and the parameters are used to construct the calibration matrix of the visual measurement system.

[0043] Parameter meaning and setting value: is the fusion weight coefficient, set to 0.6; For the The directional correction adjustment value of the coordinate point is set to , , , a total of 3 location points are extracted; For the The size error compensation value of the coordinate point is set as , , ; is the total number of coordinate points, set to 3; Substitute the parameters into the formula for calculation: ; ; The fusion error correction parameter of the first position point is calculated as: ; The results show that the standardized error correction parameters after fusion It is 1.855, which corresponds to the final error correction input value of the first measurement point. It will be used as the basis for error adjustment at this position point in the process of constructing the calibration matrix. The parameters of other position points are processed in the same way and written into the calibration matrix to complete the error correction process of the visual measurement system.

[0044] See also Figure 2 , an industrial machine vision measurement system calibration device, the industrial machine vision measurement system calibration device is used to perform the above industrial machine vision measurement system calibration method, the device comprises: The data acquisition module is based on a glass line scale. It collects multiple dimensional error measurements at multiple calibration positions and calculates the average value. It uses a circular target image to synchronously obtain residual measurement data of multiple measurement points in dynamic and static scenes, extracts the residual range of dynamic measurement points and the residual variance of static measurement points, and generates a standardized calibration data set. The error field construction module is based on the standardized calibration data set and uses the size error data set to extract the coordinates of multiple calibration points and the corresponding error averages. It performs gradient analysis and calculates smoothing correction parameters based on the error variation and spatial spacing between adjacent points. It fills the error data in the region with spatial interpolation to establish a global error gradient field. The residual analysis module is based on the standardized calibration data set, uses the dynamic residual data set, calls the fitting circle center position as the angle division origin, divides the residual distribution area into multiple equal-angle sector units, aggregates the residual measurement points in multiple sectors and calculates the extreme difference, extracts the change trend in multiple sectors in combination with the radial distance information, and establishes the residual trend map; The weight calculation module is based on the standardized calibration data set, uses the dynamic residual data set and the static residual average data set to normalize the two sets of residual data, analyzes the influence of the dynamic residual fluctuation data on the system error, analyzes the influence of the static residual stability on the system error, obtains the weight coefficients of the dynamic residual and the static residual, and obtains the residual weight coefficient table; The compensation fusion module calls the global error gradient field, residual trend map and residual weight coefficient table to extract the error compensation value, direction correction value and application intensity coefficient of each coordinate point, fuses the direction correction value and size compensation value, constructs the calibration matrix parameters, calibrates the visual measurement system and obtains the system calibration results.

[0045] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0046] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0047] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0048] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0049] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0050] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0051] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0052] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0053] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0054] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0055] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for calibrating an industrial machine vision measurement system, characterized in that: The method comprises: S1: Using a glass line ruler, collect multiple dimensional error measurement values ​​at multiple calibration positions and calculate the average value. Using a circular target image, synchronously collect the residual extreme values ​​of multiple measurement points in dynamic scenes and the residual mean value in static scenes to obtain a standardized calibration data set. S2: calling the standardized calibration data set, using the size error data set, calculating the average error value of each calibration position, analyzing the gradient change and calculating the smoothing correction parameter, using discrete points to construct a continuous error field of the entire workbench, and generating a global error gradient field; S3: calling the standardized calibration data set, using the dynamic residual data set, taking the fitting circle center as the origin, dividing the residual distribution area into multiple sector units according to a fixed angle, aggregating the residual values ​​in each sector, calculating the sector residual range, analyzing the radial distance change trend of the measuring point, and generating a residual trend map; S4: calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, calculating the weight coefficients of the dynamic residual and the static residual by analyzing the influence of the dynamic residual fluctuation data and the static residual stability on the system error, and generating a residual weight coefficient table.

2. The industrial machine vision measurement system calibration method according to claim 1, characterized in that: The standardized calibration data set includes a size error mean set, a dynamic residual range set, and a static residual mean set. The global error gradient field is specifically an error change direction layer, a smoothing correction parameter layer, and a path continuity fitting layer. The residual trend map specifically refers to a radial expansion trend layer, a sector range distribution layer, and an angle mapping layer. The residual weight coefficient table includes a dynamic error response weight, a static error stability weight, and a directional sensitivity coefficient.

3. The industrial machine vision measurement system calibration method according to claim 1, characterized in that: Using a glass linear ruler, multiple dimensional error measurements are collected at multiple calibration positions and the average value is calculated. Using a circular target image, the residual extreme values ​​of multiple measurement points in dynamic scenes and the residual mean value in static scenes are synchronously collected. The specific steps for obtaining a standardized calibration data set are as follows: S101: using a glass line ruler, perform multiple dimension measurements at each calibration position, call multiple groups of measurement values ​​corresponding to each position, calculate the average difference between each group of measurement values, and establish a set of dimension error average values ​​in combination with the position number; S102: calling the size error average value set, using the circular target image, collecting the numerical extreme difference of each measuring point in the dynamic scene, obtaining the numerical mean of each group of measuring points in the static scene, and generating a residual data group of multi-scene measuring points; S103: According to the multi-scenario measurement point residual data group, combined with the measurement position information, the dimensional error value, the dynamic residual value, and the static residual value are standardized, and measurement data sets are respectively constructed, including a dimensional error data set, a dynamic residual data set, and a static residual data set, to obtain a standardized calibration data set.

4. The industrial machine vision measurement system calibration method according to claim 3, characterized in that: The standardized calibration data set is called, the size error data set is used to calculate the average error value of each calibration position, the gradient change is analyzed and the smoothing correction parameter is calculated, and the continuous error field of the entire workbench is constructed using discrete points. The steps of generating the global error gradient field are specifically as follows: S201: calling the standardized calibration data set, using the dimension error data set, extracting each set of measured coordinate values ​​and error values, calculating the average error value of each set of coordinate points, establishing a corresponding relationship between the error value and the coordinate number in combination with the position information, and generating a coordinate error mean value list; S202: According to the coordinate error mean value list and the error change trend of adjacent coordinate points, the gradient change rate of the error data at each position is obtained, and a smoothing correction parameter is calculated; S203: calling the smoothing correction parameter, combining with spatial interpolation, constructing a continuous error field of the entire workbench, and obtaining a global error gradient field.

5. The industrial machine vision measurement system calibration method according to claim 4, characterized in that: The specific formula for constructing the continuous error field of the entire workbench domain is: ； Calculate the coordinate interpolation prediction value; in, Represents coordinate point The error interpolation prediction value of Representative Smoothing correction parameters for each measuring point, Representative The residual weight factor of each measurement point is Representative Measuring points and target points The Euclidean distance of represents the smoothing control factor, Represents the total number of measurement points involved in the interpolation calculation. Represents the measurement point number, Represents the X-axis coordinate value of the target interpolation point, Represents the Y-axis coordinate value of the target interpolation point.

6. The industrial machine vision measurement system calibration method according to claim 4, characterized in that: The standardized calibration data set is called, and the dynamic residual data set is used. The residual distribution area is divided into multiple sector units according to a fixed angle with the center of the fitting circle as the origin. The residual values ​​in each sector are aggregated, the sector residual range is calculated, and the radial distance change trend of the measuring point is analyzed. The steps of generating the residual trend map are specifically as follows: S301: calling the standardized calibration data set, using the dynamic residual data set, extracting the coordinate position data of each measuring point relative to the center of the fitting circle, dividing the residual distribution area into a plurality of sector units in turn according to the coordinate angle of each measuring point, and generating a sector division coordinate group; S302: Based on the sector division coordinate group, aggregate all residual measurement point values ​​in each sector, record the residual extreme value corresponding to each sector, analyze and mark the residual fluctuation direction, and obtain the sector residual fluctuation parameter set; S303: According to the sector residual fluctuation parameter set, using the residual extreme value and fluctuation direction of each sector, combined with the radial distance data from each measuring point to the center of the circle, extract the radial distance change of the measuring point in each sector, analyze the radial distance change trend of the measuring point, and establish a residual trend map.

7. The industrial machine vision measurement system calibration method according to claim 6, characterized in that: The steps of calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, calculating the weight coefficients of the dynamic residual and the static residual by analyzing the influence of the dynamic residual fluctuation data and the static residual stability on the system error, and generating the residual weight coefficient table are specifically as follows: S401: calling the standardized calibration data set, calling the dynamic residual data set and the static residual data set, analyzing the dynamic residual fluctuation intensity by calculating the extreme values ​​of multiple measuring points in the dynamic class, identifying the static residual stability by calculating the mean square error values ​​of multiple measuring points in the static class, and generating a residual characteristic data set; S402: Based on the residual characteristic data group, normalize the fluctuation intensity index and the stability index, and establish a residual influence analysis result by analyzing the influence of the fluctuation intensity index and the stability index on the visual measurement system error; S403: According to the residual impact analysis result, a dynamic residual coefficient and a static residual coefficient are set according to the impact degree, and a residual weight coefficient table is obtained.

8. The industrial machine vision measurement system calibration method according to claim 1, characterized in that: The method further comprises: S5: calling the global error gradient field, residual trend map and residual weight coefficient table, using the gradient field as the size error compensation reference value, using the trend map as the directional error correction value, combining the residual weight coefficient, dynamically adjusting the application intensity of the directional correction value, fusing the adjusted correction value and the size error compensation value according to the coordinate data, constructing a calibration matrix to calibrate the visual measurement system, and obtaining the system calibration result; The system calibration result specifically includes a coordinate mapping compensation value, a directional error adjustment coefficient, and a fusion correction output value.

9. The industrial machine vision measurement system calibration method according to claim 8, characterized in that: The global error gradient field, residual trend map and residual weight coefficient table are called, the gradient field is used as the size error compensation reference value, the trend map is used as the directional error correction value, and the application intensity of the directional correction value is dynamically adjusted in combination with the residual weight coefficient. According to the coordinate data, the adjusted correction value and the size error compensation value are merged to construct a calibration matrix to calibrate the visual measurement system. The steps of obtaining the system calibration result are specifically as follows: S501: calling the global error gradient field, residual trend map and residual weight coefficient table, calculating the basic compensation values ​​and directional correction values ​​of multiple coordinate points according to the size error compensation value and the directional error correction value, and generating a coordinate mapping error value group; S502: According to the coordinate mapping error value group, according to the dynamic and static residual weight values ​​corresponding to each coordinate point, the directional correction value is adjusted to obtain a correction adjustment application value set; S503: Based on the correction adjustment application value set, according to the coordinate data, the adjusted correction value and the size error compensation value are integrated, the error correction parameters of multiple positions are calculated, the calibration matrix is ​​constructed, and the visual measurement system is calibrated to obtain the system calibration result; The specific formula for calculating the error correction parameters for multiple positions is: ； Calculate the standardized error correction parameters after fusion; in, Representative The fusion standardized error correction parameter value of the coordinate position, represents the fusion weight coefficient, Representative Directional correction adjustment value for the coordinate position, Representative The size error compensation value of the coordinate position, is the total number of coordinates of the calibration points to be fused, The coordinate point number currently being calculated.

10. An industrial machine vision measurement system calibration device, characterized in that: According to the industrial machine vision measurement system calibration method according to any one of claims 1 to 9, the device comprises: The data acquisition module is based on a glass line scale. It collects multiple dimensional error measurements at multiple calibration positions and calculates the average value. It uses a circular target image to synchronously obtain residual measurement data of multiple measurement points in dynamic and static scenes, extracts the residual range of dynamic measurement points and the residual variance of static measurement points, and generates a standardized calibration data set. The error field building module is based on the standardized calibration data set and uses the size error data set to extract the coordinates of multiple calibration points and the corresponding error averages, and performs gradient analysis according to the error variation amplitude and spatial spacing between adjacent points, calculates smoothing correction parameters, and fills the error data in the region in combination with spatial interpolation to establish a global error gradient field; The residual analysis module is based on the standardized calibration data set, uses the dynamic residual data set, calls the fitting circle center position as the angle division origin, divides the residual distribution area into multiple equal-angle sector units, aggregates the residual measurement points in multiple sectors and calculates the extreme difference, extracts the change trend in multiple sectors in combination with the radial distance information, and establishes a residual trend map; The weight calculation module is based on the standardized calibration data set, uses the dynamic residual data set and the static residual mean data set to normalize the two sets of residual data, analyzes the influence of the dynamic residual fluctuation data on the system error, analyzes the influence of the static residual stability on the system error, obtains the weight coefficients of the dynamic residual and the static residual, and obtains the residual weight coefficient table; The compensation fusion module calls the global error gradient field, residual trend map and residual weight coefficient table, extracts the error compensation value, direction correction value and application intensity coefficient of each coordinate point, fuses the direction correction value and size compensation value, constructs the calibration matrix parameters, calibrates the visual measurement system, and obtains the system calibration result.

Images