Bearing Surface Quality Detection and Evaluation System Based on Visual Cognition Model

Abstract

The present invention belongs to the technical field of bearing detection, and specifically relates to a bearing surface quality detection and evaluation system based on a visual cognitive model, including a bearing scanning and image acquisition unit, a surface processing and extraction unit, a visual cognitive model unit, a quantitative evaluation and output unit, a bearing sorting unit, and a display and alarm terminal; through the bearing scanning and image acquisition unit and the surface processing and extraction unit, the present invention performs image acquisition and processing on the bearing surface, combines the visual cognitive model and machine learning algorithms for defect identification and quantitative evaluation, realizes the rapid detection and evaluation of the bearing surface quality, significantly improves the detection efficiency and reduces the labor cost, significantly reduces the workload of bearing production management personnel, and through the progressive and reasonable analysis and accurate early warning of the stability and accuracy of the bearing surface quality detection and evaluation process, as well as the bearing production performance and management status, effectively reduces the difficulty of bearing production detection and management.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing detection, specifically a bearing surface quality detection and evaluation system based on a visual cognitive model. Background Technique

[0002] As a core component in mechanical equipment, the surface quality of bearings has an important impact on the operating performance and lifespan of the equipment. Traditional bearing surface quality detection mainly relies on manual inspection. This method is not only inefficient but also vulnerable to human factors, resulting in low accuracy of detection results. With the development of computer vision and artificial intelligence technologies, automated detection systems based on machine vision have gradually become the mainstream;

[0003] In a Chinese invention patent with the publication number CN112308832A, a bearing quality detection method based on machine vision is disclosed. First, it is detected based on a two-dimensional detection method. After detecting whether there are defects under specific pixels, three-dimensional detection of local detail features is then used. For bearings with two-dimensional defect features, three-dimensional detection and reconstruction are carried out to obtain the bearing surface defect category. It is not necessary to perform three-dimensional detection on all bearings, which improves the detection efficiency and saves the detection cost;

[0004] However, in the actual application process of the above invention technical solution, it is impossible to achieve automated and comprehensive control of the bearing surface quality detection and evaluation process, and it is difficult to conduct progressive and reasonable analysis and accurate early warning on the stability and accuracy of the bearing surface quality detection and evaluation process, as well as the bearing production performance and management status. The intelligent level is low, and it cannot effectively reduce the difficulty of bearing production management;

[0005] In view of the above technical deficiencies, a solution is now proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a bearing surface quality detection and evaluation system based on a visual cognitive model, which solves the problems that the prior art cannot achieve automated and comprehensive control of the bearing surface quality detection and evaluation process, and it is difficult to conduct progressive and reasonable analysis and accurate early warning on the stability and accuracy of the bearing surface quality detection and evaluation process, as well as the bearing production performance and management status. The intelligent level is low and it cannot effectively reduce the difficulty of bearing production management.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] Bearing surface quality detection and evaluation system based on visual cognitive model, including bearing scanning and imaging unit, surface processing and extraction unit, visual cognitive model unit, quantitative evaluation and output unit, bearing sorting unit and display and alarm terminal; The bearing scanning and acquisition unit is configured with a high-resolution industrial camera, multi-angle annular light source and rotating stage, acquires visible light images, high-dynamic range images and three-dimensional topography data of the bearing surface, eliminates reflection interference through an adaptive light compensation algorithm, generates a clear surface image with multi-scale fusion, and sends the bearing surface image to the surface processing and extraction unit;

[0009] The surface processing and extraction unit performs preprocessing operations of denoising and enhancing contrast on the received bearing surface image, extracts the feature information of the bearing surface using image processing algorithms, and sends the bearing surface feature extraction information to the visual cognitive model unit; The visual cognitive model unit constructs a visual cognitive model using a convolutional neural network, the visual cognitive model analyzes the bearing surface feature extraction information, identifies the defects on the bearing surface, and sends the identification result to the quantitative evaluation and output unit;

[0010] The quantitative evaluation and output unit quantitatively evaluates the defects according to the identification result of the visual cognitive model, outputs the type, position and size of the defects, generates a bearing surface quality detection report, and sends the bearing surface quality detection report to the display and alarm terminal and the bearing sorting unit; The bearing sorting unit sorts the bearings that do not meet the requirements to the scrap area and the bearings that meet the requirements to the qualified area.

[0011] Furthermore, the display and alarm terminal is communicatively connected to a stability analysis and output module, and the stability analysis and output module is used to set the detection period, analyze the stability performance of the bearing surface quality detection and evaluation during the detection period, generate a stability qualified signal or a stability abnormal signal through the analysis, and send the stability qualified signal or the stability abnormal signal to the display and alarm terminal, and the display and alarm terminal issues a warning when receiving the stability abnormal signal.

[0012] Furthermore, the specific analysis process of the stability analysis and output module includes:

[0013] When performing the surface quality detection and evaluation of the corresponding bearing, mark the moment when the bearing scanning and acquisition unit performs image acquisition as moment one, mark the moment when the bearing sorting unit completes the bearing sorting as moment two, and mark the interval duration between moment one and moment two as the selected duration;

[0014] When the selected duration exceeds the preset selected duration threshold, the evaluation symbol ZP-1 is assigned. The number of times the evaluation symbol ZP-1 is assigned during the detection period is obtained, and the ratio is calculated with the total number of bearings subjected to surface quality detection and evaluation during the detection period to obtain the evaluation warning value. Also, the average value of all selected durations during the detection period is calculated to obtain the selection time decision value. The evaluation warning value and the selection time decision value are respectively compared numerically with the preset evaluation warning threshold and the preset selection time decision threshold. If the evaluation warning value or the selection time decision value exceeds the corresponding preset threshold, a stability anomaly signal is generated.

[0015] Furthermore, if neither the evaluation warning value nor the selection time decision value exceeds the corresponding preset threshold, the total duration of the surface quality detection and evaluation of the bearings during the detection period is collected and marked as the total detection time value, and the number of times of failures during the surface quality detection and evaluation of the bearings during the detection period is marked as the detection obstacle value. The ratio of the detection obstacle value to the total detection time value is calculated to obtain the obstacle judgment value.

[0016] By performing weighted summation calculation on the evaluation warning value, the selection time decision value, and the obstacle judgment value, the stability output difference value is obtained. The stability output difference value is compared numerically with the preset stability output difference threshold. If the stability output difference value exceeds the preset stability output difference threshold, a stability anomaly signal is generated; if the stability output difference value does not exceed the preset stability output difference threshold, a stability qualified signal is generated.

[0017] Furthermore, the stability analysis output module is communicatively connected to the accuracy test module. The stability analysis output module sends the stability qualified signal to the accuracy test module. When the accuracy test module receives the stability qualified signal, it analyzes the accuracy status of the surface quality detection and evaluation of the bearings, generates an accuracy qualified signal or an accuracy anomaly signal through the analysis, and sends the accuracy qualified signal or the accuracy anomaly signal to the display and alarm terminal. When the display and alarm terminal receives the accuracy anomaly signal, it issues a warning.

[0018] Furthermore, the specific analysis process of the accuracy test module includes:

[0019] Perform several surface quality detection and evaluations on bearings known to have multiple defects, collect the recognition accuracy rate during each surface quality detection and evaluation, calculate the variance of all recognition accuracy rates to obtain the result consistency analysis value, compare the result consistency analysis value numerically with the preset result consistency analysis threshold. If the result consistency analysis value exceeds the preset result consistency analysis threshold, the recognition judgment symbol XP-1 is assigned to the test results of the corresponding bearings.

[0020] If the result consistency analysis value does not exceed the preset result consistency analysis threshold, the ratio of the number of detection times with the recognition accuracy not exceeding the preset recognition accuracy threshold is marked as the recognition deviation value, and the average value of all recognition accuracies is calculated to obtain the recognition judgment value. The recognition deviation value and the recognition judgment value are respectively compared numerically with the preset recognition deviation threshold and the preset recognition judgment threshold. If the recognition deviation value or the recognition judgment value exceeds the corresponding preset threshold, the recognition judgment symbol XP-1 is assigned to the test result of the corresponding bearing.

[0021] After completing the tests on several bearings with multiple defects, if the number of times the recognition judgment symbol XP-1 is assigned is zero, a precision qualified signal is generated; if the number of times the recognition judgment symbol XP-1 is assigned is not zero, a precision anomaly signal is generated.

[0022] Furthermore, the precision test module is communicatively connected to the bearing production line optimization judgment module. The precision test module sends the precision qualified signal to the bearing production line optimization judgment module. When the bearing production line optimization judgment module receives the precision qualified signal, it analyzes the production status of the bearing production line during the detection period, generates a bearing production qualified signal or a bearing production anomaly signal through the analysis, and sends the bearing production qualified signal or the bearing production anomaly signal to the display and alarm terminal. When the display and alarm terminal receives the bearing production anomaly signal, it issues a warning.

[0023] Furthermore, the specific analysis process of the bearing production line optimization judgment module includes:

[0024] Collect the bearing surface quality inspection reports of all bearings entering the scrap area during the detection period. Based on the bearing surface quality inspection reports, obtain all the defect information existing on the corresponding bearings. Calculate the ratio of the number of bearings involved in the corresponding type of defect to the total number of bearings tested during the detection period to obtain the bearing defect matching value. Compare the bearing defect matching value numerically with the corresponding preset bearing defect matching threshold. If the bearing defect matching value exceeds the preset bearing defect matching threshold, assign the easy-occurrence judgment symbol WP-1 to the corresponding type of defect.

[0025] If there is a defect type corresponding to the easy-occurrence judgment symbol WP-1 during the detection period, a bearing production anomaly signal is generated; if there is no defect type corresponding to the easy-occurrence judgment symbol WP-1 during the detection period, calculate the ratio of the number of bearings entering the scrap area during the detection period to the total number of bearings tested during the detection period to obtain the bearing scrap analysis value. Compare the bearing scrap analysis value numerically with the preset bearing scrap analysis threshold. If the bearing scrap analysis value exceeds the preset bearing scrap analysis threshold, a bearing production anomaly signal is generated; if the bearing scrap analysis value does not exceed the preset bearing scrap analysis threshold, a bearing production qualified signal is generated.

[0026] Further, the bearing production line optimization judgment module is communicatively connected to the bearing production management judgment module. The bearing production management module sends a bearing production qualified signal to the bearing production management judgment module. When the bearing production management judgment module receives the bearing production qualified signal, it analyzes the management performance of bearing production during the detection period, generates a management qualified signal or a management abnormal signal through the analysis, and sends the management qualified signal or the management abnormal signal to the display and alarm terminal. When the display and alarm terminal receives the management abnormal signal, it issues a corresponding early warning.

[0027] Further, the specific analysis process of the bearing production management judgment module is as follows:

[0028] Monitor the bearing production line, collect the average interval duration for the maintenance of the corresponding production equipment in the bearing production line during the detection period, and compare its value with the corresponding preset average interval duration threshold. Mark the production equipment with an average interval duration exceeding the preset average interval duration threshold as a poorly managed equipment, obtain the ratio of the number of equipment marked as poorly managed equipment during the detection period and mark it as a non-good management value;

[0029] And mark the ratio of the average interval duration for the maintenance of the corresponding production equipment in the bearing production line to the corresponding preset average interval duration threshold as a maintenance interval measurement value. Calculate the mean value of the maintenance interval measurement values of all production equipment to obtain a production line maintenance value, and collect the operation error frequency of the production line operators during the detection period. Calculate the management judgment coefficient by weighted summation of the non-good management value, the production line maintenance value, and the operation error frequency, and compare the management judgment coefficient with the preset management judgment coefficient threshold. If the management judgment coefficient exceeds the preset management judgment coefficient threshold, generate a management abnormal signal; if the management judgment coefficient does not exceed the preset management judgment coefficient threshold, generate a management qualified signal.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] 1. In the present invention, by collecting and processing the images of the bearing surface and combining the visual cognition model and machine learning algorithm for defect identification and quantitative evaluation, the rapid detection and evaluation of the bearing surface quality are realized, the detection efficiency is significantly improved, the labor cost is reduced, the workload of the bearing production management personnel is significantly reduced, and through the stability analysis output module, the stability performance of the bearing surface quality detection and evaluation during the detection period is analyzed. When generating a stability qualified signal, the accuracy of the bearing surface quality detection and evaluation system is tested and analyzed, which is beneficial to realizing the accurate evaluation of the bearing surface quality;

[0032] 2. In the present invention, when the bearing production line optimization judgment module receives a signal indicating qualified accuracy, it analyzes the production status of the bearing production line during the detection period. When generating a bearing production abnormal signal, it strengthens the subsequent supervision of the bearing production line to ensure the quality of the produced bearings and reduce the production cost of the bearings. And when generating a bearing production qualified signal, it analyzes the management performance of bearing production during the detection period through the bearing production management judgment module. When generating a management abnormal signal, it strengthens the control of the bearing production process to further ensure the smooth progress of the subsequent bearing production process and improve the bearing quality, with a high level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the accompanying drawings;

[0034] Figure 1 It is the system block diagram of the first embodiment in the present invention;

[0035] Figure 2 It is the system block diagram of the second and third embodiments in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0037] Embodiment 1: As Figure 1 shown, the bearing surface quality detection and evaluation system based on the visual cognitive model proposed by the present invention includes a bearing scanning and imaging unit, a surface processing and extraction unit, a visual cognitive model unit, a quantitative evaluation and output unit, a bearing sorting unit, a stability analysis and output module, a precision test module, and a display and alarm terminal;

[0038] Among them, the bearing scanning and acquisition unit is configured with a high-resolution industrial camera, a multi-angle annular light source, and a rotating stage to collect visible light images, high-dynamic range images, and three-dimensional topography data of the bearing surface. It eliminates the reflection interference through an adaptive light compensation algorithm to generate a clear surface image with multi-scale fusion, and sends the bearing surface image to the surface processing and extraction unit;

[0039] The surface processing and extraction unit performs preprocessing operations of denoising and enhancing contrast on the received bearing surface image, and uses image processing algorithms to extract the feature information of the bearing surface, such as edges, textures, etc., and sends the bearing surface feature extraction information to the visual cognitive model unit;

[0040] The visual recognition model unit constructs a visual recognition model using a convolutional neural network (CNN) or other deep learning architectures (the model is trained with a large amount of bearing surface image data so that it can accurately identify various types of surface defects). The visual recognition model analyzes the information extracted from the bearing surface features, identifies the defects on the bearing surface, and sends the recognition results to the quantization evaluation output unit;

[0041] The quantization evaluation output unit quantitatively evaluates the defects according to the recognition results of the visual recognition model, outputs detailed information such as the type, location, and size of the defects, generates a bearing surface quality inspection report, and sends the bearing surface quality inspection report to the display alarm end and the bearing sorting unit; the bearing sorting unit sorts the bearings that do not meet the requirements to the scrap area and sorts the qualified bearings to the qualified area, realizing the automatic sorting of bearings at the end of the bearing surface quality inspection and evaluation.

[0042] The present invention adopts a high-resolution industrial camera and advanced image processing algorithms, can achieve high-precision detection of tiny defects on the bearing surface, combines a visual recognition model and machine learning algorithms to intelligently evaluate the detected defects, outputs detailed defect information, provides strong support for subsequent bearing quality control and improvement, and realizes rapid detection of bearings through an automated system, greatly improving the detection efficiency, reducing the labor cost, and significantly reducing the workload of bearing production management personnel.

[0043] The stability analysis output module sets the detection period, analyzes the stability performance of the bearing surface quality inspection and evaluation during the detection period, generates a stability qualified signal or a stability abnormal signal through the analysis, and sends the stability qualified signal or the stability abnormal signal to the display alarm end. When the display alarm end receives the stability abnormal signal, it issues a warning to remind the management personnel to strengthen the subsequent detection supervision and optimize the detection evaluation system in time to ensure the stability of the subsequent detection evaluation process; the specific analysis process of the stability analysis output module is as follows:

[0044] When performing the surface quality inspection and evaluation of the corresponding bearing, mark the moment when the bearing scanning and acquisition unit performs image acquisition as moment one, mark the moment when the bearing sorting unit completes the bearing sorting as moment two, and mark the interval duration between moment one and moment two as the selected duration; among them, the larger the value of the selected duration, the slower the operation efficiency of the surface quality inspection and evaluation of the corresponding bearing.

[0045] Compare the selected duration with a preset selected duration numerically. When the selected duration exceeds the preset selected duration threshold, assign the evaluation symbol ZP-1. Obtain the number of times the evaluation symbol ZP-1 is assigned during the detection period and calculate the ratio with the total number of bearings subjected to surface quality inspection and evaluation during the detection period to obtain the evaluation warning value. Also, calculate the average value of all selected durations during the detection period to obtain the selection time decision value.

[0046] Compare the evaluation warning value and the selection time decision value with the preset evaluation warning threshold and the preset selection time decision threshold respectively. If the evaluation warning value or the selection time decision value exceeds the corresponding preset threshold, it indicates that the operation stability of the bearing surface quality inspection and evaluation operation is poor, and a stability anomaly signal is generated.

[0047] Furthermore, if neither the evaluation warning value nor the selection time decision value exceeds the corresponding preset threshold, collect the total duration of the bearing surface quality inspection and evaluation during the detection period and mark it as the total detection time value. Also, mark the number of times of failure during the bearing surface quality inspection and evaluation during the detection period as the detection obstacle value. Calculate the ratio of the detection obstacle value to the total detection time value to obtain the obstacle judgment value.

[0048] Calculate the stability deviation value by weighted summation of the evaluation warning value, the selection time decision value, and the obstacle judgment value. That is, by respectively assigning preset weight coefficients greater than zero to the evaluation warning value, the selection time decision value, and the obstacle judgment value, multiplying the evaluation warning value, the selection time decision value, and the obstacle judgment value by the corresponding preset weight coefficients respectively, and summing the three product results to obtain the stability deviation value. Moreover, the larger the value of the stability deviation value, the worse the overall operation stability of the bearing surface quality inspection and evaluation operation.

[0049] Compare the stability deviation value with the preset stability deviation threshold. If the stability deviation value exceeds the preset stability deviation threshold, it indicates that the overall operation stability of the bearing surface quality inspection and evaluation operation is poor, and a stability anomaly signal is generated. If the stability deviation value does not exceed the preset stability deviation threshold, it indicates that the overall operation stability of the bearing surface quality inspection and evaluation operation is good, and a stability qualified signal is generated.

[0050] Moreover, the stability analysis output module is communicatively connected to the accuracy test module. The stability analysis output module sends the stability qualified signal to the accuracy test module. When the accuracy test module receives the stability qualified signal, it analyzes the accuracy status of the bearing surface quality inspection and evaluation and generates an accuracy qualified signal or an accuracy anomaly signal through the analysis.

[0051] And send the accuracy qualified signal or accuracy abnormal signal to the display and alarm terminal. When the display and alarm terminal receives the accuracy abnormal signal, it issues a warning to remind the management personnel to optimize and improve the defect recognition detection model, improve the subsequent recognition accuracy, and facilitate the accurate assessment of the bearing surface quality. The specific analysis process of the accuracy test module is as follows:

[0052] Conduct several surface quality detection and evaluations on bearings known to have multiple defects, collect the recognition accuracy rates during each surface quality detection and evaluation, calculate the variance of all recognition accuracy rates to obtain the result consistency analysis value, and compare the result consistency analysis value with the preset result consistency analysis threshold. If the result consistency analysis value exceeds the preset result consistency analysis threshold, it indicates that the differences in the recognition results for the corresponding bearings are relatively large, and then assign the recognition judgment symbol XP-1 to the test results of the corresponding bearings;

[0053] If the result consistency analysis value does not exceed the preset result consistency analysis threshold, then mark the ratio of the number of detection times with recognition accuracy rates not exceeding the preset recognition accuracy rate threshold as the recognition accuracy difference value, and calculate the average value of all recognition accuracy rates to obtain the recognition accuracy judgment value. Compare the recognition accuracy difference value and the recognition accuracy judgment value with the preset recognition accuracy difference threshold and the preset recognition accuracy judgment threshold respectively. If the recognition accuracy difference value or the recognition accuracy judgment value exceeds the corresponding preset threshold, it indicates that the defect recognition accuracy for the corresponding bearings is relatively poor, and then assign the recognition judgment symbol XP-1 to the test results of the corresponding bearings;

[0054] After completing the tests on several bearings with multiple defects, if the number of times the recognition judgment symbol XP-1 is assigned is zero, it indicates that the recognition performance for bearing surface defects is relatively good, and then generate an accuracy qualified signal; if the number of times the recognition judgment symbol XP-1 is assigned is not zero, it indicates that the recognition performance for bearing surface defects is relatively poor, which is not conducive to ensuring the accurate assessment of the bearing surface quality, and then generate an accuracy abnormal signal.

[0055] Embodiment 2: As Figure 2 shown, the difference between this embodiment and Embodiment 1 is that the accuracy test module is communicatively connected to the bearing production line optimization judgment module. The accuracy test module sends the accuracy qualified signal to the bearing production line optimization judgment module, and when the bearing production line optimization judgment module receives the accuracy qualified signal, it analyzes the production status of the bearing production line during the detection period;

[0056] Generate a bearing production qualified signal or a bearing production abnormal signal through analysis, and send the bearing production qualified signal or the bearing production abnormal signal to the display and alarm terminal. When the display and alarm terminal receives the bearing production abnormal signal, it issues a warning to remind the management personnel to strengthen the subsequent supervision of the bearing production line, ensure the quality of the produced bearings, and reduce the production cost of the bearings. The specific analysis process of the bearing production line optimization judgment module is as follows:

[0057] Collect the bearing surface quality inspection reports of all bearings entering the scrap area during the inspection period. Based on the bearing surface quality inspection reports, obtain all the defect information existing on the corresponding bearings. Calculate the ratio of the number of bearings involved in the corresponding type of defect to the total number of bearings inspected during the inspection period to obtain the bearing defect matching value. Compare the bearing defect matching value with the corresponding preset bearing defect matching threshold. If the bearing defect matching value exceeds the preset bearing defect matching threshold, assign the easy-to-occur judgment symbol WP-1 to the corresponding type of defect;

[0058] If there is a defect type corresponding to the easy-to-occur judgment symbol WP-1 during the inspection period, it indicates that there is a frequently occurring bearing defect type in the production process of the bearing production line during the inspection period, and it is necessary to timely analyze the occurrence reasons of the corresponding type of defect and overcome it, then generate a bearing production abnormal signal; if there is no defect type corresponding to the easy-to-occur judgment symbol WP-1 during the inspection period, then calculate the ratio of the number of bearings entering the scrap area during the inspection period to the total number of bearings inspected during the inspection period to obtain the bearing scrap analysis value;

[0059] Compare the bearing scrap analysis value with the preset bearing scrap analysis threshold. If the bearing scrap analysis value exceeds the preset bearing scrap analysis threshold, it indicates that the quality of the bearings produced during the inspection period is poor, then generate a bearing production abnormal signal; if the bearing scrap analysis value does not exceed the preset bearing scrap analysis threshold, it indicates that the quality of the bearings produced during the inspection period is good, then generate a bearing production qualified signal.

[0060] Embodiment 3: As Figure 2 shown, the difference between this embodiment and Embodiment 1 and Embodiment 2 is that the bearing production line optimization judgment module is communicatively connected to the bearing production management judgment module. The bearing production management module sends the bearing production qualified signal to the bearing production management judgment module. When the bearing production management judgment module receives the bearing production qualified signal, it analyzes the management performance of the bearing production during the inspection period;

[0061] Generate a management qualified signal or a management abnormal signal through analysis, and send the management qualified signal or the management abnormal signal to the display and alarm terminal. When the display and alarm terminal receives the management abnormal signal, it issues a corresponding warning to remind the management personnel to strengthen the control of the bearing production process, and then timely repair the production equipment and train the production line operators to further ensure the smooth progress of the subsequent bearing production process and improve the bearing quality; the specific analysis process of the bearing production management judgment module is as follows:

[0062] Monitor the bearing production line, collect the average interval time for maintenance of the corresponding production equipment in the bearing production line during the detection period, compare it numerically with the corresponding preset average interval time threshold, mark the production equipment with an average interval time exceeding the preset average interval time threshold as poorly managed equipment, obtain the ratio of the number of equipment marked as poorly managed equipment during the detection period and mark it as the non-good management value;

[0063] And mark the ratio of the average interval time for maintenance of the corresponding production equipment in the bearing production line to the corresponding preset average interval time threshold as the inspection and maintenance interval measurement value, calculate the mean value of the inspection and maintenance interval measurement values of all production equipment to obtain the production line maintenance value, and collect the operation error frequency of the production line operators during the detection period (i.e., the number of errors of the production line operators recorded during the detection period);

[0064] Calculate the management judgment coefficient by weighted summation of the non-good management value, production line maintenance value, and operation error frequency; that is, assign preset weight coefficients greater than zero to the non-good management value, production line maintenance value, and operation error frequency respectively, multiply the non-good management value, production line maintenance value, and operation error frequency by the corresponding preset weight coefficients respectively, and sum the results of the three groups of multiplications to obtain the management judgment coefficient; moreover, the larger the value of the management judgment coefficient, the worse the comprehensive management status of bearing production during the detection period;

[0065] Compare the management judgment coefficient numerically with the preset management judgment coefficient threshold. If the management judgment coefficient exceeds the preset management judgment coefficient threshold, indicating that the comprehensive management status of bearing production during the detection period is poor, then generate a management exception signal; if the management judgment coefficient does not exceed the preset management judgment coefficient threshold, indicating that the comprehensive management status of bearing production during the detection period is good, then generate a management qualified signal.

[0066] The working principle of the present invention: When in use, through image acquisition and processing of the bearing surface and combining with a visual recognition model and machine learning algorithm for defect identification and quantitative evaluation, rapid detection and evaluation of the bearing surface quality are realized, significantly improving the detection efficiency and reducing the labor cost, significantly reducing the workload of bearing production management personnel, and analyzing the stability performance of the bearing surface quality detection and evaluation during the detection period through the stability analysis output module. When a stability exception signal is generated, strengthen subsequent detection supervision and optimize the detection evaluation system in a timely manner, and when a stability qualified signal is generated, test and analyze the accuracy status of the bearing surface quality detection and evaluation system through the accuracy test module. When an accuracy exception signal is generated, optimize and improve the defect identification detection model to improve the subsequent identification accuracy, which is conducive to achieving accurate evaluation of the bearing surface quality and has a high level of intelligence.

[0067] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A bearing surface quality detection and evaluation system based on a visual cognitive model, characterized in that It includes a bearing scanning and imaging unit, a surface treatment and extraction unit, a visual recognition model unit, a quantitative evaluation and output unit, a bearing sorting unit, and a display and alarm terminal; the bearing scanning and acquisition unit acquires visible light images, high-dynamic range images, and three-dimensional topography data of the bearing surface, and generates a clear surface image with multi-scale fusion; The surface treatment and extraction unit performs preprocessing operations on the received bearing surface image and extracts the characteristic information of the bearing surface. The visual recognition model unit constructs a visual recognition model using a convolutional neural network, and the visual recognition model identifies the defects on the bearing surface; The quantitative evaluation and output unit quantitatively evaluates the defects, generates a bearing surface quality inspection report, and sends the bearing surface quality inspection report to the display and alarm terminal and the bearing sorting unit; the bearing sorting unit sorts the bearings that do not meet the requirements to the scrap area and sorts the bearings that meet the requirements to the qualified area; The display and alarm terminal is communicatively connected to a stability analysis and output module. The stability analysis and output module is used to set the detection period, analyze the stability performance of the bearing surface quality inspection and evaluation during the detection period, generate a stability qualified signal or a stability abnormal signal through the analysis, and send the stability qualified signal or the stability abnormal signal to the display and alarm terminal; The stability analysis and output module is communicatively connected to a precision test module. When the precision test module receives the stability qualified signal, it analyzes the precision status of the bearing surface quality inspection and evaluation, generates a precision qualified signal or a precision abnormal signal through the analysis, and sends the precision qualified signal or the precision abnormal signal to the display and alarm terminal.

2. The bearing surface quality detection and evaluation system based on the visual cognition model according to claim 1, characterized in that The specific analysis process of the stability analysis and output module includes: When performing the surface quality inspection and evaluation of the corresponding bearing, when the selected duration exceeds the preset selection duration threshold, the effectiveness evaluation symbol ZP-1 is assigned. The number of times the effectiveness evaluation symbol ZP-1 is assigned during the detection period is obtained and its ratio is calculated with the total number of bearings subjected to surface quality inspection and evaluation during the detection period to obtain the effectiveness evaluation warning value, and the average value of all the selected durations during the detection period is calculated to obtain the selection time decision value. If the effectiveness evaluation warning value or the selection time decision value exceeds the corresponding preset threshold, a stability abnormal signal is generated.

3. The bearing surface quality detection and evaluation system based on a visual cognitive model according to claim 2, characterized in that, If both the effectiveness evaluation warning value and the selection time decision value do not exceed the corresponding preset threshold, the stability output difference value is calculated by weighted summation of the effectiveness evaluation warning value, the selection time decision value, and the obstacle judgment value. If the stability output difference value exceeds the preset stability output difference threshold, a stability abnormal signal is generated; If the stability output difference value does not exceed the preset stability output difference threshold, a stability qualified signal is generated.

4. The bearing surface quality detection and evaluation system based on a visual perception model according to claim 1, characterized in that The specific analysis process of the precision test module includes: Perform several surface quality inspection and evaluations on the bearings known to have multiple defects. If the result consistency analysis value exceeds the preset result consistency analysis threshold, the recognition judgment symbol XP-1 is assigned to the test result of the corresponding bearing; if the result consistency analysis value does not exceed the preset result consistency analysis threshold, when the recognition accuracy difference value or the recognition accuracy judgment value exceeds the corresponding preset threshold, the recognition judgment symbol XP-1 is assigned to the test result of the corresponding bearing; After completing the test on several bearings with multiple defects, if the number of times the identification judgment symbol XP-1 is assigned is zero, a qualified accuracy signal is generated; otherwise, an abnormal accuracy signal is generated.

5. The bearing surface quality detection and evaluation system based on the visual cognitive model according to claim 1, characterized in that The accuracy test module is communicatively connected to the bearing production line optimization judgment module. When the bearing production line optimization judgment module receives the accuracy qualified signal, it analyzes the production status of the bearing production line during the detection period, generates a bearing production qualified signal or a bearing production abnormality signal through analysis, and sends the bearing production qualified signal or the bearing production abnormality signal to the display alarm terminal.

6. The bearing surface quality detection and evaluation system based on the visual cognitive model according to claim 5, wherein The specific analysis process of the bearing production line optimization judgment module includes: If a defect type corresponding to the prone judgment symbol WP-1 exists during the detection period, a bearing production abnormality signal is generated; if a defect type corresponding to the prone judgment symbol WP-1 exists during the detection period, a bearing production abnormality signal is generated when the bearing scrap analysis value exceeds the preset bearing scrap analysis threshold; otherwise, a bearing production qualified signal is generated.

7. The bearing surface quality detection and evaluation system based on the visual cognitive model according to claim 5, characterized in that, The bearing production line optimization judgment module is communicatively connected to the bearing production management judgment module. When the bearing production management judgment module receives the bearing production qualified signal, it analyzes the management performance of the bearing production during the detection period, generates a management qualified signal or a management abnormality signal through the analysis, and sends the management qualified signal or the management abnormality signal to the display alarm terminal.

8. The bearing surface quality detection and evaluation system based on a visual cognitive model according to claim 7, characterized in that The specific analysis process of the bearing production management judgment module is as follows: The management judgment coefficient is calculated by weighted summing up the management non-good value, production line maintenance value and operation error frequency. If the management judgment coefficient exceeds the preset management judgment coefficient threshold, a management abnormality signal is generated; if the management judgment coefficient does not exceed the preset management judgment coefficient threshold, a management qualification signal is generated.

Images