Small-module gear pitch deviation correction method and device based on visual measurement
Through visual measurement technology dynamic gamma correction and multi-scale local enhancement, a three-dimensional point cloud model of gear is constructed, the tooth pitch deviation is calculated and intelligent correction strategies are formulated, which solves the problem of tooth pitch deviation in small-module gear manufacturing and improves the accuracy of gears and performance, extending service life.
Patent Information
- Application Number
- CN202411240810.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-05
AI Technical Summary
The prior art is difficult to effectively solve the problem of tooth pitch deviation in small module gears during manufacturing, resulting in low transmission accuracy, high noise, short service life, and low efficiency and low accuracy of traditional measurement methods.
Using a small-module gear tooth pitch deviation correction method based on visual measurement, the gear multi-directional image is obtained through dynamic gamma correction and multi-scale local enhancement, a body point cloud model is constructed, the tooth pitch deviation is calculated, and the global deviation superposition analysis is carried out, and an intelligent Correction strategy.
Improve gear accuracy and performance, reduce noise, extend service life, and improve production efficiency and product quality.
Smart Images

Figure CN118762150B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of pitch deviation correction, and in particular to a method and a device for correcting the pitch deviation of a small-module gear based on visual measurement. Background Art
[0002] With the development of intelligent manufacturing industry, small modulus precision gears are widely used in robot joint drive, micro reducer and other fields. Plastic gears have nonlinear shrinkage when they are manufactured by injection molding process, and the manufactured plastic gear products are also affected by many factors such as the structure, parameters, mold design scheme, mold processing accuracy, mold assembly accuracy, injection molding process parameters, injection molding machine model, etc. Therefore, the current manufacturing accuracy of plastic gears at home and abroad is distributed in 8-11 levels according to GB / T 38192-2019, and most of them are in 9~11 levels, which makes it impossible to make the overall noise of plastic gearbox smaller, which seriously restricts the development of plastic gears.
[0003] This type of small modulus gear is small in size and difficult to process, which easily leads to serious pitch deviation. The existence of pitch deviation not only affects the transmission accuracy, but also causes problems such as increased gear wear and noise, which seriously affects the service life and working performance of the equipment. The traditional pitch deviation measurement method mainly relies on contact measuring tools, such as micrometers and vernier calipers. This method not only has low measurement efficiency, but also its accuracy is affected by human operation, and the deviation accuracy is low. Therefore, an intelligent gear pitch deviation correction method is needed. Summary of the invention
[0004] In order to solve the above technical problems, the present invention proposes a method and device for correcting the pitch deviation of small-module gears based on visual measurement, so as to solve at least one of the above technical problems.
[0005] To achieve the above object, the present invention provides a method for correcting the pitch deviation of a small-module gear based on visual measurement, comprising the following steps:
[0006] Preferably, step S1 comprises the following steps:
[0007] Step S1: obtaining a dynamic multi-directional image of the gear and processing mold parameters; performing dynamic gamma correction on the dynamic multi-directional image of the gear, and performing multi-scale local enhancement, thereby constructing a detail-enhanced multi-directional image of the gear;
[0008] Step S2: Calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform stereo point cloud modeling based on the center position coordinates of each tooth to construct a gear stereo point cloud model;
[0009] Step S3: performing mold intermediate reference calculation on the processing mold parameters to generate intermediate reference surface position coordinates; performing theoretical distance parameter modeling on the intermediate reference surface position coordinates to obtain a theoretical intermediate reference model;
[0010] Step S4: calculating the gear pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values;
[0011] Step S5: performing global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value;
[0012] Step S6: Based on the gear pitch deviation trend law and the gear pitch global deviation value, a pitch deviation correction decision is made, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
[0013] The present invention performs dynamic gamma correction and multi-scale local enhancement on the dynamic multi-directional image of the gear, effectively improving the clarity and detail expression of the image, thereby making the shape and structure of the gear more clearly visible, accurately obtaining various detailed features of the gear, including tooth shape, tooth pitch, etc., and providing a more accurate data basis for subsequent analysis and modeling. Through spatial center position calculation and three-dimensional point cloud modeling, the center position coordinates of each tooth are generated and a three-dimensional point cloud model of the gear is constructed. The three-dimensional point cloud model provides richer and three-dimensional gear shape information, making the understanding of the gear structure more comprehensive and in-depth, accurately evaluating the pitch deviation and shape deviation, and providing a reliable basis for optimization in the manufacturing process. Through mold intermediate reference calculation and theoretical distance parameterized modeling, a theoretical intermediate reference model is generated. This model provides a theoretical reference in gear manufacturing and identifies actual Deviations and errors in actual manufacturing are crucial to understanding the accuracy and precision of gear manufacturing, helping adjustments and optimizations during the manufacturing process. Through pitch deviation calculation, the deviation of each tooth can be accurately evaluated, so as to understand the shape differences in actual gear manufacturing, analyze and discover potential problems and sources of errors in the manufacturing process, and provide a reference for improving manufacturing quality. Through global deviation superposition calculation and trend analysis, the laws and trends of gear pitch deviations are revealed, and common deviation patterns and problems in manufacturing are discovered, providing guidance for formulating correction strategies and optimization measures. Through intelligent pitch deviation correction strategies, correction decisions are made according to the trend laws of pitch deviations and global deviation values. Intelligent correction strategies adjust pitch deviations to improve the accuracy and performance of gears. Through timely correction measures, product quality is guaranteed to meet requirements, and production efficiency and product reliability are improved.
[0014] Preferably, step S1 comprises the following steps:
[0015] Step S11: real-time monitoring of the small modulus gear processing process, obtaining dynamic multi-directional images of the gear and processing mold parameters;
[0016] Step S12: Calculate the pixel brightness of the dynamic multi-directional image of the gear to obtain the pixel brightness value;
[0017] Step S13: performing image brightness distribution analysis on the pixel brightness values to generate image brightness distribution data;
[0018] Step S14: performing dynamic gamma correction on the dynamic multi-directional image of the gear according to the image brightness distribution data, thereby obtaining a brightness corrected image;
[0019] Step S15: Perform multi-scale local enhancement on the brightness-corrected image to construct a detail-enhanced gear multi-directional image.
[0020] The present invention timely acquires dynamic multi-directional images of gears and processing mold parameters by real-time monitoring of the processing process of small-modulus gears, instantly understands the situation in the processing process, helps detect problems occurring in the processing, and provides real-time data support for subsequent data analysis and pitch deviation correction. By performing pixel brightness calculation on the dynamic multi-directional images of gears, the pixel brightness values are accurately acquired, the brightness information of the image is quantified, and an accurate data basis is provided for subsequent brightness distribution analysis. Image brightness distribution analysis is performed on the pixel brightness values to generate detailed image brightness distribution data. This analysis understands the distribution of brightness characteristics of the image and provides guidance for subsequent image processing. By performing dynamic gamma correction on the dynamic multi-directional images of gears according to the image brightness distribution data, the brightness contrast of the image is effectively adjusted, the image clarity and detail expression are improved, the detail characteristics of the gear are more accurately observed, and a more accurate image basis is provided for pitch deviation correction. By performing multi-scale local enhancement on the brightness correction image, the detail expression of the image is further enhanced, the various detail characteristics of the gear are highlighted, the image clarity and visual effect are improved, and better image support is provided for accurate evaluation and correction of pitch deviation.
[0021] Preferably, step S15 specifically comprises the following steps:
[0022] Performing discrete wavelet transform on the brightness correction image to obtain component images of different frequency bands; the component images of different frequency bands include a high-frequency band detail layer image and a low-frequency band rough layer image;
[0023] Perform multi-scale detail recognition on the high-frequency band detail layer image to obtain multiple detail features of the image;
[0024] Perform large-scale corrosion denoising on the high-frequency band detail layer image to obtain the denoised component image;
[0025] Based on multiple detail features of the image, the denoised component image is locally expanded to generate a local detail enhanced component image;
[0026] Wavelet image reconstruction is performed on the local detail enhanced component image and the low-frequency band rough layer image to construct a detail enhanced multi-directional image of the gear.
[0027] The present invention decomposes the brightness correction image into component images of different frequency bands, including a high-frequency band detail layer image and a low-frequency band coarse layer image, through discrete wavelet transform. This decomposition can better understand the frequency characteristics of the image and provide a basis for subsequent image processing. By performing multi-scale detail recognition on the high-frequency band detail layer image, multiple detail characteristics of the image are obtained, various detail information in the image is captured, and a more accurate basis is provided for subsequent detail enhancement. Large-scale corrosion denoising is performed to effectively reduce noise interference in the image, obtain a cleaner denoised component image, improve the quality and clarity of the image, and better retain the detail information of the image in subsequent processing. By performing local detail expansion on the denoised component image based on multiple detail characteristics of the image, the local detail characteristics of the image are enhanced, the detail performance of the image is improved, the important details in the image are highlighted, and the visualization effect of the gear image is enhanced. By performing wavelet image reconstruction on the local detail enhancement component image and the low-frequency band coarse layer image, the information of different frequency bands is recombined to construct a multi-directional image of the gear with enhanced details, and the various frequency components of the image are comprehensively utilized to obtain a clearer and more detailed image result.
[0028] Preferably, step S2 specifically comprises the following steps:
[0029] Step S21: performing dynamic continuous frame analysis on the detail-enhanced gear multi-directional image to generate a gear multi-directional image frame sequence;
[0030] Step S22: visually identifying each tooth on the gear multi-directional image frame sequence and marking tooth feature points;
[0031] Step S23: Calculate the spatial center position of the tooth feature points to generate the center position coordinates of each tooth;
[0032] Step S24: performing a three-dimensional space registration analysis on the gear multi-directional image frame sequence according to the center position coordinates of each tooth, thereby generating three-dimensional space registration data of tooth feature points;
[0033] Step S25: performing gear number layout structure analysis on the gear multi-directional image frame sequence to obtain gear number layout structure data;
[0034] Step S26: Performing three-dimensional point cloud modeling on the gear tooth number layout structure data based on the three-dimensional spatial registration data of the tooth feature points to construct a three-dimensional point cloud model of the gear.
[0035] The present invention generates a frame sequence of the multi-directional gear image by performing dynamic continuous frame analysis on the detail-enhanced multi-directional gear image, captures the dynamic changes of the gear at different angles and positions, provides more comprehensive image information for subsequent processing, performs visual recognition of each tooth on the multi-directional gear image frame sequence, marks the tooth feature points, identifies and tracks the position and features of each tooth, which provides basic data for subsequent pitch deviation correction, and generates the center position coordinates of each tooth by calculating the spatial center position of the tooth feature points. This calculation accurately locates the position of each tooth, providing accurate data support for subsequent analysis and correction. The center position coordinates of the gear are used for stereoscopic spatial registration analysis to generate stereoscopic spatial registration data of the tooth feature points, understand the positional relationship of the gears in three-dimensional space, and provide an important basis for subsequent precise correction. The gear multi-directional image frame sequence is analyzed for the tooth number layout structure to obtain the gear tooth number layout structure data, understand the overall structure and layout of the gear, and provide a global perspective for the correction of pitch deviation. Based on the stereoscopic spatial registration data of the tooth feature points, stereoscopic point cloud modeling of the gear tooth number layout structure data is performed to construct a stereoscopic point cloud model of the gear, which provides a more realistic and accurate gear model and provides a more reliable basis for the correction of pitch deviation.
[0036] Preferably, the specific steps of step S24 are:
[0037] Perform gear contour line recognition on the gear multi-directional image frame sequence and extract the gear contour lines of multiple frames;
[0038] Perform multi-vision posture difference analysis on multi-frame gear contour lines to generate multi-frame contour line posture difference data;
[0039] According to the center position coordinates of each tooth, a multi-posture coordinate transformation calculation is performed on the multi-frame contour line posture difference data to obtain multiple visual contour coordinates;
[0040] Performing stereoscopic spatial registration of tooth feature points based on multiple visual contour coordinates to obtain stereoscopic feature parameters of the tooth feature points;
[0041] The stereoscopic registration error is corrected for the stereoscopic feature parameters of the tooth feature points, thereby generating stereoscopic space registration data of the tooth feature points.
[0042] The present invention identifies the gear contour line of the gear multi-orientation image frame sequence, accurately extracts the gear shape profile information, captures the gear geometric shape features, and provides basic data for subsequent posture analysis and error correction. By performing multi-visual posture difference analysis on the multi-frame gear contour lines, the posture difference data of the multi-frame contour lines is generated, and the posture changes of the gear at different angles and positions are understood, providing a basis for subsequent coordinate transformation calculations. According to the center position coordinates of each tooth, multi-posture coordinate transformation calculations are performed on the multi-frame contour line posture difference data to obtain multiple visual contour coordinates, and the posture information of the gear contour line is converted to a unified coordinate system for analysis. Based on the multiple visual contour coordinates, the tooth feature points are three-dimensionally registered to ensure that the positional relationship of each visual contour of the gear in the three-dimensional space is consistent, which lays the foundation for subsequent feature parameter extraction and error correction. By performing stereo registration error correction on the three-dimensional feature parameters of the tooth feature points, the spatial registration accuracy of the tooth feature points is improved, the accuracy of each feature point of the gear in the three-dimensional space is ensured, and the pitch deviation is detected and corrected more accurately.
[0043] Preferably, step S3 specifically comprises the following steps:
[0044] Step S31: performing mold intermediate reference calculation on the processing mold parameters, thereby obtaining the mold intermediate reference value;
[0045] Step S32: Projecting the intermediate reference value of the mold onto a reference plane to generate the position coordinates of the intermediate reference plane;
[0046] Step S33: Calculating the theoretical distance of the gear position on the intermediate reference surface position coordinates based on the preset gear processing parameters, thereby obtaining the distance value from each gear to the reference surface;
[0047] Step S34: perform parameterized modeling calculation on the distance values from each gear to the reference surface to obtain a theoretical intermediate reference model.
[0048] The present invention determines the intermediate datum of the mold, that is, the reference point in the mold processing process, by performing mold intermediate datum calculation on the processing mold parameters, establishes a unified datum standard, provides a basis for subsequent precise processing and measurement, projects the mold intermediate datum value on the datum plane, generates intermediate datum plane position coordinates, determines the center position of the mold processing, provides an accurate reference point and positioning basis for subsequent gear processing, and calculates the theoretical tooth position distance of the intermediate datum plane position coordinates according to preset gear processing parameters to obtain the distance value from each gear to the datum plane, theoretically determines the accurate distance between each gear and the datum plane, and obtains a theoretical intermediate datum model by performing parameterized modeling calculation on the distance value from each gear to the datum plane. This model helps to understand the theoretical relationship between the gear and the datum plane, and provides a mathematical model and calculation basis for the correction of the pitch deviation.
[0049] Preferably, the specific steps of step S4 are:
[0050] Step S41: performing overall position deviation registration of the gear stereo point cloud model according to the theoretical intermediate reference model to obtain a reference registered gear model;
[0051] Step S42: extracting the position of each tooth vertex of the reference registration gear model one by one, and generating the point cloud coordinates of each tooth vertex;
[0052] Step S43: Calculate the pitch deviation of the tooth vertex point cloud coordinates based on the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values.
[0053] The present invention performs overall position deviation registration of the gear three-dimensional point cloud model according to the theoretical intermediate reference model, aligns the actually measured gear point cloud model with the theoretical model, accurately measures the overall position deviation of the gear, extracts the tooth vertex position of each tooth vertex of the reference registered gear model one by one, generates point cloud coordinates of each tooth vertex, captures the accurate position information of each tooth of the gear, provides detailed data for the calculation of the pitch deviation, and calculates the pitch deviation of the tooth vertex point cloud coordinates based on the theoretical intermediate reference model, thereby generating multiple gear pitch deviation values, and evaluating the difference between the actual position and theoretical position of each tooth.
[0054] Preferably, the specific steps of step S5 are:
[0055] Step S51: performing a gear number distribution analysis on the gear 3D point cloud model to obtain gear tooth number distribution data;
[0056] Step S52: marking the gear tooth number distribution data with corresponding teeth based on multiple gear pitch deviation values, and performing sequential deviation value fitting to generate a gear pitch deviation value sequence;
[0057] Step S53: performing deviation trend evolution analysis on the gear pitch deviation value sequence, thereby obtaining the gear pitch deviation trend law;
[0058] Step S54: performing global deviation superposition calculation on the gear pitch deviation value sequence to generate a gear pitch global deviation value.
[0059] The present invention obtains gear tooth number distribution data by performing tooth number distribution analysis on the gear three-dimensional point cloud model, understands the tooth number distribution of each gear, marks the gear tooth number distribution data with corresponding teeth based on multiple gear pitch deviation values, performs sequential deviation value fitting, establishes a gear pitch deviation value sequence, corresponds the pitch deviation with the specific tooth mark, and provides labeled data for subsequent deviation analysis, obtains the trend law of the gear pitch deviation by performing deviation trend evolution analysis on the gear pitch deviation value sequence, understands the changing trend of the pitch deviation, and provides guidance for formulating correction strategies, generates a global deviation value of the gear pitch by performing global deviation superposition calculation on the gear pitch deviation value sequence, comprehensively considers the influence of each deviation, and provides comprehensive data support for the correction of the overall pitch.
[0060] Preferably, the specific steps of step S6 are:
[0061] Step S61: optimizing the mold process parameters of the processing mold according to the gear pitch deviation trend law to generate the processing mold deviation optimization parameters;
[0062] Step S62: performing pitch deviation compensation calculation on the gear solid point cloud model based on the gear pitch global deviation value, thereby obtaining the gear pitch deviation compensation value;
[0063] Step S63: based on the machining mold deviation optimization parameters and the gear pitch deviation compensation value, a pitch deviation correction decision is made for the small modulus gear to be machined, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
[0064] The present invention optimizes machining mold parameters according to the trend law of gear pitch deviation, generates machining mold deviation optimization parameters, adjusts machining parameters, improves machining accuracy, reduces the generation of pitch deviation, performs pitch deviation compensation calculation on the gear three-dimensional point cloud model based on the gear pitch global deviation value, obtains the gear pitch deviation compensation value, performs fine deviation compensation on the gear according to the global deviation value, improves the accuracy and performance of the gear, makes pitch deviation correction decisions for small modulus gears to be processed based on machining mold deviation optimization parameters and gear pitch deviation compensation values, constructs an intelligent pitch deviation correction strategy, formulates an intelligent correction strategy according to the optimization parameters and compensation values, and improves the machining quality and accuracy of the gear.
[0065] In this specification, a device for correcting the pitch deviation of a small modulus gear based on visual measurement is provided, which is used to perform the method for correcting the pitch deviation of a small modulus gear based on visual measurement as described above, comprising:
[0066] The image enhancement module is used to obtain dynamic multi-directional images of gears and processing mold parameters; perform dynamic gamma correction on the dynamic multi-directional images of gears, and perform multi-scale local enhancement to construct detail-enhanced multi-directional images of gears;
[0067] The three-dimensional model module is used to calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform three-dimensional point cloud modeling based on the center position coordinates of each tooth to construct a three-dimensional point cloud model of the gear;
[0068] Theoretical reference module, used to calculate the intermediate reference of the processing mold parameters and generate the position coordinates of the intermediate reference surface; perform theoretical distance parameter modeling on the position coordinates of the intermediate reference surface to obtain a theoretical intermediate reference model;
[0069] A pitch deviation calculation module is used to calculate the pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating multiple gear pitch deviation values;
[0070] The global deviation module is used to perform global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value;
[0071] The deviation correction module is used to make pitch deviation correction decisions based on the gear pitch deviation trend law and the gear pitch global deviation value, and to build an intelligent pitch deviation correction strategy to perform small modulus gear pitch deviation correction operations.
[0072] The present invention realizes detail enhancement of dynamic multi-directional images of gears through dynamic gamma correction and multi-scale local enhancement, improves image quality and clarity, obtains more accurate multi-directional images of gears and processing mold parameters, and provides a high-quality data basis for subsequent steps. Through spatial center position calculation and three-dimensional point cloud modeling, the center position coordinates of each tooth are accurately calculated, and a three-dimensional point cloud model of the gear is constructed, which provides a gear model with a spatial stereoscopic perspective and provides an accurate data basis for subsequent pitch deviation calculation. Through mold intermediate benchmark calculation and theoretical distance parameterized modeling, a theoretical intermediate benchmark model is generated, which provides a theoretical benchmark reference for gear processing, ensures accurate theoretical modeling of processing mold parameters, and provides a basis for subsequent pitch deviation calculation. The necessary theoretical foundation is based on the theoretical intermediate benchmark model, and the pitch deviation is calculated on the gear three-dimensional point cloud model to generate accurate gear pitch deviation values, providing specific deviation values for each tooth, providing detailed data support for global deviation calculation, and performing superposition calculation based on multiple gear pitch deviation values to obtain the trend law and global deviation value of the gear pitch deviation, providing a global understanding of the overall gear pitch deviation situation, and providing an overall judgment basis for subsequent deviation correction. Based on the pitch deviation trend law and global deviation value, an intelligent pitch deviation correction strategy is constructed, and the small modulus gear pitch deviation correction operation is performed, realizing the intelligent correction of pitch deviation, ensuring that the quality and precision of the gear meet the requirements, and improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0073] Figure 1 A schematic flow chart of the steps of a method for correcting a pitch deviation of a small-module gear based on visual measurement according to the present invention;
[0074] Figure 2 Detailed implementation flow chart of step S1;
[0075] Figure 3 Detailed implementation flow chart of step S2;
[0076] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION
[0077] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0078] The present application example provides a method and device for correcting the pitch deviation of a small modulus gear based on visual measurement. The execution subject of the method and device for correcting the pitch deviation of a small modulus gear based on visual measurement includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio image management system, an information management system, and a cloud data management system.
[0079] See also Figures 1 to 4 The present invention provides a method for correcting a pitch deviation of a small-module gear based on visual measurement, and the method for correcting a pitch deviation of a small-module gear based on visual measurement comprises the following steps:
[0080] Step S1: obtaining a dynamic multi-directional image of the gear and processing mold parameters; performing dynamic gamma correction on the dynamic multi-directional image of the gear, and performing multi-scale local enhancement, thereby constructing a detail-enhanced multi-directional image of the gear;
[0081] Step S2: Calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform stereo point cloud modeling based on the center position coordinates of each tooth to construct a gear stereo point cloud model;
[0082] Step S3: performing mold intermediate reference calculation on the processing mold parameters to generate intermediate reference surface position coordinates; performing theoretical distance parameter modeling on the intermediate reference surface position coordinates to obtain a theoretical intermediate reference model;
[0083] Step S4: calculating the gear pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values;
[0084] Step S5: performing global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value;
[0085] Step S6: Based on the gear pitch deviation trend law and the gear pitch global deviation value, a pitch deviation correction decision is made, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
[0086] The present invention performs dynamic gamma correction and multi-scale local enhancement on the dynamic multi-directional image of the gear, effectively improving the clarity and detail of the image, thereby making the shape and structure of the gear more clearly visible, accurately acquiring various detailed features of the gear, including tooth shape, tooth pitch, etc., and providing a more accurate data basis for subsequent analysis and modeling. Through spatial center position calculation and three-dimensional point cloud modeling, the center position coordinates of each tooth are generated and a three-dimensional point cloud model of the gear is constructed. This three-dimensional point cloud model provides richer and three-dimensional gear shape information, making the understanding of the gear structure more comprehensive and in-depth, accurately evaluating the pitch deviation and shape deviation, and providing a reliable basis for optimization in the manufacturing process. Through mold intermediate reference calculation and theoretical distance parameterized modeling, a theoretical intermediate reference model is generated. This model provides a theoretical reference in gear manufacturing and identifies actual Deviations and errors in manufacturing. This information is crucial to understanding the accuracy and precision of gear manufacturing, and helps with adjustments and optimizations in the manufacturing process. Through pitch deviation calculation, the deviation of each tooth can be accurately evaluated, so as to understand the shape differences in actual gear manufacturing. This analysis finds potential problems and sources of errors in the manufacturing process, and provides a reference for improving manufacturing quality. Through global deviation superposition calculation and trend analysis, the laws and trends of gear pitch deviations are revealed, and common deviation patterns and problems in manufacturing are discovered, providing guidance for formulating correction strategies and optimization measures. Through intelligent pitch deviation correction strategies, correction decisions are made according to the trend laws of pitch deviations and global deviation values. This intelligent correction strategy adjusts pitch deviations to improve the accuracy and performance of gears. Through timely correction measures, product quality is guaranteed to meet requirements, and production efficiency and product reliability are improved.
[0087] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for correcting a pitch deviation of a small modulus gear based on visual measurement of the present invention. In this example, the steps of the method for correcting a pitch deviation of a small modulus gear based on visual measurement include:
[0088] Step S1: obtaining a dynamic multi-directional image of the gear and processing mold parameters; performing dynamic gamma correction on the dynamic multi-directional image of the gear, and performing multi-scale local enhancement, thereby constructing a detail-enhanced multi-directional image of the gear;
[0089] In this embodiment, a plurality of high-speed cameras are arranged around a small-modulus gear processing machine. These cameras can capture dynamic images of the gears during the processing process at a high frame rate (e.g., hundreds of frames per second). Cameras at different angles obtain a multi-directional image sequence of the gears. At the same time, it is also necessary to collect the operating parameters of the processing mold, such as the spindle speed, feed speed, etc. The dynamic image sequence is subjected to pixel brightness analysis frame by frame, the RGB value of each pixel is calculated, and converted into a corresponding brightness value, the brightness distribution characteristics of the entire image sequence, such as the mean value and variance, are analyzed, and a gamma correction algorithm is used to dynamically adjust the brightness of each frame of the image so that the brightness distribution of the entire image sequence is more uniform. Based on the brightness correction image sequence, a multi-scale local contrast enhancement algorithm, such as a Laplace pyramid and a Gaussian pyramid, is used to enhance the detailed texture characteristics of the gear surface in the image, and generate a clearer and sharper multi-directional image sequence of the gear.
[0090] Step S2: Calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform stereo point cloud modeling based on the center position coordinates of each tooth to construct a gear stereo point cloud model;
[0091] In this embodiment, the detail enhanced image sequence is analyzed, and the contour of each tooth in the image is identified by using image processing techniques such as edge detection and contour extraction, and the center of mass position coordinates of each tooth contour are calculated to obtain the center position of each tooth in the 2D image plane. The image sequences taken at multiple angles are combined, and the true coordinate position of each tooth in the 3D space is calculated through triangulation, depth estimation and other methods. The 3D coordinate points of all the teeth are combined to construct a complete three-dimensional point cloud model of the gear.
[0092] Step S3: performing mold intermediate reference calculation on the processing mold parameters to generate intermediate reference surface position coordinates; performing theoretical distance parameter modeling on the intermediate reference surface position coordinates to obtain a theoretical intermediate reference model;
[0093] In this embodiment, the geometric parameters of the processing mold, such as size and shape, are obtained through measurement and modeling. According to the structural characteristics of the mold, the appropriate intermediate reference plane position is determined. The spatial coordinates of the intermediate reference plane are calculated according to the mold parameters using mathematical calculation methods, and a three-dimensional coordinate system is established, in which the position of the intermediate reference plane is represented. According to the theoretical distance between the reference plane and other key surfaces, a parametric model is constructed. This model completely describes the positional relationship of the intermediate reference plane in space. The intermediate reference plane coordinate information obtained in the above steps and the theoretical distance parameterized model are integrated into a complete theoretical intermediate reference model. The model not only includes the absolute position of the intermediate reference plane, but also describes the relative positional relationship between it and other key surfaces. This theoretical model will serve as a reference in subsequent processing.
[0094] Step S4: calculating the gear pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values;
[0095] In this embodiment, based on the aligned theoretical benchmark model, the actual pitch of each tooth on the gear point cloud model is measured, the actual pitch is compared with the standard pitch preset in the theoretical benchmark model, and the pitch deviation value of each tooth is calculated to form a set of pitch deviation data. Statistical analysis is performed on multiple pitch deviation values, and a distribution histogram of the pitch deviation is plotted. The discrete degree of the deviation value is analyzed, and the tooth position with abnormal deviation is identified to provide a basis for subsequent optimization.
[0096] Step S5: performing global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value;
[0097] In this embodiment, according to the analysis of multiple groups of pitch deviation value sequences, time series analysis, grey prediction and other methods are used to analyze these pitch deviation value sequences, and the trend of their overall changes is excavated to provide a basis for predicting the future deviation evolution. To predict the evolution direction of the future pitch deviation, the arithmetic sum of all deviation values of the pitch deviation value sequence of each gear sample is calculated, and the global deviation value of each gear sample is calculated, that is, all pitch deviation values are added to obtain an indicator describing the overall deviation level of the gear.
[0098] Step S6: Based on the gear pitch deviation trend law and the gear pitch global deviation value, a pitch deviation correction decision is made, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
[0099] In this embodiment, the overall variation pattern of the pitch deviation under different batches or processes is understood, the future deviation evolution direction is predicted, the overall deviation level of the current gear sample is judged, and compared with the ideal state or standard requirements, and a set of targeted pitch deviation correction strategies are formulated, including correction amount, correction method, correction timing, etc. The overall deviation level of each gear sample is evaluated and compared with the ideal state, and a targeted intelligent correction decision strategy is formulated in combination with the deviation trend and the global deviation situation. According to the formulated correction strategy, accurate deviation correction operations are performed on small modulus gears.
[0100] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:
[0101] Step S11: real-time monitoring of the small modulus gear processing process, obtaining dynamic multi-directional images of the gear and processing mold parameters;
[0102] Step S12: Calculate the pixel brightness of the dynamic multi-directional image of the gear to obtain the pixel brightness value;
[0103] Step S13: performing image brightness distribution analysis on the pixel brightness values to generate image brightness distribution data;
[0104] Step S14: performing dynamic gamma correction on the dynamic multi-directional image of the gear according to the image brightness distribution data, thereby obtaining a brightness corrected image;
[0105] Step S15: Perform multi-scale local enhancement on the brightness-corrected image to construct a detail-enhanced gear multi-directional image.
[0106] In this embodiment, multiple high-speed cameras are arranged around the gear processing equipment to capture dynamic images of the gear during the processing at different angles. These images can comprehensively record the changing details of the gear surface. At the same time, it is also necessary to monitor and record the key process parameters of the gear processing equipment in real time, such as spindle speed, feed speed, cutting depth, etc. These parameters are closely related to the final gear quality. The collected dynamic image sequence and process parameter data are time-synchronized to provide basic data for subsequent image analysis and quality assessment. Based on each frame of dynamic image, the brightness value of each pixel is calculated. Commonly used brightness calculation methods include RGB weighted average, HSV color space brightness, etc. These methods obtain the precise brightness value of each pixel. The brightness values of all pixels are integrated into a two-dimensional matrix as the pixel brightness data of the frame image. According to the distribution of pixel brightness values, the brightness value of each pixel is calculated. , draw the brightness histogram of the image, analyze the morphological characteristics of the brightness histogram, such as the number of peaks, peak positions, histogram width, etc., understand the overall brightness distribution law of the image, organize these brightness distribution characteristic parameters into a data set as the basis for subsequent image enhancement, calculate the appropriate gamma correction parameters, and use the dynamic gamma correction algorithm to nonlinearly adjust the pixel brightness values of each frame of the image to make the overall brightness distribution of the image more uniform. Apply multi-resolution image enhancement algorithms, such as Laplace pyramid and bilateral filtering, to perform targeted enhancement on different scale areas of the brightness corrected image. These algorithms can highlight key information such as texture details and edge contours in the image, and highlight the tiny defect features on the gear surface. Finally, a multi-directional image sequence of gears with multi-scale local enhancement is obtained, providing high-quality input data for subsequent defect detection and quality analysis.
[0107] In this embodiment, the specific steps of step S15 are:
[0108] Performing discrete wavelet transform on the brightness correction image to obtain component images of different frequency bands; the component images of different frequency bands include a high-frequency band detail layer image and a low-frequency band rough layer image;
[0109] Perform multi-scale detail recognition on the high-frequency band detail layer image to obtain multiple detail features of the image;
[0110] Perform large-scale corrosion denoising on the high-frequency band detail layer image to obtain the denoised component image;
[0111] Based on multiple detail features of the image, the denoised component image is locally expanded to generate a local detail enhanced component image;
[0112] Wavelet image reconstruction is performed on the local detail enhanced component image and the low-frequency band rough layer image to construct a detail enhanced multi-directional image of the gear.
[0113] In this embodiment, the image is decomposed into a plurality of sub-band signals of different frequencies, including a high-frequency detail layer and a low-frequency coarse layer. The high-frequency band contains the detail information of the image, and the low-frequency band contains the overall structural information of the image. The high-frequency band detail layer image reflects the detail features of the image, and the low-frequency band coarse layer image reflects the overall structural information of the image. For the high-frequency band detail layer image, a multi-scale feature extraction algorithm (such as Gabor filtering, SIFT, etc.) is used to identify detail features of different scales and directions in the image, such as texture, edge, corner point, etc., and these detail features are organized into a feature set to provide a basis for subsequent detail enhancement processing. For the nodal layer image, a large-scale morphological corrosion filter is applied. This operation can effectively remove the noise interference in the image and retain the main detail information of the image to obtain the component image after the noise is removed, so as to prepare for the subsequent detail enhancement processing. The local area of the denoised component image is subjected to detail expansion processing in a targeted manner to highlight and enhance the key texture, edge and other detail information in the image to obtain the component image after local detail enhancement. The wavelet reconstruction algorithm is used to fuse and reconstruct the two component images to obtain the final detail enhanced gear multi-directional image sequence with rich details and less noise. These images can provide high-quality input data for subsequent defect detection and quality analysis.
[0114] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:
[0115] Step S21: performing dynamic continuous frame analysis on the detail-enhanced gear multi-directional image to generate a gear multi-directional image frame sequence;
[0116] Step S22: visually identifying each tooth on the gear multi-directional image frame sequence and marking tooth feature points;
[0117] Step S23: Calculate the spatial center position of the tooth feature points to generate the center position coordinates of each tooth;
[0118] Step S24: performing a three-dimensional space registration analysis on the gear multi-directional image frame sequence according to the center position coordinates of each tooth, thereby generating three-dimensional space registration data of tooth feature points;
[0119] Step S25: performing gear number layout structure analysis on the gear multi-directional image frame sequence to obtain gear number layout structure data;
[0120] Step S26: Performing three-dimensional point cloud modeling on the gear tooth number layout structure data based on the three-dimensional spatial registration data of the tooth feature points to construct a three-dimensional point cloud model of the gear.
[0121] In this embodiment, video processing technology is used to extract continuous image frames to form a complete image frame sequence. This image frame sequence is used for subsequent gear dynamic analysis. An image recognition algorithm is applied to locate and identify each tooth of the gear in each frame of the image, and key feature points such as tooth top and tooth root are marked on the contour of each tooth to obtain the feature point position information of all teeth in each frame of the image, calculate the geometric center position of each tooth, and record the center point coordinate information of each tooth to form a coordinate data set. These coordinate data reflect the position distribution of each tooth of the gear in space. According to the center coordinate information of each tooth, three-dimensional spatial matching is performed on the multi-view image frames. The corresponding relationship of each tooth in images of different viewing angles is determined, and the registration data of the tooth feature points in three-dimensional space is generated. These data lay the foundation for the subsequent construction of a three-dimensional point cloud model of the gear. The distribution of each tooth in the multi-directional image sequence of the gear is analyzed to determine the total number of teeth of the gear. The layout and arrangement rules of each tooth around the gear are further analyzed to generate the tooth number layout structure data. These data reflect the overall geometric structure characteristics of the gear. According to the obtained three-dimensional registration data of the tooth feature points, it is fused with the tooth number layout structure data, and a complete three-dimensional point cloud model of the gear is constructed using the three-dimensional point cloud modeling algorithm. The model contains the precise geometric shape and structural information of the gear in three-dimensional space.
[0122] In this embodiment, the specific steps of step S24 are:
[0123] Perform gear contour line recognition on the gear multi-directional image frame sequence and extract the gear contour lines of multiple frames;
[0124] Perform multi-vision posture difference analysis on multi-frame gear contour lines to generate multi-frame contour line posture difference data;
[0125] According to the center position coordinates of each tooth, a multi-posture coordinate transformation calculation is performed on the multi-frame contour line posture difference data to obtain multiple visual contour coordinates;
[0126] Performing stereoscopic spatial registration of tooth feature points based on multiple visual contour coordinates to obtain stereoscopic feature parameters of the tooth feature points;
[0127] The stereoscopic registration error is corrected for the stereoscopic feature parameters of the tooth feature points, thereby generating stereoscopic space registration data of the tooth feature points.
[0128] In this embodiment, an image edge detection algorithm is applied to extract the gear contour in each frame of the image, and contour recognition and extraction are performed on each frame of the image to obtain a series of gear contour data. These contour data reflect the changes in the geometric shape of the gear at different times. The relative posture difference between the frames is calculated based on the extracted multi-frame gear contour data, and the characteristic parameters such as the angle and position of the contour are used to quantify the posture difference. The posture differences between all frames are organized into a data set as a basis for subsequent analysis. The obtained center position coordinates of each tooth are used to perform coordinate transformation on the posture difference data. Through coordinate transformation, the multiple frames are transformed into a data set. The frame contour line data is mapped to a unified three-dimensional coordinate system to obtain unified coordinate data reflecting the geometric shape of the multi-view contour line. Based on the multi-view contour line coordinate data, it is associated with the tooth feature point information. The corresponding relationship of each tooth feature point under different viewing angles is determined through the three-dimensional geometric alignment algorithm, and the precise position and posture parameters of these tooth feature points in the three-dimensional space are extracted. The obtained three-dimensional feature parameters of the tooth feature points are analyzed, and the alignment accuracy is evaluated. The optimization algorithm is used to correct and calibrate the existing alignment errors, and the error-corrected three-dimensional space alignment data of the tooth feature points is obtained, providing high-precision input for subsequent three-dimensional modeling.
[0129] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:
[0130] Step S31: performing mold intermediate reference calculation on the processing mold parameters, thereby obtaining the mold intermediate reference value;
[0131] Step S32: Projecting the intermediate reference value of the mold onto a reference plane to generate the position coordinates of the intermediate reference plane;
[0132] Step S33: Calculating the theoretical distance of the gear position on the intermediate reference surface position coordinates based on the preset gear processing parameters, thereby obtaining the distance value from each gear to the reference surface;
[0133] Step S34: perform parameterized modeling calculation on the distance values from each gear to the reference surface to obtain a theoretical intermediate reference model.
[0134] In this embodiment, the geometric parameters of the mold used for gear processing are collected, and the intermediate reference position of the mold relative to the machining center is calculated based on these parameters to obtain the intermediate reference value of the mold as a reference for subsequent steps. Based on the intermediate reference value of the mold, it is projected onto the reference plane for gear processing, and the intermediate reference position coordinates on the reference plane are determined as the reference position of each tooth of the gear. These coordinate information reflects the theoretical position distribution of the gear during the processing. The preset parameters of gear processing, such as the number of teeth and module, are used to calculate the theoretical position of each tooth on the reference plane. For each tooth position coordinate, its theoretical distance value relative to the reference plane is calculated to obtain a set of distance data reflecting the theoretical positions of each tooth of the gear. Through parametric modeling, these discrete distance data are converted into a continuous mathematical model, which describes the intermediate reference geometry of the gear under ideal conditions, i.e., the theoretical intermediate reference model.
[0135] In this embodiment, step S4 includes the following steps:
[0136] Step S41: performing overall position deviation registration of the gear stereo point cloud model according to the theoretical intermediate reference model to obtain a reference registered gear model;
[0137] Step S42: extracting the position of each tooth vertex of the reference registration gear model one by one, and generating the point cloud coordinates of each tooth vertex;
[0138] Step S43: Calculate the pitch deviation of the tooth vertex point cloud coordinates based on the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values.
[0139] In this embodiment, the actual gear 3D point cloud model is aligned with the theoretical intermediate reference model in terms of overall position and posture. The overall position deviation of the actual gear model relative to the theoretical model is eliminated through the alignment algorithm to obtain the 3D gear model after reference alignment. The 3D gear model after reference alignment is analyzed to extract the tooth vertex position coordinates of each tooth. This requires the use of point cloud segmentation, feature extraction and other technologies to accurately determine the three-dimensional coordinates of each tooth vertex, and organize the coordinate information of all tooth vertices into a point cloud data set to prepare for subsequent deviation analysis. With reference to the standard pitch of the theoretical intermediate reference model, the deviation of the actual tooth vertex coordinates relative to the theoretical model is calculated. For each tooth vertex position, its deviation from the theoretical pitch is calculated to form a set of pitch deviation values. These deviation values reflect the degree of difference between the actual gear manufacturing quality and the theoretical model, providing a basis for subsequent defect diagnosis.
[0140] In this embodiment, step S5 includes the following steps:
[0141] Step S51: performing a gear number distribution analysis on the gear 3D point cloud model to obtain gear tooth number distribution data;
[0142] Step S52: marking the gear tooth number distribution data with corresponding teeth based on multiple gear pitch deviation values, and performing sequential deviation value fitting to generate a gear pitch deviation value sequence;
[0143] Step S53: performing deviation trend evolution analysis on the gear pitch deviation value sequence, thereby obtaining the gear pitch deviation trend law;
[0144] Step S54: performing global deviation superposition calculation on the gear pitch deviation value sequence to generate a gear pitch global deviation value.
[0145] In this embodiment, the distribution of point clouds contained in the three-dimensional point cloud model of the gear is analyzed, and the point cloud data is divided into subsets corresponding to each tooth through point cloud segmentation and clustering algorithms. The point cloud distribution characteristics of each tooth are statistically obtained to form the gear tooth number distribution data. These data reflect the distribution layout of each tooth in the actual gear. By analyzing the corresponding relationship, the corresponding deviation value is assigned to each tooth, and the deviation values of all teeth are arranged in the order of the teeth to form a continuous deviation value sequence. The sequence is fitted and analyzed to obtain a smoother pitch deviation value sequence. The generated pitch deviation value sequence is analyzed and its change trend is evaluated. Time series analysis, Fourier analysis and other methods can be used to extract the evolution law of the deviation value and determine the overall change trend of the pitch deviation in the circumferential direction of the gear, so as to provide a basis for subsequent fault diagnosis. Based on the obtained pitch deviation value sequence, the contribution of each pitch deviation to the overall gear is calculated. By weighted superposition, a global deviation value reflecting the overall deviation level of the gear is obtained. This global deviation value can be used as a comprehensive indicator for evaluating the manufacturing quality of the gear.
[0146] In this embodiment, step S6 includes the following steps:
[0147] Step S61: optimizing the mold process parameters of the processing mold according to the gear pitch deviation trend law to generate the processing mold deviation optimization parameters;
[0148] Step S62: performing pitch deviation compensation calculation on the gear solid point cloud model based on the gear pitch global deviation value, thereby obtaining the gear pitch deviation compensation value;
[0149] Step S63: based on the machining mold deviation optimization parameters and the gear pitch deviation compensation value, a pitch deviation correction decision is made for the small modulus gear to be machined, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
[0150] In this embodiment, according to the trend law of gear pitch deviation, the mold process parameters used in gear processing are optimized and adjusted in a targeted manner. After optimization calculation, a set of processing mold parameters that can suppress pitch deviation is obtained. The optimization of the mold process parameters is specifically as follows: the cavity processing technology is optimized, the assembly method of the cavity and the rear mold core is changed, and the positioning of the cavity is the outer ring positioning. Therefore, the tooth shape of the cavity is first processed, and then the tooth top runout is corrected to 0.001mm by the micrometer, and the center of the cavity is found. Then, the outer circle is ground based on this, so that the eccentricity between the cavity gear ring and the reference is controlled below 0.002mm, which greatly improves the concentricity between the cavity gear ring and the reference. Based on the obtained global deviation value of the gear, the required deviation compensation is calculated, and this compensation is applied to the constructed gear three-dimensional point cloud model. The coordinates of each tooth vertex are adjusted. Through this deviation compensation, the overall size deviation existing in the actual gear model is corrected. The obtained optimized mold parameters are integrated with the calculated deviation compensation value for analysis, and a complete set of intelligent decision-making strategies are established to guide the actual processing of small-module gears. Real-time monitoring is performed during the processing, and mold parameters and processes are automatically adjusted according to the decision-making strategy, thereby suppressing and eliminating pitch deviation. Through this intelligent deviation correction, high-quality small-module gears are produced.
[0151] In this embodiment, a small-module gear pitch deviation correction device based on visual measurement is provided, which is used to execute the small-module gear pitch deviation correction method based on visual measurement as described above, comprising:
[0152] The image enhancement module is used to obtain dynamic multi-directional images of gears and processing mold parameters; perform dynamic gamma correction on the dynamic multi-directional images of gears, and perform multi-scale local enhancement to construct detail-enhanced multi-directional images of gears;
[0153] The three-dimensional model module is used to calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform three-dimensional point cloud modeling based on the center position coordinates of each tooth to construct a three-dimensional point cloud model of the gear;
[0154] Theoretical reference module, used to calculate the intermediate reference of the processing mold parameters and generate the position coordinates of the intermediate reference surface; perform theoretical distance parameter modeling on the position coordinates of the intermediate reference surface to obtain a theoretical intermediate reference model;
[0155] A pitch deviation calculation module is used to calculate the pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating multiple gear pitch deviation values;
[0156] The global deviation module is used to perform global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value;
[0157] The deviation correction module is used to make pitch deviation correction decisions based on the gear pitch deviation trend law and the gear pitch global deviation value, and to build an intelligent pitch deviation correction strategy to perform small modulus gear pitch deviation correction operations.
[0158] The present invention realizes detail enhancement of dynamic multi-directional images of gears through dynamic gamma correction and multi-scale local enhancement, improves image quality and clarity, obtains more accurate multi-directional images of gears and processing mold parameters, and provides a high-quality data basis for subsequent steps. Through spatial center position calculation and three-dimensional point cloud modeling, the center position coordinates of each tooth are accurately calculated, and a three-dimensional point cloud model of the gear is constructed, which provides a gear model with a spatial stereoscopic perspective and provides an accurate data basis for subsequent pitch deviation calculation. Through mold intermediate benchmark calculation and theoretical distance parameterized modeling, a theoretical intermediate benchmark model is generated, which provides a theoretical benchmark reference for gear processing, ensures accurate theoretical modeling of processing mold parameters, and provides a basis for subsequent pitch deviation calculation. The necessary theoretical foundation is based on the theoretical intermediate benchmark model, and the pitch deviation is calculated on the gear three-dimensional point cloud model to generate accurate gear pitch deviation values, providing specific deviation values for each tooth, providing detailed data support for global deviation calculation, and performing superposition calculation based on multiple gear pitch deviation values to obtain the trend law and global deviation value of the gear pitch deviation, providing a global understanding of the overall gear pitch deviation situation, and providing an overall judgment basis for subsequent deviation correction. Based on the pitch deviation trend law and global deviation value, an intelligent pitch deviation correction strategy is constructed, and the small modulus gear pitch deviation correction operation is performed, realizing the intelligent correction of pitch deviation, ensuring that the quality and precision of the gear meet the requirements, and improving production efficiency and product quality.
[0159] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.
[0160] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A method for correcting the pitch deviation of small modulus gears based on visual measurement. It is characterized in that The following steps are involved: Step S1: obtaining a dynamic multi-directional image of the gear and processing mold parameters; performing dynamic gamma correction on the dynamic multi-directional image of the gear, and performing multi-scale local enhancement, thereby constructing a detail-enhanced multi-directional image of the gear; Step S2: Calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform stereo point cloud modeling based on the center position coordinates of each tooth to construct a gear stereo point cloud model; Step S3: performing mold intermediate reference calculation on the processing mold parameters to generate intermediate reference surface position coordinates; performing theoretical distance parameter modeling on the intermediate reference surface position coordinates to obtain a theoretical intermediate reference model; Among them, the specific steps of step S3 are: Step S31: performing mold intermediate reference calculation on the processing mold parameters, thereby obtaining the mold intermediate reference value; Step S32: Projecting the intermediate reference value of the mold onto a reference plane to generate the position coordinates of the intermediate reference plane; Step S33: Calculating the theoretical distance of the gear position on the intermediate reference surface position coordinates based on the preset gear processing parameters, thereby obtaining the distance value from each gear to the reference surface; Step S34: performing parameterized modeling calculation on the distance values from each gear to the reference surface to obtain a theoretical intermediate reference model; Step S4: calculating the gear pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values; The specific steps of step S4 are: Step S41: performing overall position deviation registration of the gear stereo point cloud model according to the theoretical intermediate reference model to obtain a reference registered gear model; Step S42: extracting the position of each tooth vertex of the reference registration gear model one by one, and generating the point cloud coordinates of each tooth vertex; Step S43: calculating the pitch deviation of the tooth vertex point cloud coordinates based on the theoretical intermediate reference model, thereby generating a plurality of gear pitch deviation values; Step S5: performing global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value; The specific steps of step S5 are: Step S51: performing a gear number distribution analysis on the gear 3D point cloud model to obtain gear tooth number distribution data; Step S52: marking the gear tooth number distribution data with corresponding teeth based on multiple gear pitch deviation values, and performing sequential deviation value fitting to generate a gear pitch deviation value sequence; Step S53: performing deviation trend evolution analysis on the gear pitch deviation value sequence, thereby obtaining the gear pitch deviation trend law; Step S54: performing global deviation superposition calculation on the gear pitch deviation value sequence to generate a gear pitch global deviation value; Step S6: Based on the gear pitch deviation trend law and the gear pitch global deviation value, a pitch deviation correction decision is made, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
2. According to the method for correcting the pitch deviation of small modulus gears based on visual measurement according to claim 1, It is characterized in that The specific steps of step S1 are: Step S11: real-time monitoring of the small modulus gear processing process, obtaining dynamic multi-directional images of the gear and processing mold parameters; Step S12: Calculate the pixel brightness of the dynamic multi-directional image of the gear to obtain the pixel brightness value; Step S13: performing image brightness distribution analysis on the pixel brightness values to generate image brightness distribution data; Step S14: performing dynamic gamma correction on the dynamic multi-directional image of the gear according to the image brightness distribution data, thereby obtaining a brightness corrected image; Step S15: Perform multi-scale local enhancement on the brightness-corrected image to construct a detail-enhanced gear multi-directional image.
3. According to claim 2, the method for correcting the pitch deviation of small modulus gears based on visual measurement, It is characterized in that The specific steps of step S15 are: Performing discrete wavelet transform on the brightness correction image to obtain component images of different frequency bands; the component images of different frequency bands include a high-frequency band detail layer image and a low-frequency band rough layer image; Perform multi-scale detail recognition on the high-frequency band detail layer image to obtain multiple detail features of the image; Perform large-scale corrosion denoising on the high-frequency band detail layer image to obtain the denoised component image; Based on multiple detail features of the image, the denoised component image is locally expanded to generate a local detail enhanced component image; Wavelet image reconstruction is performed on the local detail enhanced component image and the low-frequency band rough layer image to construct a detail enhanced multi-directional image of the gear.
4. The method for correcting the pitch deviation of small-module gears based on visual measurement according to claim 1, It is characterized in that The specific steps of step S2 are: Step S21: performing dynamic continuous frame analysis on the detail-enhanced gear multi-directional image to generate a gear multi-directional image frame sequence; Step S22: visually identifying each tooth on the gear multi-directional image frame sequence and marking tooth feature points; Step S23: Calculate the spatial center position of the tooth feature points to generate the center position coordinates of each tooth; Step S24: performing a three-dimensional space registration analysis on the gear multi-directional image frame sequence according to the center position coordinates of each tooth, thereby generating three-dimensional space registration data of tooth feature points; Step S25: performing gear number layout structure analysis on the gear multi-directional image frame sequence to obtain gear tooth number layout structure data; Step S26: Performing three-dimensional point cloud modeling on the gear tooth number layout structure data based on the three-dimensional spatial registration data of the tooth feature points to construct a three-dimensional point cloud model of the gear.
5. The method according to claim 4, It is characterized in that The specific steps of step S24 are: Perform gear contour line recognition on the gear multi-directional image frame sequence and extract the gear contour lines of multiple frames; Perform multi-vision posture difference analysis on multi-frame gear contour lines to generate multi-frame contour line posture difference data; According to the center position coordinates of each tooth, a multi-posture coordinate transformation calculation is performed on the multi-frame contour line posture difference data, thereby obtaining multiple visual contour coordinates; Performing stereoscopic spatial registration of tooth feature points based on multiple visual contour coordinates to obtain stereoscopic feature parameters of the tooth feature points; The stereoscopic registration error is corrected for the stereoscopic feature parameters of the tooth feature points, thereby generating stereoscopic space registration data of the tooth feature points.
6. The method for correcting the pitch deviation of small-module gears based on visual measurement according to claim 1, It is characterized in that The specific steps of step S6 are: Step S61: optimizing the mold process parameters of the processing mold according to the gear pitch deviation trend law to generate the processing mold deviation optimization parameters; Step S62: performing pitch deviation compensation calculation on the gear solid point cloud model based on the gear pitch global deviation value, thereby obtaining the gear pitch deviation compensation value; Step S63: based on the machining mold deviation optimization parameters and the gear pitch deviation compensation value, a pitch deviation correction decision is made for the small modulus gear to be machined, and an intelligent pitch deviation correction strategy is constructed to perform the small modulus gear pitch deviation correction operation.
7. A small modulus gear pitch deviation correction device based on visual measurement, It is characterized in that The method for correcting the pitch deviation of a small-module gear based on visual measurement according to claim 1 comprises: The image enhancement module is used to obtain dynamic multi-directional images of gears and processing mold parameters; perform dynamic gamma correction on the dynamic multi-directional images of gears, and perform multi-scale local enhancement to construct detail-enhanced multi-directional images of gears; The three-dimensional model module is used to calculate the spatial center position of the detail-enhanced gear multi-directional image to generate the center position coordinates of each tooth; perform three-dimensional point cloud modeling based on the center position coordinates of each tooth to construct a three-dimensional point cloud model of the gear; Theoretical reference module, used to calculate the intermediate reference of the processing mold parameters and generate the position coordinates of the intermediate reference surface; perform theoretical distance parameter modeling on the position coordinates of the intermediate reference surface to obtain a theoretical intermediate reference model; A pitch deviation calculation module is used to calculate the pitch deviation of the gear solid point cloud model according to the theoretical intermediate reference model, thereby generating multiple gear pitch deviation values; The global deviation module is used to perform global deviation superposition calculation based on multiple gear pitch deviation values to obtain the gear pitch deviation trend law and the gear pitch global deviation value; The deviation correction module is used to make pitch deviation correction decisions based on the gear pitch deviation trend law and the gear pitch global deviation value, and to build an intelligent pitch deviation correction strategy to perform small modulus gear pitch deviation correction operations.
