A method and system for intelligent acceptance of machine tool assembly based on computer vision

Abstract

The present invention is applicable to the technical field of machine tool assembly, and provides a method and system for intelligent acceptance of machine tool assembly based on computer vision, which can obtain detection standards and visual detection data; obtain visual results, and preprocess the visual results, wherein the preprocessing includes noise reduction processing and feature extraction; obtain detection results according to the visual results, obtain detection standards, wherein the detection standards include geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection, compare the detection results with the detection standards, and judge whether the assembly results are qualified; obtain acceptance standards, judge whether the qualified assembly can be accepted according to the acceptance standards, obtain the reasons for failure, obtain improvement methods, and send the improvement methods to a terminal, thereby solving the problems that the existing intelligent assembly of machine tools cannot perform real-time detection of assembly effect, assembly progress and machine tool quality, and cannot timely provide processing means and processing methods for the reasons for failure in detection.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine tool assembly, and in particular relates to a computer vision-based intelligent acceptance method and system for machine tool assembly. Background Art

[0002] In modern industrial manufacturing, the quality of machine tool assembly directly determines the processing accuracy, stability and service life of the equipment. Existing assembly inspection methods mainly rely on manual measurement and single precision analysis tools, such as micrometers, laser interferometers and optical microscopes. Although they can meet some geometric accuracy inspection requirements, it is difficult to achieve a comprehensive evaluation of dynamic performance, assembly clearance and surface quality. In addition, the assembly process lacks a real-time feedback mechanism, resulting in errors that cannot be corrected in the early stages, thus affecting the performance of the entire machine and the efficiency of acceptance.

[0003] In the existing technology, technologies such as high-speed cameras and vibration sensors are used for dynamic performance testing, and multi-spectral and 3D scanning technologies are used for surface and assembly status analysis. However, these technologies are usually applied in a scattered manner, lack linkage, and rely more on experience-based operations during the assembly process, making it difficult to achieve automation and precise adjustment. At the same time, current acceptance standards usually focus on static indicators and ignore the impact of actual operating conditions on assembly quality, resulting in problems such as reduced accuracy and performance failure of machine tools during long-term use.

[0004] However, in the existing intelligent assembly process of machine tools, there is a problem that the assembly effect, assembly progress and machine tool quality cannot be detected in real time. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a method for intelligent acceptance of machine tool assembly based on computer vision, aiming to solve the problems raised in the background technology.

[0006] The embodiment of the present invention is implemented as follows: a method for intelligent acceptance of machine tool assembly based on computer vision, the method comprising: Acquire inspection standards and obtain visual inspection data, wherein the visual inspection data includes visual inspection equipment and equipment parameters; Acquire visual results, the visual results include assembly structures, moving parts and internal cavities, and preprocess the visual results, the preprocessing includes noise reduction and feature extraction; Obtain the test results based on the visual results, obtain the test standards, which include geometric accuracy test, dynamic performance test, surface quality test and assembly test, compare the test results with the test standards, and determine whether the assembly results are qualified; Obtain the acceptance criteria, determine whether the qualified assembly can be accepted according to the acceptance criteria, obtain the reasons for failure, obtain the improvement methods, and send the improvement methods to the terminal.

[0007] Preferably, the visual result is obtained, the visual result includes the assembly structure, the moving parts and the internal cavity, and the visual result is preprocessed, and the preprocessing includes the steps of noise reduction and feature extraction, specifically including: Obtaining visual results, including assembly structures, moving parts and internal cavities, and recording the motion process through the visual results, wherein the motion process includes key actions during startup, processing and stopping; Preprocessing the visual results, wherein the preprocessing includes noise reduction and feature extraction, to obtain preprocessing results; Obtain multiple preprocessing results, compare the multiple preprocessing results, and obtain the best preprocessing result.

[0008] Preferably, the step of obtaining the test results according to the visual results, acquiring the test standards, wherein the test standards include geometric accuracy test, dynamic performance test, surface quality test and assembly test, and comparing the test results with the test standards to determine whether the assembly results are qualified, specifically includes: Obtaining the test results according to the best preprocessing results and obtaining the test standards, wherein the test standards include geometric accuracy test, dynamic performance test, surface quality test and assembly test; Compare the test results with the test standards to determine whether the assembly results are qualified, and obtain error information if the test fails; If the test result is determined to be greater than the test standard, the test is considered qualified.

[0009] Preferably, the steps of obtaining the acceptance criteria, determining whether the qualified assembly can be accepted according to the acceptance criteria, obtaining the reasons for failure, obtaining the improvement method, and sending the improvement method to the terminal specifically include: Obtain the acceptance criteria and determine whether the qualified assembly can be accepted based on the acceptance criteria. The acceptance criteria are greater than the inspection criteria. Obtain the reasons for non-conformity and the improvement methods, which include reassembly, improvement of matching design, surface repair and improvement of installation process; Determine the specific improvement method based on the cause of non-conformity and send the specific improvement method to the terminal.

[0010] Preferably, the inspection standard is an inspection standard for machine tool assembly.

[0011] Another object of an embodiment of the present invention is to provide a machine tool assembly intelligent acceptance system based on computer vision, the system comprising: A standard module, which obtains the inspection standard and the visual inspection data, wherein the visual inspection data includes the visual inspection equipment and equipment parameters; A preprocessing module, which obtains visual results, including assembly structures, moving parts and internal cavities, and preprocesses the visual results, including noise reduction and feature extraction; The inspection module obtains the inspection results according to the visual results, obtains the inspection standards, which include geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection, compares the inspection results with the inspection standards, and determines whether the assembly results are qualified; The acceptance module obtains the acceptance criteria, determines whether the qualified assembly can be accepted according to the acceptance criteria, obtains the reasons for failure, obtains the improvement methods, and sends the improvement methods to the terminal.

[0012] Preferably, the preprocessing module comprises: A visual result unit, which obtains visual results, including assembly structure, moving parts and internal cavity, and records the movement process through the visual results, and the movement process includes key actions during startup, processing and stopping; A preprocessing unit, which preprocesses the visual result, wherein the preprocessing includes noise reduction processing and feature extraction, and obtains a preprocessing result; The best unit obtains multiple preprocessing results, compares the multiple preprocessing results, and obtains the best preprocessing result.

[0013] Preferably, the detection module comprises: A detection unit, which obtains a detection result according to the best preprocessing result and acquires a detection standard, wherein the detection standard includes geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection; The judgment unit compares the test results with the test standards, judges whether the assembly results are qualified, and obtains error information if the test fails; For a qualified unit, if the test result is determined to be greater than the test standard, it is determined to be qualified.

[0014] Preferably, the acceptance module includes: Acceptance standard unit, obtains acceptance standards, and determines whether qualified assembly can be accepted based on the acceptance standards. The acceptance standards are greater than the inspection standards. Improve the unit, obtain the reasons for non-conformity, and obtain improvement methods, which include reassembly, improvement of matching design, surface repair, and improvement of installation process; The terminal unit determines the specific improvement method according to the non-conformity reason and sends the specific improvement method to the terminal.

[0015] Preferably, the inspection standard is an inspection standard for machine tool assembly.

[0016] An embodiment of the present invention provides an intelligent acceptance method for machine tool assembly based on computer vision, which includes obtaining a detection standard, obtaining visual detection data, the visual detection data including visual detection equipment and equipment parameters, obtaining a visual result, the visual result including an assembly structure, moving parts and an internal cavity, recording a movement process through the visual result, preprocessing the visual result, the preprocessing including noise reduction and feature extraction, obtaining a preprocessing result, obtaining multiple preprocessing results, comparing the multiple preprocessing results, obtaining an optimal preprocessing result, obtaining a detection result according to the optimal preprocessing result, obtaining a detection standard, comparing the detection result with the detection standard, and judging whether the assembly result is qualified. According to the unqualified test, error information is obtained. If the test result is determined to be greater than the test standard, the test is determined to be qualified, and the acceptance standard is obtained. According to the acceptance standard, it is determined whether the qualified assembly can be accepted. If the acceptance standard is greater than the test standard, the unqualified reason is obtained, and the improvement method is obtained. The improvement method includes reassembly, improvement of matching design, repair of surface and improvement of installation process. The specific improvement method is determined according to the unqualified reason, and the specific improvement method is sent to the terminal, which solves the problem that the existing machine tool intelligent assembly process cannot perform real-time detection of assembly effect, assembly progress and machine tool quality, and cannot promptly provide processing means and processing methods for the cause of unqualified test. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method for intelligent acceptance of machine tool assembly based on computer vision provided by an embodiment of the present invention; Figure 2 A flowchart of the steps of obtaining visual results and preprocessing the visual results provided by an embodiment of the present invention; Figure 3 A flowchart of the steps of obtaining a test result based on a visual result, obtaining a test standard, and determining whether an assembly result is qualified according to the test provided by an embodiment of the present invention; Figure 4 A flowchart of the steps of obtaining acceptance criteria, obtaining reasons for non-conformity, and obtaining improvement methods provided in an embodiment of the present invention; Figure 5 An architecture diagram of a machine tool assembly intelligent acceptance system based on computer vision provided by an embodiment of the present invention; Figure 6 An architectural diagram of a preprocessing module provided in an embodiment of the present invention; Figure 7 An architectural diagram of a detection module provided in an embodiment of the present invention; Figure 8 This is an architecture diagram of the acceptance module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0020] like Figure 1 As shown, a method for intelligent acceptance of machine tool assembly based on computer vision is provided in an embodiment of the present invention, and the method includes: S100, acquiring inspection standards and visual inspection data, wherein the visual inspection data includes visual inspection equipment and equipment parameters.

[0021] In this step, the inspection standards are obtained. The inspection standards are the target basis of visual inspection, clarifying the allowable range and inspection requirements of various parameters, obtaining the CAD drawings and technical documents of the machine tool assembly, clarifying the size, geometric accuracy, tolerance and matching requirements, and at the same time, it is necessary to refer to the special requirements proposed and formulate personalized inspection standards; Obtain visual inspection data. Visual inspection data is the basis for realizing automated inspection, including the entire process from equipment selection to parameter configuration. Visual inspection data includes visual inspection equipment and equipment parameters. Visual inspection equipment includes industrial cameras and infrared cameras. Industrial cameras are used to collect images of machine tool appearance, surface defects or key dimensions. Infrared cameras are used to detect specific problems such as coating thickness, cracks, thermal effects, etc. The configuration of equipment parameters directly determines the accuracy and efficiency of detection. Camera parameters include resolution, exposure time and frame rate. Machine tool assembly detection based on computer vision can not only accurately capture complex assembly problems, but also improve efficiency and detection quality through standardized and automated processes.

[0022] S200, obtaining visual results, wherein the visual results include assembly structures, moving parts, and internal cavities, and preprocessing the visual results, wherein the preprocessing includes noise reduction processing and feature extraction.

[0023] In this step, visual results are obtained, including assembly structure, moving parts and internal cavity. The assembly structure is used to obtain the appearance data of the overall assembly of the machine tool, verify the assembly gap, joint position and surface defects; the moving parts include obtaining the dynamic state data of the moving parts such as the spindle, guide rail, slider, etc., and evaluating their running stability and accuracy; the internal cavity includes detecting whether there are processing defects, contamination or assembly abnormalities in the internal cavity of the machine tool, such as the bearing cavity and the transmission cavity; Preprocess the visual results, including noise reduction and feature extraction. Visual data may be affected by external ambient light, equipment noise or acquisition accuracy. Noise reduction can improve the quality of images and point clouds. Feature extraction is used to extract the core information of assembly structures, moving parts and cavities from the denoised visual data for subsequent analysis.

[0024] S300, obtaining the inspection result according to the visual result, obtaining the inspection standard, which includes geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection, comparing the inspection result with the inspection standard, and judging whether the assembly result is qualified.

[0025] In this step, the test results are obtained based on the visual results. After the assembly test results are obtained through visual inspection technology, the test results are compared with the predefined test standards to determine whether the assembly quality is qualified. Geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection are performed through visual inspection equipment; Obtain the inspection standards. The inspection standards are the basis for judging whether the assembly is qualified, including design documents, industry standards and user requirements. Compare the inspection results with the inspection standards, compare the inspection results with the inspection standards item by item, record whether the qualified conditions are met, and judge whether the assembly results are qualified. Qualified means that all inspection results meet the inspection standards, and unqualified means that any inspection result exceeds the inspection standard range.

[0026] S400, obtaining acceptance criteria, determining whether qualified assemblies can be accepted according to the acceptance criteria, obtaining reasons for non-compliance, obtaining improvement methods, and sending the improvement methods to the terminal.

[0027] In this step, the acceptance criteria are obtained. The formulation of the acceptance criteria includes design drawings and technical documents, industry and national standards, and customer customization requirements. The acceptance criteria are used to determine whether the qualified assembly can be accepted. For unqualified items, the specific reasons need to be analyzed in order to provide effective improvement measures. The reasons for unqualified items include processing errors, assembly problems, material defects, and environmental factors. Obtain improvement methods, formulate targeted improvement measures based on the reasons for non-conformity, generate a list of improvement measures and send it to the terminal, such as machine tool operators or maintenance engineers. The improvement list should include descriptions of non-conformities, cause analysis and improvement measures.

[0028] like Figure 2 As shown, as a preferred embodiment of the present invention, the visual result is obtained, and the visual result includes an assembly structure, a moving part, and an internal cavity, and the visual result is preprocessed, and the preprocessing includes the steps of noise reduction and feature extraction, specifically including: S201, obtaining visual results, the visual results including assembly structure, moving parts and internal cavity, recording the movement process through the visual results, the movement process including key actions during startup, processing and stopping.

[0029] In this step, visual results are obtained, including assembly structure, moving parts and internal cavity. The assembly structure is used to obtain the appearance data of the overall assembly of the machine tool, verify the assembly gap, joint position and surface defects; the moving parts include obtaining the dynamic state data of the moving parts such as the spindle, guide rail, slider, etc., and evaluating their running stability and accuracy; the internal cavity includes detecting whether there are processing defects, contamination or assembly abnormalities in the internal cavity of the machine tool, such as the bearing cavity and the transmission cavity; The motion process is recorded through visual results. The motion process is divided into three stages: startup, processing and stop. The visual results record the key actions and performance characteristics of each stage to ensure that the transition process of the machine tool from static to startup is smooth without abnormal vibration, noise or jitter. The dynamic stability and processing accuracy of the moving parts of the machine tool in the processing task are analyzed to ensure that the transition process of the machine tool from running to stopping is safe and reliable without abnormal vibration or thermal damage.

[0030] S202, preprocessing the visual result, wherein the preprocessing includes noise reduction and feature extraction, and obtaining a preprocessing result.

[0031] In this step, the visual results are preprocessed, including noise reduction and feature extraction. The goal of noise reduction is to reduce the random noise caused by the detection equipment or environment, improve the quality of image and point cloud data, and enhance the extraction accuracy of effective information. For example, Gaussian filtering can be used to eliminate the small bright spot noise in the image taken on the machine tool surface. The goal of feature extraction is to extract key information describing assembly status, dynamic performance or surface quality from the denoised data. The extracted content includes edge features, texture features and key point features. The contact boundary between the guide rail and the slider can be identified through edge features, scratches or cracks can be identified through texture features, and turning points in the motion trajectory of the workbench can be detected through key point features. Obtaining the preprocessing results, through noise reduction and feature extraction, clear image data can be generated, noise is removed, and key information is enhanced, such as assembly structure diagrams with clear edges, geometric features, texture features or dynamic characteristics extracted from feature data for further analysis.

[0032] S203, obtaining multiple preprocessing results, comparing the multiple preprocessing results, and obtaining the best preprocessing result.

[0033] In this step, multiple preprocessing results are obtained, and the machine tool can be inspected multiple times. Different preprocessing results are visualized to intuitively evaluate the noise removal effect and edge clarity. According to the comparison results, the preprocessing result that best meets the inspection target is selected as the best result. The best result should achieve the best balance between data accuracy, feature integrity, processing efficiency and applicability. Therefore, it is necessary to compare and analyze multiple preprocessing results and select the best preprocessing result to support subsequent inspection and judgment.

[0034] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of obtaining the test result according to the visual result, obtaining the test standard, the test standard including geometric accuracy test, dynamic performance test, surface quality test and assembly test, comparing the test result with the test standard, and judging whether the assembly result is qualified, specifically includes: S301, obtaining a test result according to the best preprocessing result, and acquiring a test standard, wherein the test standard includes geometric accuracy test, dynamic performance test, surface quality test and assembly test.

[0035] In this step, the test results are obtained according to the best preprocessing results. After the assembly test results are obtained by visual inspection technology, the test results are compared with the predefined test standards to determine whether the assembly quality is qualified. Geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection are performed by visual inspection equipment; Geometric accuracy testing is to test the geometric shape and dimensional tolerance of key machine tool components to ensure assembly accuracy; dynamic performance testing is to evaluate the motion accuracy and stability of the machine tool under dynamic operation; surface quality testing is to check the surface roughness, coating uniformity and defects of machine tool components; assembly testing is to ensure that the clearance, position accuracy and matching relationship of the assembled components meet the design requirements; In the geometric accuracy inspection standard, the straightness tolerance of the guide rail is ≤ 0.02 mm, the parallelism tolerance of the slider and the guide rail is ≤ 0.01 mm, and the dimensional tolerance is ±0.01 mm for example. The spindle eccentricity in the dynamic performance test standard is: spindle eccentricity ≤ 0.02 mm, motion track deviation: slider track deviation ≤ 0.03 mm; Surface quality inspection standards: surface roughness: surface roughness Ra ≤ 0.8 µm, coating thickness: coating thickness deviation ≤ ±5%, defect size: crack length ≤ 0.3 mm, scratch depth ≤ 0.1 mm; The assembly inspection standard stipulates that the gap between the guide rail and the slider is ≤ 0.02 mm, and the position deviation is ≤ 0.01 mm.

[0036] S302, comparing the test result with the test standard, determining whether the assembly result is qualified, and obtaining error information if the test is unqualified.

[0037] In this step, the test results are compared with the test standards, and the four aspects of geometric accuracy, dynamic performance, surface quality and assembly test are classified and compared with the corresponding test standards to determine whether they are within the allowable range. For items that exceed the test standard range, their specific error values ​​and deviations are recorded; Determine whether the assembly results are qualified. A qualified result means that all test results are within the test standard range, and a failed result means that any test result exceeds the test standard range. Error information is obtained based on the unqualified test. By comparing the test results with the standards and analyzing the unqualified items, the source of the problem can be accurately located, the impact can be assessed, and specific repair measures can be provided. This process helps to improve the overall quality and reliability of machine tool assembly.

[0038] S303: If it is determined that the test result is greater than the test standard, the test is determined to be qualified.

[0039] In this step, if it is determined that the test result is greater than the test standard, if the test result ≥ the test standard, it is determined to be qualified, for example, including rigidity testing, coating thickness and friction coefficient.

[0040] like Figure 4 As shown, as a preferred embodiment of the present invention, the steps of obtaining acceptance criteria, judging whether the qualified assembly can be accepted according to the acceptance criteria, obtaining the reasons for failure, obtaining improvement methods, and sending the improvement methods to the terminal specifically include: S401, obtaining acceptance criteria, and judging whether qualified assembly can be accepted according to the acceptance criteria, wherein the acceptance criteria is greater than the inspection criteria.

[0041] In this step, the acceptance criteria are obtained. The formulation of the acceptance criteria includes design drawings and technical documents, industry and national standards, and customer customization requirements. The acceptance criteria are used to determine whether the qualified assembly can be accepted. For unqualified items, the specific reasons need to be analyzed in order to provide effective improvement measures. The reasons for unqualified items include processing errors, assembly problems, material defects, and environmental factors. The acceptance standard is the benchmark for the final judgment of whether the assembly is qualified. The acceptance standard is greater than the inspection standard. The acceptance standard is set higher than the inspection standard because the actual tolerance range is larger: the assembly allows a certain error but can still operate normally. For example, the inspection standard: straightness tolerance ≤ 0.02 mm, the acceptance standard: straightness tolerance ≤ 0.03 mm.

[0042] S402, obtaining the reasons for non-conformity and improvement methods, which include reassembly, improvement of matching design, surface repair and improvement of installation process.

[0043] In this step, the reasons for failure are obtained. Geometric accuracy problems include accumulated errors in part processing, inaccurate positioning during assembly, and assembly deviation caused by stress deformation of parts. Dynamic performance problems include excessive or uneven clearance, lubrication system failure, and unbalanced moving parts. Surface quality problems include scratches, cracks, pits and other defects on the surface of parts, uneven coating thickness or improper spraying, and surface roughness that does not meet requirements. Assembly problems include assembly clearance deviation, position error of mating parts, and improper tightening or calibration during assembly. Obtain improvement methods. For different types of non-conformities, improvement methods include reassembly, improving matching design, repairing the surface, and improving installation technology. Through the above process, the causes of non-conformities can be accurately analyzed and targeted improvement measures can be formulated, ultimately improving assembly quality and ensuring that acceptance standards are met.

[0044] S403, determining a specific improvement method based on the non-conformity reason, and sending the specific improvement method to the terminal.

[0045] In this step, the specific improvement method is determined based on the cause of non-conformity. For non-conforming items found in machine tool assembly inspection, the specific improvement method needs to be determined based on the analysis of the cause of non-conformity, and detailed operation steps are generated and sent to the terminal, such as the equipment of quality inspectors, assembly workers or engineers.

[0046] like Figure 5 As shown, a machine tool assembly intelligent acceptance system based on computer vision is provided in an embodiment of the present invention, and the system includes: The standard module 100 is used to obtain the inspection standard and obtain the visual inspection data, wherein the visual inspection data includes the visual inspection equipment and equipment parameters.

[0047] In this system, the standard module 100 obtains the inspection standard, which is the target basis of visual inspection, clarifies the allowable range and inspection requirements of various parameters, obtains the CAD drawings and technical documents of machine tool assembly, clarifies the size, geometric accuracy, tolerance and matching requirements, and also needs to refer to the special requirements proposed to formulate personalized inspection standards; Obtain visual inspection data. Visual inspection data is the basis for realizing automated inspection, including the entire process from equipment selection to parameter configuration. Visual inspection data includes visual inspection equipment and equipment parameters. Visual inspection equipment includes industrial cameras and infrared cameras. Industrial cameras are used to collect images of machine tool appearance, surface defects or key dimensions. Infrared cameras are used to detect specific problems such as coating thickness, cracks, thermal effects, etc. The configuration of equipment parameters directly determines the accuracy and efficiency of detection. Camera parameters include resolution, exposure time and frame rate. Machine tool assembly detection based on computer vision can not only accurately capture complex assembly problems, but also improve efficiency and detection quality through standardized and automated processes.

[0048] The preprocessing module 200 is used to obtain visual results, which include assembly structures, moving parts and internal cavities, and preprocess the visual results, which includes noise reduction and feature extraction.

[0049] In this system, the preprocessing module 200 obtains visual results, which include assembly structure, moving parts and internal cavity. The assembly structure is used to obtain the appearance data of the overall assembly of the machine tool, verify the assembly gap, joint position and surface defects; the moving parts include obtaining the dynamic state data of moving parts such as the spindle, guide rails, and sliders, and evaluating their operating stability and accuracy; the internal cavity includes detecting whether there are processing defects, contamination or assembly abnormalities in the internal cavity of the machine tool, such as the bearing cavity and the transmission cavity; Preprocess the visual results, including noise reduction and feature extraction. Visual data may be affected by external ambient light, equipment noise or acquisition accuracy. Noise reduction can improve the quality of images and point clouds. Feature extraction is used to extract the core information of assembly structures, moving parts and cavities from the denoised visual data for subsequent analysis.

[0050] The detection module 300 is used to obtain the detection results based on the visual results, obtain the detection standards, which include geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection, compare the detection results with the detection standards, and determine whether the assembly results are qualified.

[0051] In this system, the inspection module 300 obtains the inspection result based on the visual result. After obtaining the assembly inspection result through visual inspection technology, the inspection result is compared with the predefined inspection standard to determine whether the assembly quality is qualified. The geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection are performed through visual inspection equipment; Obtain the inspection standards. The inspection standards are the basis for judging whether the assembly is qualified, including design documents, industry standards and user requirements. Compare the inspection results with the inspection standards, compare the inspection results with the inspection standards item by item, record whether the qualified conditions are met, and judge whether the assembly results are qualified. Qualified means that all inspection results meet the inspection standards, and unqualified means that any inspection result exceeds the inspection standard range.

[0052] The acceptance module 400 is used to obtain the acceptance criteria, determine whether the qualified assembly can be accepted according to the acceptance criteria, obtain the reasons for failure, obtain improvement methods, and send the improvement methods to the terminal.

[0053] In this system, the acceptance module 400 obtains the acceptance criteria. The formulation of the acceptance criteria includes design drawings and technical documents, industry and national standards, and customer customization requirements. The acceptance criteria are used to determine whether the qualified assembly can be accepted. For unqualified items, the specific reasons need to be analyzed in order to provide effective improvement measures. The reasons for unqualified items include processing errors, assembly problems, material defects, and environmental factors. Obtain improvement methods, formulate targeted improvement measures based on the reasons for non-conformity, generate a list of improvement measures and send it to the terminal, such as machine tool operators or maintenance engineers. The improvement list should include descriptions of non-conformities, cause analysis and improvement measures.

[0054] like Figure 6 As shown, as a preferred embodiment of the present invention, the preprocessing module 200 includes: The visual result unit 201 is used to obtain visual results, which include assembly structures, moving parts and internal cavities. The movement process is recorded through the visual results, and the movement process includes key actions during startup, processing and stopping.

[0055] In this module, the visual result unit 201 obtains visual results, which include assembly structure, moving parts and internal cavity. The assembly structure is used to obtain the appearance data of the overall assembly of the machine tool, verify the assembly gap, joint position and surface defects; the moving parts include obtaining the dynamic state data of moving parts such as the spindle, guide rails, and sliders, and evaluating their operating stability and accuracy; the internal cavity includes detecting whether there are processing defects, contamination or assembly abnormalities in the internal cavity of the machine tool, such as the bearing cavity and the transmission cavity; The motion process is recorded through visual results. The motion process is divided into three stages: startup, processing and stop. The visual results record the key actions and performance characteristics of each stage to ensure that the transition process of the machine tool from static to startup is smooth without abnormal vibration, noise or jitter. The dynamic stability and processing accuracy of the moving parts of the machine tool in the processing task are analyzed to ensure that the transition process of the machine tool from running to stopping is safe and reliable without abnormal vibration or thermal damage.

[0056] The preprocessing unit 202 is used to preprocess the visual result, wherein the preprocessing includes noise reduction processing and feature extraction to obtain a preprocessing result.

[0057] In this module, the preprocessing unit 202 preprocesses the visual results. The preprocessing includes noise reduction and feature extraction. The goal of noise reduction is to reduce the random noise caused by the detection equipment or environment, improve the quality of image and point cloud data, and enhance the extraction accuracy of effective information. For example, Gaussian filtering can be used to eliminate the small bright spot noise in the image taken on the machine tool surface. The goal of feature extraction is to extract key information describing assembly status, dynamic performance or surface quality from the denoised data. The extracted content includes edge features, texture features and key point features. The contact boundary between the guide rail and the slider can be identified through edge features, scratches or cracks can be identified through texture features, and turning points in the motion trajectory of the workbench can be detected through key point features. Obtaining the preprocessing results, through noise reduction and feature extraction, clear image data can be generated, noise is removed, and key information is enhanced, such as assembly structure diagrams with clear edges, geometric features, texture features or dynamic characteristics extracted from feature data for further analysis.

[0058] The optimal unit 203 is used to obtain multiple preprocessing results, compare the multiple preprocessing results, and obtain the optimal preprocessing result.

[0059] In this module, the optimal unit 203 obtains multiple preprocessing results, can perform multiple inspections on the machine tool, visualize different preprocessing results, intuitively evaluate the noise removal effect and edge clarity, and select the preprocessing result that best meets the inspection target as the optimal result based on the comparison results. The optimal result should achieve the best balance between data accuracy, feature integrity, processing efficiency and applicability. Therefore, it is necessary to compare and analyze multiple preprocessing results and select the best preprocessing result from them to support subsequent inspection and judgment.

[0060] like Figure 7 As shown, as a preferred embodiment of the present invention, the detection module 300 includes: The detection unit 301 is used to obtain the detection result according to the best preprocessing result and acquire the detection standard, wherein the detection standard includes geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection.

[0061] In this module, the detection unit 301 obtains the detection result according to the best preprocessing result. After obtaining the assembly detection result through visual inspection technology, the detection result is compared with the predefined detection standard to determine whether the assembly quality is qualified. The geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection are performed through visual inspection equipment; Geometric accuracy testing is to test the geometric shape and dimensional tolerance of key machine tool components to ensure assembly accuracy; dynamic performance testing is to evaluate the motion accuracy and stability of the machine tool under dynamic operation; surface quality testing is to check the surface roughness, coating uniformity and defects of machine tool components; assembly testing is to ensure that the clearance, position accuracy and matching relationship of the assembled components meet the design requirements; In the geometric accuracy inspection standard, the straightness tolerance of the guide rail is ≤ 0.02 mm, the parallelism tolerance of the slider and the guide rail is ≤ 0.01 mm, and the dimensional tolerance is ±0.01 mm for example. The spindle eccentricity in the dynamic performance test standard is: spindle eccentricity ≤ 0.02 mm, motion track deviation: slider track deviation ≤ 0.03 mm; Surface quality inspection standards: surface roughness: surface roughness Ra ≤ 0.8 µm, coating thickness: coating thickness deviation ≤ ±5%, defect size: crack length ≤ 0.3 mm, scratch depth ≤ 0.1 mm; The assembly inspection standard stipulates that the gap between the guide rail and the slider is ≤ 0.02 mm, and the position deviation is ≤ 0.01 mm.

[0062] The judgment unit 302 is used to compare the test result with the test standard, judge whether the assembly result is qualified, and obtain error information according to the unqualified test result.

[0063] In this module, the judgment unit 302 compares the test results with the test standards, and compares the four aspects of geometric accuracy, dynamic performance, surface quality and assembly detection with the corresponding test standards to determine whether they are within the allowable range. For items that exceed the test standard range, the specific error value and deviation are recorded; Determine whether the assembly results are qualified. A qualified result means that all test results are within the test standard range, and a failed result means that any test result exceeds the test standard range. Error information is obtained based on the unqualified test. By comparing the test results with the standards and analyzing the unqualified items, the source of the problem can be accurately located, the impact can be assessed, and specific repair measures can be provided. This process helps to improve the overall quality and reliability of machine tool assembly.

[0064] The qualified detection unit 303 is used to determine that the detection is qualified if it is determined that the detection result is greater than the detection standard.

[0065] In this module, if the qualified detection unit 303 determines that the detection result is greater than the detection standard, if the detection result ≥ the detection standard, it is determined to be qualified, for example, including rigidity detection, coating thickness and friction coefficient.

[0066] like Figure 8 As shown, as a preferred embodiment of the present invention, the acceptance module 400 includes: The acceptance standard unit 401 is used to obtain the acceptance standard and determine whether the qualified assembly can be accepted according to the acceptance standard. The acceptance standard is greater than the inspection standard.

[0067] In this module, the acceptance standard unit 401 obtains the acceptance standard. The formulation of the acceptance standard includes design drawings and technical documents, industry and national standards and customer customization requirements. According to the acceptance standard, it is judged whether the qualified assembly can be accepted. For unqualified items, the specific reasons need to be analyzed in order to provide effective improvement measures. The reasons for unqualified items include processing errors, assembly problems, material defects and environmental factors; The acceptance standard is the benchmark for the final judgment of whether the assembly is qualified. The acceptance standard is greater than the inspection standard. The acceptance standard is set higher than the inspection standard because the actual tolerance range is larger: the assembly allows a certain error but can still operate normally. For example, the inspection standard: straightness tolerance ≤ 0.02 mm, the acceptance standard: straightness tolerance ≤ 0.03 mm.

[0068] The improvement unit 402 is used to obtain the reasons for non-conformity and obtain improvement methods, which include reassembly, improvement of matching design, surface repair and improvement of installation process.

[0069] In this module, the improvement unit 402 obtains the reasons for non-conformity. Geometric accuracy problems include the accumulation of part processing errors, inaccurate positioning during assembly, and assembly deviation caused by stress deformation of parts; dynamic performance problems include excessive or uneven matching clearance, lubrication system failure, and unbalanced moving parts; surface quality problems include scratches, cracks, pits and other defects on the surface of parts, uneven coating thickness or improper spraying, and surface roughness that does not meet the requirements; assembly problems include assembly clearance deviation, matching part position error, and improper tightening or calibration during assembly; Obtain improvement methods. For different types of non-conformities, improvement methods include reassembly, improving matching design, repairing the surface, and improving installation technology. Through the above process, the causes of non-conformities can be accurately analyzed and targeted improvement measures can be formulated, ultimately improving assembly quality and ensuring that acceptance standards are met.

[0070] The terminal unit 403 is used to determine a specific improvement method according to the non-conformity reason and send the specific improvement method to the terminal.

[0071] In this module, the terminal unit 403 determines the specific improvement method based on the cause of non-conformity. For the non-conforming items found in the machine tool assembly inspection, the specific improvement method needs to be determined based on the analysis of the cause of non-conformity, and detailed operation steps are generated and sent to the terminal, such as the equipment of the quality inspector, assembly worker or engineer.

[0072] In one embodiment, a computer device is provided, the computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are implemented: Acquire inspection standards and obtain visual inspection data, wherein the visual inspection data includes visual inspection equipment and equipment parameters; Acquire visual results, the visual results include assembly structures, moving parts and internal cavities, and preprocess the visual results, the preprocessing includes noise reduction and feature extraction; Obtain the test results based on the visual results, obtain the test standards, which include geometric accuracy test, dynamic performance test, surface quality test and assembly test, compare the test results with the test standards, and determine whether the assembly results are qualified; Obtain the acceptance criteria, determine whether the qualified assembly can be accepted according to the acceptance criteria, obtain the reasons for failure, obtain the improvement methods, and send the improvement methods to the terminal.

[0073] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor performs the following steps: Acquire inspection standards and obtain visual inspection data, wherein the visual inspection data includes visual inspection equipment and equipment parameters; Acquire visual results, the visual results include assembly structures, moving parts and internal cavities, and preprocess the visual results, the preprocessing includes noise reduction and feature extraction; Obtain the test results based on the visual results, obtain the test standards, which include geometric accuracy test, dynamic performance test, surface quality test and assembly test, compare the test results with the test standards, and determine whether the assembly results are qualified; Obtain the acceptance criteria, determine whether the qualified assembly can be accepted according to the acceptance criteria, obtain the reasons for failure, obtain the improvement methods, and send the improvement methods to the terminal.

[0074] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0075] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0076] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0077] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent acceptance of machine tool assembly based on computer vision, characterized in that: The method comprises: Acquire inspection standards and obtain visual inspection data, wherein the visual inspection data includes visual inspection equipment and equipment parameters; Acquire visual results, the visual results include assembly structures, moving parts and internal cavities, and preprocess the visual results, the preprocessing includes noise reduction and feature extraction; Obtain the inspection results based on the visual results, obtain the inspection standards, which include geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection, compare the inspection results with the inspection standards, and determine whether the assembly results are qualified; Obtain the acceptance criteria, determine whether the qualified assembly can be accepted according to the acceptance criteria, obtain the reasons for failure, obtain the improvement methods, and send the improvement methods to the terminal.

2. According to the computer vision-based intelligent acceptance method for machine tool assembly according to claim 1, it is characterized in that: The obtaining of visual results, which include assembly structures, moving parts and internal cavities, and preprocessing of the visual results, which includes noise reduction and feature extraction steps, specifically includes: Obtaining visual results, including assembly structures, moving parts and internal cavities, and recording the motion process through the visual results, wherein the motion process includes key actions during startup, processing and stopping; Preprocessing the visual results, wherein the preprocessing includes noise reduction and feature extraction, to obtain preprocessing results; Obtain multiple preprocessing results, compare the multiple preprocessing results, and obtain the best preprocessing result.

3. The method for intelligent acceptance of machine tool assembly based on computer vision according to claim 1, characterized in that: The step of obtaining the test results according to the visual results, obtaining the test standards, wherein the test standards include geometric accuracy test, dynamic performance test, surface quality test and assembly test, comparing the test results with the test standards, and judging whether the assembly results are qualified includes: Obtaining the test results according to the best preprocessing results and obtaining the test standards, wherein the test standards include geometric accuracy test, dynamic performance test, surface quality test and assembly test; Compare the test results with the test standards to determine whether the assembly results are qualified, and obtain error information if the test fails; If the test result is determined to be greater than the test standard, the test is considered qualified.

4. The method for intelligent acceptance of machine tool assembly based on computer vision according to claim 1, characterized in that: The steps of obtaining the acceptance criteria, judging whether the qualified assembly can be accepted according to the acceptance criteria, obtaining the reasons for failure, obtaining the improvement method, and sending the improvement method to the terminal specifically include: Obtain the acceptance criteria and determine whether the qualified assembly can be accepted based on the acceptance criteria. The acceptance criteria are greater than the inspection criteria. Obtain the reasons for non-conformity and the improvement methods, which include reassembly, improvement of matching design, surface repair and improvement of installation process; Determine the specific improvement method based on the cause of non-conformity and send the specific improvement method to the terminal.

5. The method for intelligent acceptance of machine tool assembly based on computer vision according to claim 1, characterized in that: The inspection standards are inspection standards for machine tool assembly.

6. A machine tool assembly intelligent acceptance system based on computer vision, characterized in that: The system comprises: A standard module, which obtains the inspection standard and the visual inspection data, wherein the visual inspection data includes the visual inspection equipment and equipment parameters; A preprocessing module, which obtains visual results, including assembly structures, moving parts and internal cavities, and preprocesses the visual results, including noise reduction and feature extraction; The inspection module obtains the inspection results according to the visual results, obtains the inspection standards, which include geometric accuracy inspection, dynamic performance inspection, surface quality inspection and assembly inspection, compares the inspection results with the inspection standards, and determines whether the assembly results are qualified; The acceptance module obtains the acceptance criteria, determines whether the qualified assembly can be accepted according to the acceptance criteria, obtains the reasons for failure, obtains the improvement methods, and sends the improvement methods to the terminal.

7. The intelligent acceptance system for machine tool assembly based on computer vision according to claim 6, characterized in that: The preprocessing module comprises: A visual result unit, which obtains visual results, including assembly structure, moving parts and internal cavity, and records the movement process through the visual results, and the movement process includes key actions during startup, processing and stopping; A preprocessing unit, which preprocesses the visual result, wherein the preprocessing includes noise reduction processing and feature extraction, and obtains a preprocessing result; The best unit obtains multiple preprocessing results, compares the multiple preprocessing results, and obtains the best preprocessing result.

8. The intelligent acceptance system for machine tool assembly based on computer vision according to claim 7, characterized in that: The detection module comprises: A detection unit, which obtains a detection result according to the best preprocessing result and acquires a detection standard, wherein the detection standard includes geometric accuracy detection, dynamic performance detection, surface quality detection and assembly detection; The judgment unit compares the test results with the test standards, judges whether the assembly results are qualified, and obtains error information if the test fails; For a qualified unit, if the test result is determined to be greater than the test standard, it is determined to be qualified.

9. The intelligent acceptance system for machine tool assembly based on computer vision according to claim 8, characterized in that: The acceptance module includes: Acceptance standard unit, obtains acceptance standards, and determines whether qualified assembly can be accepted based on the acceptance standards. The acceptance standards are greater than the inspection standards. Improve the unit, obtain the reasons for non-conformity, and obtain improvement methods, which include reassembly, improvement of matching design, surface repair, and improvement of installation process; The terminal unit determines the specific improvement method according to the non-conformity reason and sends the specific improvement method to the terminal.

10. The intelligent acceptance system for machine tool assembly based on computer vision according to claim 9, characterized in that: The inspection standards are inspection standards for machine tool assembly.

Images