Off-axis encoder production monitoring system based on image analysis

Abstract

The invention discloses an off-axis code disc production monitoring system based on image analysis, relates to the technical field of image processing, and is used to improve the problem of decreased monitoring effect caused by off-axis code disc defects appearing outside fixed points or angles. The system comprises the following steps: obtaining an overall image of the off-axis code disc, screening out a plurality of high-noise positions according to image pixel values ​​and determining the area of ​​each high-noise position, obtaining symmetry information of the off-axis code disc to determine whether the off-axis code disc is symmetrical, and obtaining the position of the center of gravity and the surface roughness of the off-axis code disc if it is asymmetrical, calculating a scanning ratio by using a polynomial regression algorithm, amplifying the area of ​​each noise position according to the scanning ratio, setting a scanning area threshold, classifying the quality of the code disc according to the amplified area and the scanning area threshold, collecting the test temperature of the off-axis code disc and the fixed-point position deviation of similar code discs, and re-evaluating the monitoring area by using a logistic regression algorithm based on the test temperature of the off-axis code disc and the fixed-point position deviation of the code disc.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and more specifically, to an off-axis code disc production monitoring system based on image analysis. Background Art

[0002] Image processing technology is a series of methods and techniques for analyzing and processing images, aiming to improve image quality or extract useful information. Image processing technology is applied in the off-axis code disc production monitoring process to improve the quality resolution of the off-axis code disc and increase the quality screening efficiency.

[0003] The prior art has the following deficiencies:

[0004] In the past, when monitoring the production of off-axis code discs, monitoring equipment was installed at fixed points, so only a few fixed points or angles of the off-axis code discs could be monitored. However, there are quality differences between individual off-axis code discs. When code disc defects appear outside of fixed points or angles, they can also be monitored, resulting in a decline in the overall quality of off-axis code disc production. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an off-axis code disc production monitoring system based on image analysis, which screens out multiple high-noise positions in the image through the overall image of the off-axis code disc, analyzes the stability of the off-axis code disc and calculates the monitoring area in combination with the high-noise positions, analyzes the quality of the off-axis code disc based on the monitoring area and makes targeted adjustments to the monitoring area to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An off-axis encoder production monitoring system based on image analysis includes a data acquisition module, an encoder identification module, a quality assessment module, and a monitoring reset module, and signal connections between the modules;

[0008] The data acquisition module is used to collect the overall image of the off-axis code disk and obtain the image pixel value, obtain the symmetry information of the off-axis code disk, and transmit the image pixel value and the symmetry information of the off-axis code disk to the code disk recognition module; obtain the center of gravity position and surface roughness of the off-axis code disk and transmit them to the code disk recognition module; collect the test temperature of the off-axis code disk and the distance between the fixed points of the code disk and send them to the monitoring reset module;

[0009] The code disc recognition module determines multiple key monitoring positions in the overall image of the off-axis code disc according to the image pixel value, classifies the off-axis code disc after determining the symmetry information of the off-axis code disc, calculates the center of gravity deviation according to the center of gravity position of the off-axis code disc, calculates the scanning ratio by using a polynomial regression algorithm based on the received surface roughness of the off-axis code disc and the center of gravity deviation, determines the monitoring area of ​​each key monitoring position based on the off-axis code disc category, scanning ratio and key monitoring position, and sends the monitoring area to the quality assessment module;

[0010] The quality assessment module receives the monitoring area transmitted by the code disc identification module, sets the scanning area threshold, compares each key monitoring position with the scanning area threshold, classifies the off-axis code disc into a high-quality off-axis code disc or a low-quality off-axis code disc, and sends the off-axis code disc quality type and monitoring area to the monitoring reset module;

[0011] The monitoring reset module selects a monitoring area adjustment mode according to the off-axis code disk quality type, and re-evaluates the monitoring area using a logistic regression algorithm using the test temperature and the code disk fixed point position spacing according to the monitoring area adjustment mode.

[0012] In a preferred embodiment, after receiving the image pixel values ​​of the off-axis code disk overall image, the code disk recognition module combines the pixel values ​​of all pixels into a pixel data set;

[0013] Calculate the mean and standard deviation of the image pixel values ​​in the pixel data set, use the sum of the mean and standard deviation of the image pixel values ​​as the noise screening threshold, and mark the image pixel values ​​exceeding the noise screening threshold in the overall image of the off-axis code disk as noise;

[0014] The adjacent marked pixel points are merged and compared with the preset pixel number threshold. If the number of marked pixels after merging exceeds the preset pixel number threshold, the area after the marked pixel points are merged is regarded as a high-noise area, and each high-noise area in the overall image of the off-axis code disk is regarded as a key monitoring position.

[0015] In a preferred embodiment, the symmetry information of the off-axis code disc is a regular attribute when the off-axis code disc is produced, and the regular attribute is divided into regular and irregular. If the regular attribute of the off-axis code disc is regular, the off-axis code disc is judged to be symmetrical, and the off-axis code disc is marked as a symmetrical code disc; if the regular attribute of the off-axis code disc is irregular, the off-axis code disc is judged to be asymmetrical, and the off-axis code disc is marked as an asymmetrical code disc;

[0016] When the off-axis code disc mark is an asymmetric code disc, the off-axis code disc mark is transmitted to the data acquisition module. After receiving the off-axis code disc mark, the data acquisition module obtains the center of gravity position and surface roughness of the off-axis code disc through the document management system and sends them to the code disc identification module.

[0017] In a preferred embodiment, the code disc identification module uses the center of gravity position of the off-axis code disc as the intersection of two vertical straight lines, obtains the length values ​​of the two vertical straight lines, screens out the straight line with the high length value and takes the midpoint thereof as the deviation point, and uses the distance between the center of gravity position and the deviation point position as the center of gravity deviation amount.

[0018] In a preferred embodiment, the code disc recognition module calculates the scanning ratio by using a polynomial regression algorithm based on the surface roughness of the off-axis code disc and the center of gravity deviation as follows:

[0019] Data processing: The surface roughness of the off-axis encoder and the center of gravity deviation are standardized using logarithmic transformation;

[0020] Data input: The result after the off-axis code disc surface roughness is standardized is marked as a, and the result after the off-axis code disc gravity center deviation is standardized is marked as b;

[0021] Construct the polynomial: Take the normalized results of the off-axis encoder surface roughness and the center of gravity deviation as input, and set the scan ratio as output to construct the polynomial: , where s is the scan ratio and s>1, and c is the preset conversion parameter.

[0022] In a preferred embodiment, the code disc recognition module determines the monitoring area of ​​each key monitoring position according to the following rules:

[0023] Rule 1: When the off-axis code disc is marked as a symmetrical code disc, the area of ​​the key monitoring position is used as the monitoring area;

[0024] Rule 2: When the off-axis code disc is marked as an asymmetric code disc, the product of the area of ​​the key monitoring position and the scanning ratio is used as the monitoring area.

[0025] In a preferred embodiment, the quality assessment module accesses the historical database to obtain multiple off-axis code disk monitoring areas and merges them into a monitoring area data set. The scanning area threshold is set according to the monitoring area data set using the percentile method. The data in the monitoring area data set are arranged from small to large, and N% is set as the percentile ratio. The data value exceeding N% of the data in the monitoring area data set is used as the scanning area threshold.

[0026] The quality assessment module receives the monitoring area transmitted by the code disc recognition module and compares it with the scanning area threshold. When the monitoring area exceeds the scanning area threshold, the off-axis code disc is classified as a low-quality off-axis code disc; when the monitoring area is lower than the scanning area threshold, the off-axis code disc is classified as a high-quality off-axis code disc.

[0027] In a preferred embodiment, the monitoring reset module selects the monitoring area adjustment mode according to the off-axis code disk quality type as follows:

[0028] When the off-axis code disc is a high-quality off-axis code disc, the test temperatures of multiple off-axis code discs are called in the historical database, the maximum test temperature is selected as the upper limit temperature, the ratio of the upper limit temperature to the test temperature of the off-axis code disc is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area;

[0029] When the off-axis code disc is a low-quality off-axis code disc, the test temperatures of multiple off-axis code discs and the fixed-point position spacings of multiple similar code discs are called in the historical database, the average fixed-point position spacings of similar code discs are calculated, and the fixed-point position spacings of similar code discs are subtracted from the average fixed-point position spacings of similar code discs to obtain the spacing difference of similar code discs;

[0030] The test temperature and the spacing difference of similar code disks are standardized by logarithmic transformation and then merged into the temperature coefficient data set and the spacing difference data set respectively.

[0031] In a preferred embodiment, the correction coefficient is calculated by combining the temperature coefficient data set and the spacing difference data set using a logistic regression algorithm, and the specific steps are as follows:

[0032] Randomly select a data in the temperature coefficient data set and the spacing difference data set as input, and calculate the logistic regression coefficient using the logistic regression formula: , where L is the logistic regression coefficient, e is the natural base, z is the logistic regression parameter and the absolute value of the difference between the two inputs; set the number of selections to n, randomly select a data in the temperature coefficient data set and the spacing coefficient data set as input and calculate the logistic regression coefficient, and take the median of the logistic regression coefficient as the correction coefficient;

[0033] The code disk spacing difference of the current off-axis code disk is obtained by subtracting the code disk fixed-point position spacing of the current off-axis code disk from the average fixed-point position spacing of similar code disks. The test temperature of the current off-axis code disk and the code disk spacing difference are logarithmically transformed and the logistic regression coefficient is calculated. The ratio of the logistic regression coefficient calculated for the current off-axis code disk to the correction coefficient is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area.

[0034] The technical effects and advantages of the off-axis code disc production monitoring system based on image analysis of the present invention are as follows:

[0035] The present invention obtains an overall image of an off-axis code disk, screens out a plurality of high-noise positions according to image pixel values ​​and determines the area of ​​each high-noise position, screens out the high-noise positions to narrow the analysis direction and reduce the monitoring cost, obtains the symmetry information of the off-axis code disk to determine whether the off-axis code disk is symmetrical, and obtains the center of gravity position and surface roughness of the off-axis code disk if it is asymmetrical, calculates the scanning ratio, and amplifies the area of ​​each noise position according to the scanning ratio. The area amplification is used to obtain more monitoring information, which is beneficial to supervision, sets a scanning area threshold, classifies the code disk quality according to the amplified area and the scanning area threshold, collects the test temperature of the off-axis code disk and the fixed-point position deviation of the same type of code disk, and re-evaluates the monitoring area based on the test temperature of the off-axis code disk and the fixed-point position deviation of the code disk, thereby greatly improving the production monitoring efficiency of the off-axis code disk. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of the off-axis code disc production monitoring system based on image analysis of the present invention.

[0037] Figure 2 This is a flow chart of the off-axis code disc production monitoring system based on image analysis of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] The present invention obtains the overall image of the off-axis code disk, screens out multiple high-noise positions according to the image pixel values ​​and determines the area of ​​each high-noise position, obtains the symmetry information of the off-axis code disk to determine whether the off-axis code disk is symmetrical, and if it is asymmetric, obtains the center of gravity position and surface roughness of the off-axis code disk, calculates the scanning ratio, amplifies the area of ​​each noise position according to the scanning ratio, sets the scanning area threshold, classifies the code disk quality according to the amplified area and the scanning area threshold, collects the test temperature of the off-axis code disk and the fixed-point position deviation of the same type of code disk, and re-evaluates the monitoring area based on the test temperature of the off-axis code disk and the fixed-point position deviation of the code disk, thereby greatly improving the production monitoring efficiency of the off-axis code disk.

[0040] Embodiment, an off-axis code disc production monitoring system based on image analysis, such as Figure 1 As shown, it includes a data acquisition module, a code disc recognition module, a quality assessment module and a monitoring reset module, and the signals between the modules are connected;

[0041] The functions of each module are as follows:

[0042] The data acquisition module is used to collect the overall image of the off-axis code disk and obtain the image pixel value, obtain the symmetry information of the off-axis code disk, and transmit the image pixel value and the symmetry information of the off-axis code disk to the code disk recognition module; after receiving the off-axis code disk mark transmitted by the code disk recognition module, the center of gravity position and surface roughness of the off-axis code disk are obtained and transmitted back to the code disk recognition module; after receiving the off-axis code disk quality type transmitted by the quality assessment module, the test temperature of the off-axis code disk and the distance between the fixed points of the code disk are collected and sent to the monitoring reset module;

[0043] The code disc recognition module determines multiple key monitoring positions in the overall image of the off-axis code disc according to the image pixel value, classifies the off-axis code disc after determining the symmetry information of the off-axis code disc, marks the off-axis code disc differently according to the category and transmits it to the data acquisition module, calculates the center of gravity deviation after receiving the center of gravity position of the off-axis code disc, calculates the scanning ratio by using a polynomial regression algorithm based on the received surface roughness of the off-axis code disc and the center of gravity deviation, determines the monitoring area of ​​each key monitoring position based on the off-axis code disc category, scanning ratio and key monitoring position, and sends the monitoring area to the quality assessment module and the monitoring reset module;

[0044] The quality assessment module receives the monitoring area transmitted by the code disc recognition module, sets the scanning area threshold, compares each key monitoring position with the scanning area threshold to classify the off-axis code disc quality, and transmits the off-axis code disc quality type to the data acquisition module and the monitoring reset module;

[0045] The monitoring reset module selects a monitoring area adjustment mode according to the off-axis code disk quality type, and re-evaluates the monitoring area using a logistic regression algorithm using the test temperature and the code disk fixed point position spacing according to the monitoring area adjustment mode.

[0046] It should be noted that the system calculates and obtains different key monitoring areas through off-axis code disk overall image analysis to improve monitoring efficiency.

[0047] After the data acquisition module acquires the overall image of the off-axis encoder, it obtains the image pixel value through the image processing library.

[0048] The symmetry information of the off-axis code disk is obtained, the image pixel value and the symmetry information of the off-axis code disk are transmitted to the code disk recognition module, the document management system is accessed to obtain the symmetry information of the current off-axis code disk, and the data acquisition module obtains the image pixel value of the overall image of the off-axis code disk and the symmetry information of the off-axis code disk and transmits them to the code disk recognition module.

[0049] It should be noted that the image processing library is a tool set for processing and analyzing images, and is used to obtain the image pixel values ​​of the overall image of the off-axis code disk. The document management system is a software system for storing and managing electronic documents and images. In this example, it is used to obtain the design documents of the off-axis code disk. The design documents of the off-axis code disk contain the symmetry information of the off-axis code disk.

[0050] After receiving the image pixel values ​​of the off-axis code disk overall image, the code disk recognition module merges the pixel values ​​of all pixels into a pixel data set, calculates the average value and standard deviation of the image pixel values ​​in the pixel data set, takes the sum of the average value and standard deviation of the image pixel values ​​as the noise screening threshold, marks the image pixel values ​​in the off-axis code disk overall image that exceed the noise screening threshold as noise, and merges adjacent marked pixel points and compares them with the preset pixel quantity threshold;

[0051] If the number of pixel points after merging exceeds a preset pixel number threshold, the area after the merging of the pixel points is taken as a high noise area, and each high noise area in the overall image of the off-axis code disk is taken as a key monitoring position.

[0052] It should be noted that, in the above, the noise screening threshold is only used as an example, and it can also be obtained by summing the average value and different multiples of the standard deviation. The pixel number threshold can be set according to the accuracy requirements, which will not be elaborated here.

[0053] The symmetry information of the off-axis code disc is the regular attribute when the off-axis code disc is produced. The regular attribute is divided into regular and irregular. If the regular attribute of the off-axis code disc is regular, the off-axis code disc is judged to be symmetrical and marked as a symmetrical code disc; if the regular attribute of the off-axis code disc is irregular, the off-axis code disc is judged to be asymmetrical and marked as an asymmetrical code disc;

[0054] When the off-axis code disc mark is an asymmetric code disc, the off-axis code disc mark is transmitted to the data acquisition module. After receiving the off-axis code disc mark, the data acquisition module obtains the center of gravity position and surface roughness of the off-axis code disc through the document management system and sends them to the code disc identification module;

[0055] The code disc recognition module receives the center of gravity position of the off-axis code disc and calculates the center of gravity deviation of the off-axis code disc. The center of gravity position of the off-axis code disc is used as the intersection of two vertical lines, and the length values ​​of the two vertical lines are obtained. The line with the highest length value is selected and the midpoint is taken as the deviation point. The distance between the center of gravity position and the deviation point position is taken as the center of gravity deviation;

[0056] The code disc recognition module uses a polynomial regression algorithm to calculate the scanning ratio based on the surface roughness of the off-axis code disc and the center of gravity deviation. The specific steps are as follows:

[0057] Data processing: The off-axis encoder surface roughness and center of gravity deviation are normalized using logarithmic transformation: ,in, The smoothness of the off-axis encoder surface or the center of gravity deviation, It is the normalized result of the off-axis code disc surface roughness or center of gravity deviation;

[0058] Data input: The result after the off-axis code disc surface roughness is standardized is marked as a, and the result after the off-axis code disc gravity center deviation is standardized is marked as b;

[0059] Construct the polynomial: Take the normalized results of the off-axis encoder surface roughness and the center of gravity deviation as input, and set the scan ratio as output to construct the polynomial: , where s is the scan ratio and s>1, and c is the conversion parameter.

[0060] It should be noted that the conversion parameters in the polynomial regression algorithm are not unique and are adjusted by professionals in the field according to actual conditions, and will not be analyzed here.

[0061] The code disc recognition module determines the monitoring area of ​​each key monitoring location according to the following rules:

[0062] Rule 1: When the off-axis code disc is marked as a symmetrical code disc, the area of ​​the key monitoring position is used as the monitoring area;

[0063] Rule 2: When the off-axis code disc is marked as an asymmetric code disc, the product of the area of ​​the key monitoring position and the scanning ratio is used as the monitoring area.

[0064] The code disc recognition module transmits the calculated off-axis code disc monitoring area to the quality assessment module and the monitoring reset module.

[0065] It should be noted that when the key monitoring position is expanded according to the scanning ratio, the area of ​​the key monitoring position is geometrically expanded. Geometric expansion means increasing the length or width of the boundary so that the ratio of the new area is consistent with that of the original area.

[0066] The quality assessment module accesses the historical database to obtain multiple off-axis encoder monitoring areas and merges them into a monitoring area data set. The scanning area threshold is set using the percentile method according to the monitoring area data set. The data in the monitoring area data set are arranged from small to large, and N% is set as the percentile ratio. The data value exceeding N% of the data in the monitoring area data set is used as the scanning area threshold.

[0067] For example, if the percentile ratio is set to 60%, then data with values ​​exceeding 60% of the values ​​in the monitoring area dataset are selected from the monitoring area dataset as the scanning area threshold.

[0068] The quality assessment module receives the monitoring area transmitted by the code disc recognition module and compares it with the scanning area threshold. When the monitoring area exceeds the scanning area threshold, the off-axis code disc is classified as a low-quality off-axis code disc; when the monitoring area is lower than the scanning area threshold, the off-axis code disc is classified as a high-quality off-axis code disc.

[0069] It should be noted that the historical database is a database that stores time-related historical data, and is used to call the off-axis code disk data processed in the past. In this example, it is used to obtain the off-axis code disk monitoring area in the historical data; the monitoring area is determined by the area of ​​the key monitoring position and the scanning ratio, wherein the area of ​​the key monitoring position is obtained by noise screening, and the larger the key monitoring area, the greater the image noise, the lower the image stability, and the more enhanced monitoring is needed; the scanning ratio is calculated from the surface roughness and center of gravity deviation of the off-axis code disk, and the larger the scanning ratio, the greater the surface roughness and center of gravity deviation of the off-axis code disk, the lower its stability, and the more enhanced monitoring is needed; therefore, the smaller the monitoring area, the higher the stability of the off-axis code disk and the better its quality.

[0070] After receiving the off-axis code disc quality type transmitted by the quality assessment module, the data acquisition module selects and collects the data, obtains the test temperature of the off-axis code disc through the database management system, and accesses the historical database to obtain the fixed-point position spacing of the code disc;

[0071] The test temperature is the average value of the temperature range in which the current off-axis encoder works normally; the spacing between the fixed positions of the encoder is the spacing thickness value of the same type of encoder at the same position in the top view;

[0072] The data acquisition module follows the following steps when selecting and collecting data:

[0073] When the off-axis code disc is a high-quality off-axis code disc, the test temperature of the off-axis code disc is obtained; when the off-axis code disc is a low-quality off-axis code disc, the test temperature of the off-axis code disc and the distance between the fixed-point positions of the code disc are obtained;

[0074] The data acquisition module selects different acquisition methods according to the off-axis code disc quality type to collect data and then sends the data to the monitoring reset module.

[0075] It should be noted that the database management system is a structured method for storing, retrieving and managing data, which is used to obtain the test temperature of the off-axis code disk in this example.

[0076] The monitoring reset module selects the monitoring area adjustment mode according to the off-axis code disk quality type as shown below:

[0077] When the off-axis code disc is a high-quality off-axis code disc, the test temperatures of multiple off-axis code discs are called in the historical database, the maximum test temperature is selected as the upper limit temperature, the ratio of the upper limit temperature to the test temperature of the off-axis code disc is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area;

[0078] When the off-axis code disc is a low-quality off-axis code disc, the test temperatures of multiple off-axis code discs and the fixed-point position spacings of multiple similar code discs are called in the historical database, the average fixed-point position spacings of similar code discs are calculated, and the fixed-point position spacings of similar code discs are subtracted from the average fixed-point position spacings of similar code discs to obtain the spacing difference of similar code discs;

[0079] The test temperature and the spacing difference of the same code disks are standardized by logarithmic transformation and then merged into a temperature coefficient data set and a spacing difference data set respectively;

[0080] The correction coefficients are calculated by using the logistic regression algorithm based on the temperature coefficient data set and the spacing difference data set. The specific steps are as follows:

[0081] Randomly select a data in the temperature coefficient data set and the spacing difference data set as input, and calculate the logistic regression coefficient using the logistic regression formula: , where L is the logistic regression coefficient, e is the natural base, z is the logistic regression parameter and the absolute value of the difference between the two inputs; set the number of selections to n, randomly select a data in the temperature coefficient data set and the spacing coefficient data set as input and calculate the logistic regression coefficient, and take the median of the logistic regression coefficient as the correction coefficient;

[0082] The code disk spacing difference of the current off-axis code disk is obtained by subtracting the code disk fixed-point position spacing of the current off-axis code disk from the average fixed-point position spacing of similar code disks. The test temperature of the current off-axis code disk and the code disk spacing difference are logarithmically transformed and the logistic regression coefficient is calculated. The ratio of the logistic regression coefficient calculated for the current off-axis code disk to the correction coefficient is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area.

[0083] It should be noted that when the monitoring area is adjusted using the update ratio, geometric expansion is still performed.

[0084] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0085] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0086] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0087] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0088] Finally: 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, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Off-axis encoder production monitoring system based on image analysis, characterized in that: It includes data acquisition module, code disc recognition module, quality assessment module and monitoring reset module, and the signal connection between each module; The data acquisition module is used to collect the overall image of the off-axis code disk and obtain the image pixel value, obtain the symmetry information of the off-axis code disk, and transmit the image pixel value and the symmetry information of the off-axis code disk to the code disk recognition module; obtain the center of gravity position and surface roughness of the off-axis code disk and transmit them to the code disk recognition module; collect the test temperature of the off-axis code disk and the distance between the fixed points of the code disk and send them to the monitoring reset module; The code disc recognition module determines multiple key monitoring positions in the overall image of the off-axis code disc according to the image pixel value, classifies the off-axis code disc after determining the symmetry information of the off-axis code disc, calculates the center of gravity deviation according to the center of gravity position of the off-axis code disc, calculates the scanning ratio by using a polynomial regression algorithm based on the received surface roughness of the off-axis code disc and the center of gravity deviation, determines the monitoring area of ​​each key monitoring position based on the off-axis code disc category, scanning ratio and key monitoring position, and sends the monitoring area to the quality assessment module; The quality assessment module receives the monitoring area transmitted by the code disc identification module, sets the scanning area threshold, compares each key monitoring position with the scanning area threshold, classifies the off-axis code disc into a high-quality off-axis code disc or a low-quality off-axis code disc, and sends the off-axis code disc quality type and monitoring area to the monitoring reset module; The monitoring reset module selects a monitoring area adjustment mode according to the off-axis code disk quality type, and re-evaluates the monitoring area using a logistic regression algorithm based on the test temperature and the code disk fixed point position spacing according to the monitoring area adjustment mode; The monitoring reset module selects the monitoring area adjustment mode according to the off-axis code disk quality type as shown below: When the off-axis code disc is a high-quality off-axis code disc, the test temperatures of multiple off-axis code discs are called in the historical database, the maximum test temperature is selected as the upper limit temperature, the ratio of the upper limit temperature to the test temperature of the off-axis code disc is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area; When the off-axis code disc is a low-quality off-axis code disc, the test temperatures of multiple off-axis code discs and the fixed-point position spacings of multiple similar code discs are called in the historical database, the average fixed-point position spacings of similar code discs are calculated, and the fixed-point position spacings of similar code discs are subtracted from the average fixed-point position spacings of similar code discs to obtain the spacing difference of similar code discs; The test temperature and the spacing difference of similar code disks are standardized by logarithmic transformation and then merged into the temperature coefficient data set and the spacing difference data set respectively.

2. The off-axis code disc production monitoring system based on image analysis according to claim 1 is characterized in that: After receiving the image pixel values ​​of the off-axis code disk overall image, the code disk recognition module combines the pixel values ​​of all pixels into a pixel data set; Calculate the mean and standard deviation of the image pixel values ​​in the pixel data set, use the sum of the mean and standard deviation of the image pixel values ​​as the noise screening threshold, and mark the image pixel values ​​exceeding the noise screening threshold in the overall image of the off-axis code disk as noise; The adjacent marked pixel points are merged and compared with the preset pixel number threshold. If the number of marked pixels after merging exceeds the preset pixel number threshold, the area after the marked pixel points are merged is regarded as a high-noise area, and each high-noise area in the overall image of the off-axis code disk is regarded as a key monitoring position.

3. The off-axis code disc production monitoring system based on image analysis according to claim 1 is characterized in that: The symmetry information of the off-axis code disc is the regular attribute when the off-axis code disc is produced. The regular attribute is divided into regular and irregular. If the regular attribute of the off-axis code disc is regular, the off-axis code disc is judged to be symmetrical and marked as a symmetrical code disc; if the regular attribute of the off-axis code disc is irregular, the off-axis code disc is judged to be asymmetrical and marked as an asymmetrical code disc; When the off-axis code disc mark is an asymmetric code disc, the off-axis code disc mark is transmitted to the data acquisition module. After receiving the off-axis code disc mark, the data acquisition module obtains the center of gravity position and surface roughness of the off-axis code disc through the document management system and sends them to the code disc identification module; The document management system is a software system for storing and managing electronic documents and images, and is used to obtain the design documents of the off-axis code disc, including the symmetry information of the off-axis code disc, the center of gravity position of the off-axis code disc, and the surface roughness.

4. The off-axis code disc production monitoring system based on image analysis according to claim 3 is characterized in that: The code disc recognition module uses the off-axis code disc center of gravity position as the intersection of two vertical lines, obtains the length values ​​of the two vertical lines, selects the line with the highest length value and takes its midpoint as the deviation point, and uses the distance between the center of gravity position and the deviation point position as the center of gravity deviation amount.

5. The off-axis code disc production monitoring system based on image analysis according to claim 1, characterized in that: The code disc recognition module uses the polynomial regression algorithm to calculate the scanning ratio based on the off-axis code disc surface roughness and the center of gravity deviation. The steps are as follows: Data processing: The surface roughness of the off-axis encoder and the center of gravity deviation are standardized using logarithmic transformation; Data input: The result after the off-axis code disc surface roughness is standardized is marked as a, and the result after the off-axis code disc gravity center deviation is standardized is marked as b; Construct the polynomial: Take the normalized results of the off-axis encoder surface roughness and the center of gravity deviation as input, and set the scan ratio as output to construct the polynomial: , where s is the scan ratio and s>1, and c is the preset conversion parameter; Logarithmic transformation is an algorithm that uses logarithms to standardize data, and is used to standardize the surface roughness and center of gravity deviation of the off-axis code disk.

6. The off-axis code disc production monitoring system based on image analysis according to claim 5 is characterized in that: The code disc recognition module determines the monitoring area of ​​each key monitoring location according to the following rules: Rule 1: When the off-axis code disc is marked as a symmetrical code disc, the area of ​​the key monitoring position is used as the monitoring area; Rule 2: When the off-axis code disc is marked as an asymmetric code disc, the product of the area of ​​the key monitoring position and the scanning ratio is used as the monitoring area.

7. The off-axis code disc production monitoring system based on image analysis according to claim 1 is characterized in that: The quality assessment module accesses the historical database to obtain multiple off-axis encoder monitoring areas and merges them into a monitoring area data set. The scanning area threshold is set using the percentile method according to the monitoring area data set. The data in the monitoring area data set are arranged from small to large, and N% is set as the percentile ratio. The data value exceeding N% of the data in the monitoring area data set is used as the scanning area threshold. The quality assessment module receives the monitoring area transmitted by the code disc recognition module and compares it with the scanning area threshold. When the monitoring area exceeds the scanning area threshold, the off-axis code disc is classified as a low-quality off-axis code disc; when the monitoring area is lower than the scanning area threshold, the off-axis code disc is classified as a high-quality off-axis code disc. A historical database is a database that stores time-related historical data and is used to obtain the off-axis code disk monitoring area in the historical data.

8. The off-axis code disc production monitoring system based on image analysis according to claim 1, characterized in that: The correction coefficients are calculated by using the logistic regression algorithm based on the temperature coefficient data set and the spacing difference data set. The specific steps are as follows: Randomly select a data in the temperature coefficient data set and the spacing difference data set as input, and calculate the logistic regression coefficient using the logistic regression formula: , where L is the logistic regression coefficient, e is the natural base, z is the logistic regression parameter and the absolute value of the difference between the two inputs; set the number of selections to n, randomly select a data in the temperature coefficient data set and the spacing coefficient data set as input and calculate the logistic regression coefficient, and take the median of the logistic regression coefficient as the correction coefficient; The code disk spacing difference of the current off-axis code disk is obtained by subtracting the code disk fixed-point position spacing of the current off-axis code disk from the average fixed-point position spacing of similar code disks. The test temperature of the current off-axis code disk and the code disk spacing difference are logarithmically transformed and the logistic regression coefficient is calculated. The ratio of the logistic regression coefficient calculated for the current off-axis code disk to the correction coefficient is used as the update ratio, and the product of the update ratio and the monitoring area is used as the new monitoring area.

Images