Rubber roller thread online correction method based on machine vision

Abstract

The invention relates to the technical field of rubber roller threads, and particularly discloses a rubber roller thread online correction method based on machine vision, which comprises the following steps: processing and analyzing a depth error value of a rubber roller thread surface to obtain a depth error early warning value, if the depth error early warning value is greater than a depth error early warning threshold value, generating an analysis signal, and based on the analysis signal, carrying out online correction on the rubber roller thread. The grinding pressure value and the depth error value of the thread surface of the rubber roller are processed and analyzed to obtain a coincidence influence value, if the coincidence influence value is larger than a coincidence influence threshold value, a suspected influence signal is generated, and based on the suspected influence signal, grinding pressure deviation values and depth error values of the thread surface of the rubber roller at different time points are processed and analyzed to obtain the grinding pressure deviation value of the thread surface of the rubber roller. According to the method, the deviation of the grinding pressure borne by the rubber roller thread is analyzed, whether the rubber roller thread depth error is abnormal due to the fact that the deviation of the grinding pressure borne by the surface of the rubber roller thread is different or not is judged, and if yes, the current grinding pressure deviation is adjusted.

Description

Technical Field

[0001] The invention relates to the technical field of rubber roller threads, and in particular to an online correction method for rubber roller threads based on machine vision. Background Art

[0002] With the rapid development of modern manufacturing industry, the requirements for product quality and production efficiency are getting higher and higher. As an important component widely used in many industries such as papermaking, printing, metallurgy, etc., the processing size and accuracy of the thread of the rubber roller play an important role in the processing and use of the rubber roller. Strict measurement and correction of the appearance and geometric parameters of the rubber roller thread have become the key link to ensure product quality and performance.

[0003] The Chinese invention patent with publication number CN115127479A discloses an online detection and correction method for rubber roller threads based on machine vision, including: collecting rubber roller thread images; preprocessing the collected images to remove noise and burrs of the rubber roller images; obtaining broken line contours and pixel coordinates of thread crests; obtaining pixel coordinates of rubber roller thread contours and tooth bottoms; calculating rubber roller thread geometric parameters, and converting the calculated pixel size into actual physical size; correcting rubber rollers with errors; using a non-contact measurement method to detect rubber roller threads, solving the problem that the current rubber roller thread measurement cannot be directly and non-contacted; the measurement results are sent to the machine tool CNC system through a bus communication protocol, and the CNC system corrects the position that needs to be reprocessed according to the rubber roller size measurement value, thereby improving production efficiency and ensuring the factory accuracy of the product; using an improved filtering method to remove these noises, improve the accuracy of the measurement results, and improve the measurement accuracy.

[0004] However, in the above-mentioned prior art, there is a lack of judgment on the factors affecting the surface depth data of the rubber roller thread. Analyzing the rubber roller thread depth data and grinding pressure data can avoid abnormal rubber roller thread depth errors caused by different grinding pressure deviations. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for online correction of rubber roller threads based on machine vision to solve the technical problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The present invention provides a method for online correction of rubber roller threads based on machine vision, comprising:

[0008] Step 1: Process and analyze the depth error value of the thread surface of the rubber roller to obtain a depth error warning value YJ. If the depth error warning value is greater than the depth error warning threshold, an analysis signal is generated;

[0009] Step 2: Based on the analysis signal, the grinding pressure value and depth error value of the thread surface of the rubber roller are processed and analyzed to obtain the overlap influence value CY. If the overlap influence value CY is greater than the overlap influence threshold, a suspected influence signal is generated;

[0010] Step 3: Based on the suspected influencing signal, the grinding pressure deviation value and the depth error value of the rubber roller thread surface at different time points are processed and analyzed to obtain the change synchronization value BT. If the change synchronization value is greater than the change synchronization threshold, it is determined that the grinding pressure deviation has a great influence on the rubber roller thread depth error;

[0011] Step 4: If the grinding pressure deviation has a great influence on the thread depth error of the rubber roller, a grinding pressure deviation adjustment value is obtained, and the current grinding pressure deviation is adjusted based on the grinding pressure deviation adjustment value.

[0012] As a further solution of the present invention: the depth error warning value YJ is obtained in the following manner:

[0013] Obtain the depth error value of the thread surface of the rubber roller, and analyze to obtain the analysis continuity ratio SJ and the depth approach value YC;

[0014] The continuous analysis ratio SJ and the depth error approximation value YC are processed by the formula: The depth error warning value YJ is obtained, where s1 and s2 are both preset proportional coefficients.

[0015] As a further solution of the present invention: the method for obtaining the analysis continuity ratio SJ is:

[0016] The depth error value is compared with the depth error standard value and the depth error reference value respectively. If the depth error value is less than or equal to the depth error reference value, a depth normal signal is generated. If the depth error value is greater than the depth error reference value and the depth error value is less than the depth error standard value, a depth abnormal signal is generated.

[0017] The duration of the deep abnormal signal is obtained and marked as the analysis duration. The analysis duration is ratioed with the duration corresponding to the monitoring period to obtain the analysis duration ratio, which is marked as SJ.

[0018] As a further solution of the present invention: the depth error approximation value YC is obtained in the following manner:

[0019] The time period corresponding to the continuous occurrence time of the depth anomaly signal is marked as the continuous period, and the continuous period is divided into several time nodes with equal time intervals. The depth error value corresponding to each time node is obtained, and the sum and average are taken to obtain the depth error mean. The depth error mean is subtracted from the depth error standard value, and the absolute value of the difference is taken to obtain the depth deviation value. The depth deviation value is ratioed with the depth error standard value to obtain the depth error approximation value, which is marked as YC.

[0020] As a further solution of the present invention: the overlap influence value CY is obtained in the following manner:

[0021] Obtain the grinding pressure value of the thread surface of the rubber roller during the grinding time, determine the normal pressure time and the abnormal pressure time, overlap the analysis duration with the abnormal pressure time, and obtain the abnormal overlap period ratio YH;

[0022] The analysis duration is overlapped with the normal pressure time to obtain the normal overlap period ratio EH;

[0023] The data of abnormal overlap period ratio YH and normal overlap period ratio EH are processed by the formula: The overlap influence value CY is calculated, where a1 and a2 are both preset proportional coefficients, and both a1 and a2 are greater than 0.

[0024] As a further solution of the present invention: the abnormal overlap period ratio YH is obtained in the following manner:

[0025] Compare the grinding pressure value of the rubber roller thread surface during the grinding time with the preset grinding pressure value:

[0026] If the grinding pressure value is not equal to the preset grinding pressure value, a pressure abnormality signal is generated;

[0027] Obtain the duration of the pressure abnormality signal, and mark it as the pressure abnormality time;

[0028] The analysis duration is overlapped with the pressure abnormality time, specifically:

[0029] The time period that overlaps between the analysis duration and the pressure abnormality time is extracted and marked as the abnormal overlap period. The duration corresponding to the abnormal overlap period is processed by ratio with the analysis duration to obtain the abnormal overlap period proportion YH.

[0030] As a further solution of the present invention: the normal overlap period ratio EH is obtained in the following manner:

[0031] If the grinding pressure value is equal to the preset grinding pressure value, a normal pressure signal is generated;

[0032] Obtain the duration of the normal pressure signal, and mark it as the normal pressure time;

[0033] The analysis duration is overlapped with the normal pressure time, specifically:

[0034] The time period that overlaps between the analysis duration and the normal pressure time is extracted and marked as the normal overlap period. The duration corresponding to the normal overlap period is ratio-processed with the analysis duration to obtain the normal overlap period ratio EH.

[0035] As a further solution of the present invention: the change synchronization value BT is obtained in the following manner:

[0036] Constructing a depth error variation curve within the abnormal overlap period, and determining a first synchronization value YT and a second synchronization value ET according to the depth error variation curve;

[0037] The first synchronization value and the second synchronization value are processed by the formula: The change synchronization value is calculated, wherein b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0.

[0038] As a further solution of the present invention: the first synchronization value YT and the second synchronization value ET are obtained in the following manner:

[0039] During the abnormal overlap period, the grinding pressure deviation value and the depth error value at different time points are obtained and marked in the XY two-dimensional coordinate system to obtain the depth error change curve, where the X-axis represents the grinding pressure deviation value and the Y-axis represents the depth error value;

[0040] The grinding pressure deviation value is obtained by performing difference processing between the grinding pressure value and the preset grinding pressure value and taking the absolute value;

[0041] Divide the depth error variation curve into a number of sub-curves with equal horizontal lengths;

[0042] Get the slope value of the sub-curve;

[0043] If the slope value is greater than 0, the subcurve is marked as a growth subcurve;

[0044] If the slope value is less than or equal to 0, the sub-curve is marked as a non-increasing sub-curve;

[0045] Count the number of growth sub-curves and perform ratio processing with the total number of sub-curves to obtain the first synchronization value YT;

[0046] The slope values ​​of all growth sub-curves are integrated into a slope value data group, and the variance of the slope value data group is obtained to obtain a second synchronization value ET;.

[0047] As a further solution of the present invention: the grinding pressure deviation adjustment value is obtained by:

[0048] All slope values ​​contained in the slope value data group are summed to obtain an adjustment coefficient;

[0049] The current depth error value is obtained and compared with the adjustment coefficient to obtain the grinding pressure deviation adjustment value.

[0050] Beneficial effects of the present invention:

[0051] (1) The present invention processes and analyzes the depth error value to obtain the analysis continuity ratio SJ and the warning deviation degree performance value YC, performs data processing based on the analysis continuity ratio and the depth error proximity value YC to obtain the depth error warning value YJ, and determines whether to generate an analysis signal based on the depth error warning value YJ. The present invention can timely detect abnormal changes in the thread surface of the rubber roller by real-time collection and analysis of depth data, generate analysis signals, and reduce the possibility of manual intervention and misjudgment;

[0052] (2) The present invention processes and analyzes the grinding pressure data and the depth data based on the analysis signal to obtain the overlap influence value CY, and determines whether the grinding pressure is suspected to affect the depth error of the rubber roller thread. If so, based on the processing and analysis of the grinding pressure deviation value and the depth error value at different time points, a first synchronization value YT and a second synchronization value ET are obtained. Based on the processing and analysis of the first synchronization value YT and the second synchronization value ET, a change synchronization value BT is obtained to determine whether the grinding pressure deviation affects the depth error of the rubber roller thread. If so, a grinding pressure deviation adjustment value is obtained, and the current grinding pressure is adjusted based on the grinding pressure deviation adjustment value. The present invention analyzes the grinding pressure on the rubber roller thread to determine whether the abnormal depth error of the rubber roller thread is caused by the different grinding pressure deviations on the surface of the rubber roller thread. If the grinding pressure deviation affects the abnormal depth error of the rubber roller thread, the current grinding pressure deviation is adjusted. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present invention will be further described below in conjunction with the accompanying drawings.

[0054] Figure 1 It is a flowchart of the steps of a method for online correction of rubber roller threads based on machine vision according to an embodiment of the present invention;

[0055] Figure 2 It is a flow chart for obtaining the change synchronization value of a method for online correction of rubber roller threads based on machine vision described in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] Embodiment 1:

[0058] See also Figure 1 As shown, a method for online correction of rubber roller threads based on machine vision according to an embodiment of the present invention includes:

[0059] Step 1: During the monitoring period, the depth data of the thread surface of the rubber roller is obtained in real time, wherein the depth data includes a depth error value, the depth data is processed and analyzed to obtain an analysis continuity ratio SJ and an early warning deviation degree performance value YC, data processing is performed based on the analysis continuity ratio SJ and the early warning deviation degree performance value YC to obtain a depth error early warning value YJ, and whether to generate an analysis signal is determined based on the depth error early warning value YJ;

[0060] Use a three-dimensional coordinate measuring machine to detect the depth of the rubber roller thread and obtain the depth value;

[0061] Perform difference processing on the depth value and the preset depth value, and take the absolute value of the difference to obtain the depth error value;

[0062] In some embodiments, the depth error value is compared with the depth error standard value and the depth error reference value respectively. If the depth error value is less than or equal to the depth error reference value, a depth normal signal is generated. If the depth error value is greater than the depth error reference value and the depth error value is less than the depth error standard value, a depth abnormal signal is generated.

[0063] It should be noted that the depth error reference value is a warning value set by those skilled in the art based on past experience and is smaller than the depth error standard value to prevent the depth error from exceeding the standard range;

[0064] Exemplarily, the duration of the deep abnormal signal is obtained, marked as the analysis duration, and the analysis duration is processed with the duration corresponding to the monitoring period to obtain the analysis duration ratio, which is marked as SJ;

[0065] Mark the time period corresponding to the continuous occurrence time of the depth anomaly signal as the continuous period, divide the continuous period into several time nodes with equal time intervals, obtain the depth error value corresponding to each time node, sum and average them to obtain the depth error mean, perform difference processing on the depth error mean and the depth error standard value, and take the absolute value of the difference to obtain the depth deviation value, perform ratio processing on the depth deviation value and the depth error standard value to obtain the depth error approximate value, and mark it as YC;

[0066] The continuous analysis ratio SJ and the depth error approximation value YC are processed by the formula: Obtain a depth error warning value YJ, where s1 and s2 are both preset proportional coefficients, and both s1 and s2 are greater than 0;

[0067] In some embodiments, the depth error warning value is compared with a depth error warning threshold;

[0068] If the depth error warning value is greater than the depth error warning threshold, an analysis signal is generated;

[0069] If the depth error warning value is less than or equal to the depth error warning threshold, no processing is performed;

[0070] The technical solution of the embodiment of the present invention is: within the monitoring period, the depth data of the threaded surface of the rubber roller is acquired in real time, wherein the depth data includes a depth error value; the depth data is processed and analyzed to obtain an analysis continuity ratio SJ and a warning deviation degree performance value YC; data processing is performed based on the analysis continuity ratio and the depth error proximity value YC to obtain a depth error warning value YJ; based on the depth error warning value YJ, it is determined whether to generate an analysis signal. The present invention can timely detect abnormal changes in the threaded surface of the rubber roller and generate analysis signals by real-time collection and analysis of depth data, thereby reducing the possibility of manual intervention and misjudgment.

[0071] Embodiment 2:

[0072] Based on Example 1, please refer to Figure 1 , Figure 2 As shown, the method for online correction of rubber roller thread based on machine vision described in the embodiment of the present invention also includes the following steps:

[0073] Step 2: Based on the analysis signal, the grinding pressure data of the thread surface of the rubber roller during the grinding time is obtained, wherein the grinding pressure data includes the grinding pressure value, and based on the grinding pressure data and the depth data processing and analysis, the abnormal overlap period ratio YH and the normal overlap period ratio EH are obtained, and the data is processed based on the abnormal overlap period ratio YH and the normal overlap period ratio EH to obtain the overlap influence value CY, and based on the overlap influence value CY, it is judged whether the grinding pressure is suspected of affecting the thread depth of the rubber roller, and if so, a suspected influence signal is generated;

[0074] Specifically, the grinding pressure value of the thread surface of the rubber roller during the grinding time is obtained and compared with the preset grinding pressure value:

[0075] If the grinding pressure value is not equal to the preset grinding pressure value, a pressure abnormality signal is generated;

[0076] If the grinding pressure value is equal to the preset grinding pressure value, a normal pressure signal is generated;

[0077] It should be noted that the grinding time represents the period from the start of grinding to the generation of the analysis signal;

[0078] Obtain the duration of the abnormal pressure signal and mark it as the abnormal pressure time; obtain the duration of the normal pressure signal and mark it as the normal pressure time;

[0079] The analysis duration is overlapped with the pressure abnormality time, specifically:

[0080] Extract the time period that overlaps the analysis duration and the pressure abnormality time, mark it as the abnormal overlap period, and perform ratio processing on the duration corresponding to the abnormal overlap period and the analysis duration to obtain the abnormal overlap period ratio YH;

[0081] The analysis duration is overlapped with the normal pressure time, specifically:

[0082] Extract the time period that overlaps the analysis duration and the normal pressure time, mark it as the normal overlap period, and perform ratio processing on the duration corresponding to the normal overlap period and the analysis duration to obtain the normal overlap period ratio EH;

[0083] The data of abnormal overlap period ratio YH and normal overlap period ratio EH are processed by the formula: The overlap influence value CY is calculated, wherein a1 and a2 are both preset proportional coefficients, and a1 and a2 are both greater than 0;

[0084] Compare the coincidence impact value CY with the coincidence impact threshold:

[0085] If the coincidence impact value CY is greater than the coincidence impact threshold, a suspected impact signal is generated;

[0086] If the overlap influence value CY is less than or equal to the overlap influence threshold, it means that the grinding pressure has little effect on the thread depth error of the rubber roller, and further analysis of other influencing factors is required, including but not limited to temperature, wear degree of the turning tool, etc.;

[0087] Step 3: Based on the suspected influencing signal, the grinding pressure deviation value and the depth error value at different time points are obtained, and the first synchronization value YT and the second synchronization value ET are obtained based on the processing and analysis of the grinding pressure deviation value and the depth error value at different time points. Based on the processing and analysis of the first synchronization value YT and the second synchronization value ET, a change synchronization value BT is obtained, and based on the change synchronization value BT, it is determined whether the grinding pressure deviation affects the thread depth error of the rubber roller;

[0088] During the abnormal overlap period, the grinding pressure deviation value and the depth error value at different time points are obtained and marked in the XY two-dimensional coordinate system to obtain the depth error change curve, where the X-axis represents the grinding pressure deviation value and the Y-axis represents the depth error value;

[0089] The grinding pressure deviation value is obtained by performing difference processing between the grinding pressure value and the preset grinding pressure value and taking the absolute value;

[0090] Divide the depth error variation curve into a number of sub-curves with equal horizontal lengths;

[0091] Get the slope value of the sub-curve;

[0092] If the slope value is greater than 0, the subcurve is marked as a growth subcurve;

[0093] If the slope value is less than or equal to 0, the sub-curve is marked as a non-increasing sub-curve;

[0094] Count the number of growth sub-curves and perform ratio processing with the total number of sub-curves to obtain the first synchronization value YT;

[0095] Integrate the slope values ​​of all growth sub-curves into a slope value data group, obtain the variance of the slope value data group, and obtain a second synchronization value ET;

[0096] The first synchronization value YT and the second synchronization value ET are processed by the formula: The change synchronization value is calculated, wherein b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0;

[0097] Compare the change synchronization value to the change synchronization threshold:

[0098] If the change synchronization value is greater than the change synchronization threshold, it means that the grinding pressure has a great influence on the thread depth error of the rubber roller;

[0099] If the change synchronization value is less than or equal to the change synchronization threshold, it means that the grinding pressure has little effect on the thread depth error of the rubber roller, and further analysis is needed on other influencing factors, including but not limited to the grinding path, the wear degree of the turning tool, etc.;

[0100] Step 4: If the grinding pressure deviation has a great influence on the thread depth error of the rubber roller, a grinding pressure deviation adjustment value is obtained, and the current grinding pressure deviation is adjusted based on the grinding pressure deviation adjustment value;

[0101] Specifically, all slope values ​​included in the slope value data group are summed to obtain the adjustment coefficient;

[0102] The current depth error value is obtained, and the value is compared with the adjustment coefficient to obtain the grinding pressure deviation adjustment value;

[0103] Adjust the grinding pressure deviation of the current rubber roller based on the grinding pressure deviation adjustment value;

[0104] The technical solution of the embodiment of the present invention is mainly as follows: based on the analysis signal, the grinding pressure data of the rubber roller thread surface during the grinding time is obtained, wherein the grinding pressure data includes the grinding pressure value, based on the processing and analysis of the grinding pressure data and the depth data, the overlap influence value CY is obtained, and it is judged whether the grinding pressure is suspected to affect the rubber roller thread depth error, if so, based on the processing and analysis of the grinding pressure deviation value and the depth error value at different time points, a first synchronization value YT and a second synchronization value ET are obtained, based on the processing and analysis of the first synchronization value YT and the second synchronization value ET, a change synchronization value BT is obtained, and it is determined whether the grinding pressure deviation affects the rubber roller thread depth error, if so, a grinding pressure deviation adjustment value is obtained, and the grinding pressure deviation is adjusted based on the grinding pressure deviation adjustment value. The present invention analyzes the grinding pressure on the rubber roller thread, and then judges whether the rubber roller thread depth deviation is abnormal due to the different grinding pressure deviations on the rubber roller thread surface, and if the grinding pressure deviation affects the abnormal rubber roller thread depth deviation, the current grinding pressure deviation is adjusted.

[0105] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for online correction of rubber roller threads based on machine vision, characterized in that: include: Step 1: Process and analyze the depth error value of the thread surface of the rubber roller to obtain a depth error warning value YJ. If the depth error warning value is greater than the depth error warning threshold, an analysis signal is generated; Step 2: Based on the analysis signal, the grinding pressure value and depth error value of the thread surface of the rubber roller are processed and analyzed to obtain the overlap influence value CY. If the overlap influence value CY is greater than the overlap influence threshold, a suspected influence signal is generated; Step 3: Based on the suspected influencing signal, the grinding deviation pressure value and the depth error value of the rubber roller thread surface at different time points are processed and analyzed to obtain the change synchronization value BT. If the change synchronization value is greater than the change synchronization threshold, it is determined that the grinding pressure deviation has a great influence on the rubber roller thread depth error; Step 4: If the grinding pressure deviation has a great influence on the thread depth error of the rubber roller, a grinding pressure deviation adjustment value is obtained, and the current grinding pressure deviation is adjusted based on the grinding pressure deviation adjustment value.

2. The method for online correction of rubber roller thread based on machine vision according to claim 1 is characterized in that: The depth error warning value YJ is obtained as follows: Obtain the depth error value of the thread surface of the rubber roller, and analyze to obtain the analysis continuity ratio SJ and the depth approach value YC; The continuous analysis ratio SJ and the depth error approximation value YC are processed by the formula: The depth error warning value YJ is obtained, where s1 and s2 are both preset proportional coefficients.

3. The method for online correction of rubber roller thread based on machine vision according to claim 2 is characterized in that: The method for obtaining the analysis continuity ratio SJ is as follows: The depth error value is compared with the depth error standard value and the depth error reference value respectively. If the depth error value is less than or equal to the depth error reference value, a depth normal signal is generated. If the depth error value is greater than the depth error reference value and the depth error value is less than the depth error standard value, a depth abnormal signal is generated. The duration of the deep abnormal signal is obtained and marked as the analysis duration. The analysis duration is ratioed with the duration corresponding to the monitoring period to obtain the analysis duration ratio, which is marked as SJ.

4. The method for online correction of rubber roller thread based on machine vision according to claim 2 is characterized in that: The depth error approximation value YC is obtained as follows: The time period corresponding to the continuous occurrence time of the depth anomaly signal is marked as the continuous period, and the continuous period is divided into several time nodes with equal time intervals. The depth error value corresponding to each time node is obtained, and the sum and average are taken to obtain the depth error mean. The depth error mean is subtracted from the depth error standard value, and the absolute value of the difference is taken to obtain the depth deviation value. The depth deviation value is ratioed with the depth error standard value to obtain the depth error approximation value, which is marked as YC.

5. The method for online correction of rubber roller thread based on machine vision according to claim 1 is characterized in that: The overlap influence value CY is obtained as follows: Obtain the grinding pressure value of the thread surface of the rubber roller during the grinding time, determine the normal pressure time and the abnormal pressure time, overlap the analysis duration with the abnormal pressure time, and obtain the abnormal overlap period ratio YH; The analysis duration is overlapped with the normal pressure time to obtain the normal overlap period ratio EH; The data of abnormal overlap period ratio YH and normal overlap period ratio EH are processed by the formula: The overlap influence value CY is calculated, where a1 and a2 are both preset proportional coefficients, and both a1 and a2 are greater than 0.

6. The method for online correction of rubber roller threads based on machine vision according to claim 5 is characterized in that: The abnormal overlap period ratio YH is obtained in the following manner: Compare the grinding pressure value of the rubber roller thread surface during the grinding time with the preset grinding pressure value: If the grinding pressure value is not equal to the preset grinding pressure value, a pressure abnormality signal is generated; Obtain the duration of the pressure abnormality signal, and mark it as the pressure abnormality time; The analysis duration is overlapped with the pressure abnormality time, specifically: The time period that overlaps between the analysis duration and the pressure abnormality time is extracted and marked as the abnormal overlap period. The duration corresponding to the abnormal overlap period is processed by ratio with the analysis duration to obtain the abnormal overlap period proportion YH.

7. The method for online correction of rubber roller threads based on machine vision according to claim 5 is characterized in that: The normal overlap period ratio EH is obtained as follows: If the grinding pressure value is equal to the preset grinding pressure value, a normal pressure signal is generated; Obtain the duration of the normal pressure signal, and mark it as the normal pressure time; The analysis duration is overlapped with the normal pressure time, specifically: The time period that overlaps between the analysis duration and the normal pressure time is extracted and marked as the normal overlap period. The duration corresponding to the normal overlap period is ratio-processed with the analysis duration to obtain the normal overlap period ratio EH.

8. The method for online correction of rubber roller threads based on machine vision according to claim 1 is characterized in that: The change synchronization value BT is obtained in the following manner: Constructing a depth error variation curve within the abnormal overlap period, and determining a first synchronization value YT and a second synchronization value ET according to the depth error variation curve; The first synchronization value and the second synchronization value are processed by the formula: The change synchronization value is calculated, wherein b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0.

9. The method for online correction of rubber roller threads based on machine vision according to claim 8 is characterized in that: The first synchronization value YT and the second synchronization value ET are obtained in the following manner: During the abnormal overlap period, the grinding pressure deviation value and the depth error value at different time points are obtained and marked in the XY two-dimensional coordinate system to obtain the depth error change curve, where the X-axis represents the grinding pressure deviation value and the Y-axis represents the depth error value; The grinding pressure deviation value is obtained by performing difference processing between the grinding pressure value and the preset grinding pressure value and taking the absolute value; Divide the depth error variation curve into a number of sub-curves with equal horizontal lengths; Get the slope value of the sub-curve; If the slope value is greater than 0, the subcurve is marked as a growth subcurve; If the slope value is less than or equal to 0, the sub-curve is marked as a non-increasing sub-curve; Count the number of growth sub-curves and perform ratio processing with the total number of sub-curves to obtain the first synchronization value YT; The slope values ​​of all growth sub-curves are integrated into a slope value data group, and the variance of the slope value data group is obtained to obtain a second synchronization value ET;.

10. The method for online correction of rubber roller threads based on machine vision according to claim 1, characterized in that: The grinding pressure deviation adjustment value is obtained as follows: All slope values ​​contained in the slope value data group are summed to obtain an adjustment coefficient; The current depth error value is obtained and compared with the adjustment coefficient to obtain the grinding pressure deviation adjustment value.

Images