Production quality verification method of porcelain bottle binding device based on image analysis

Through the method based on image analysis, the damage of the porcelain bottle tying device is comprehensively evaluated, which solves the problem of inability to evaluate the actual use of the product in the prior art, and achieves rapid feedback and production optimization, improving detection efficiency and maintenance efficiency.

CN118674717BActive Publication Date: 2024-11-12HONGGUANG ELECTRIC GROUP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411154824.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-12
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The existing porcelain bottle tying device production quality verification method is limited to a single detection indicator, and it is impossible to comprehensively evaluate the damage of the product in actual use. It lacks a rapid feedback and optimization mechanism, which makes it difficult to quickly solve the problem.

Method used

Using an image analysis-based method, by collecting and analyzing the image information of the porcelain bottle tying device, calculating the damage assessment index of the wires and porcelain bottles, evaluating production quality, and triggering corresponding signaling for optimization and adjustment based on the evaluation results.

Benefits of technology

The comprehensive quality evaluation of the porcelain bottle tying device has been achieved, the detection efficiency and accuracy have been improved, and the problems arise in production can be quickly feedbacked and solved, the production process is optimized, and the maintenance work efficiency has been improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118674717B_ABST
    Figure CN118674717B_ABST
Patent Text Reader

Abstract

The invention discloses a production quality verification method for a porcelain bottle binder based on image analysis, and relates to the technical field of production quality verification. The invention not only improves the detection efficiency and accuracy through automated image acquisition and analysis, but also can comprehensively evaluate the use quality of the product, including the damage of the wire and the porcelain bottle, and further analyzes and quantifies and evaluates the degree of damage to the product, and optimizes and adjusts the analysis and production process based on the evaluation results, thereby solving the problem in the prior art that only the geometric dimensions of the product can be detected, but the damage of the product in actual use cannot be evaluated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of production quality verification, and in particular to a production quality verification method for a porcelain bottle binding device based on image analysis. Background Art

[0002] The porcelain bottle tie is a special tool used to fix and bind porcelain bottles in power systems. The wires are fixed to the porcelain bottles through the tie to ensure a firm connection between the wires and the porcelain bottles, reducing electrical failures caused by poor connections. Therefore, after the production of the porcelain bottle tie is completed, the quality of the porcelain bottle tie needs to be checked to ensure product quality and safety.

[0003] Traditional production quality inspection methods mainly rely on manual inspection, which has many limitations. With the development of technology, intelligent quality inspection has gradually become a trend. However, the production quality inspection methods of porcelain bottle binding devices in the existing technology still have the following shortcomings:

[0004] They are often limited to a single testing indicator and lack the ability to conduct a comprehensive assessment of the product. For example, they are often only able to detect the geometric dimensions of the product but are unable to assess the damage to the product during actual use. In addition, when quality problems are detected, there is a lack of effective feedback and optimization mechanisms, which results in the problem being unable to be resolved quickly.

[0005] Therefore, a production quality verification method for porcelain bottle binders based on image analysis is proposed. Summary of the invention

[0006] In view of this, the present invention provides a method for verifying the production quality of porcelain bottle binders based on image analysis to solve the problems raised by the above-mentioned background technology.

[0007] The purpose of the present invention can be achieved by the following technical solution: A method for verifying the production quality of a porcelain bottle binding device based on image analysis, comprising:

[0008] Step 1: Obtain the number of porcelain bottle binders produced in the current batch, and represent the number as X. Analyze the number X of porcelain bottle binders produced in the current batch, and obtain the number of samples drawn from the current batch of porcelain bottle binders. The number of samples drawn is used as the inspection and evaluation number of porcelain bottle binders produced in the current batch. The inspection and evaluation number is represented by a number i, where i=1,2,......,k, and k is the total number of inspection and evaluation numbers.

[0009] Step 2: Based on the inspection and evaluation quantity of the porcelain bottle binders produced in the current batch, randomly select the porcelain bottle binders of the corresponding inspection and evaluation quantity from the porcelain bottle binders produced in the current batch as inspection samples, and test the inspection samples to simulate the compression action in the actual use process, and collect image information for analysis, thereby obtaining the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each inspection sample;

[0010] Step 3: Based on the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, the production quality score of each test sample is obtained;

[0011] Step 4: preset a reference threshold score for the production quality score, extract the production quality score of each test sample and compare it with the preset reference threshold score, and evaluate the overall quality of the porcelain bottle binders produced in the current batch based on the comparison results;

[0012] Step 5: Based on the overall quality assessment results of the current batch of porcelain bottle binders, the assessment results include the overall assessment results being qualified and the existence of samples with unqualified quality. Execute corresponding steps according to the overall quality assessment results of the current batch of porcelain bottle binders.

[0013] In some embodiments, the test sample is tested, specifically:

[0014] Collect the initial image information of the application wire and porcelain bottle, and operate the extracted test samples to simulate the compression action during actual use;

[0015] After a single clamping action is completed, the operation is stopped, and the time point of the stop operation is marked as a single analysis time point, and the image information of the wire and the porcelain bottle at the single analysis time point is collected and integrated into a single image data packet; the collected image information is respectively the image information of the contact area between the wire and the clamping block of the porcelain bottle binding device, and the image information of the contact area between the porcelain bottle and the left and right clamping arms;

[0016] After the acquisition is completed, continue to operate the extracted test sample to repeat the compacting action P times; P>3; stop the operation after the P compacting actions are completed, mark the time point of stopping the operation as the repeated analysis time point, collect the image information of the wire and the porcelain bottle at the repeated analysis time endpoint and integrate them into a repeated image data packet.

[0017] In some embodiments, the wire damage assessment index Sai of each test sample is obtained, specifically:

[0018] Extract the single image data package of each test sample, analyze the image information of the wire in the single image data package corresponding to each test sample, highlight the boundary of the wire surface in the image information, and segment the wire surface from the background to identify the indentation damage on the wire surface;

[0019] According to the recognition result, the position of the indentation on the conductor surface is located as the indentation area of ​​the conductor surface, and the number of pixels in the indentation area is calculated. The actual size is converted based on the resolution of the image to obtain the indentation estimation value Hai corresponding to the conductor image information of each test sample;

[0020] Before the test, feature points are marked on the original image of the test wire, and the position coordinates of the feature points of the wire before and after a single compression action are determined; for each pair of feature points corresponding to each other before and after compression, the displacement vector between them is obtained, and the size of the displacement vector is calculated according to the obtained displacement vector;

[0021] The displacement sizes of all feature points are sorted from large to small, and the maximum displacement among all feature points is determined to obtain the displacement estimation Hbi of the wire image information corresponding to each test sample;

[0022] Similarly, the repeated image data packets of each test sample are extracted, and the image information of the wire in the repeated image data packets corresponding to each test sample is analyzed to obtain the total indentation value Vai and the total displacement value Vbi of the wire image information corresponding to each test sample;

[0023] Substitute the indentation estimate Hai and displacement estimate Hbi corresponding to the single image data packet of each test sample, and the total indentation value Vai and total displacement value Vbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the wire damage assessment index Sai of each test sample; as well as They represent the highest allowable reference values ​​set for the indentation estimate Hai, displacement estimate Hbi, indentation total value Vai and displacement total value Vbi respectively; as well as They are the influencing weight factors of the indentation estimate Hai, displacement estimate Hbi, the total indentation value Vai and the total displacement value Vbi;

[0024] In some embodiments, the porcelain bottle damage assessment index Sbi of each test sample is obtained, specifically:

[0025] Extracting a single image data packet of each test sample, analyzing the image information of the porcelain bottle in the single image data packet corresponding to each test sample, identifying scratch damage and notch damage on the surface of the porcelain bottle, and locating the position of the scratch damage and notch damage in the porcelain bottle as the scratch area and notch area of ​​the porcelain bottle according to the identification result;

[0026] Calculate the number of pixels in the scratch area and the notch area, convert the actual size based on the image resolution, and obtain the scratch estimation Wai and notch estimation Wbi of the porcelain bottle image information corresponding to each test sample;

[0027] Similarly, the repeated image data packets of each test sample are extracted, and the image information of the porcelain bottle in the repeated image data packets corresponding to each test sample is analyzed to obtain the total scratch value Cai and the total notch value Cbi of the porcelain bottle image information corresponding to each test sample;

[0028] Substitute the scratch estimate Wai and gap estimate Wbi corresponding to the single image data packet of each test sample, and the total scratch value Cai and total gap value Cbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the porcelain bottle damage assessment index Sbi of each test sample; as well as They represent the highest allowable reference values ​​set for the scratch estimate Wai, the notch estimate Wbi, the total scratch value Cai and the total notch value Cbi respectively; as well as They are the influencing weight factors of scratch valuation Wai, notch valuation Wbi, total scratch value Cai and total notch value Cbi.

[0029] In some embodiments, the production quality score of each test sample is obtained, specifically:

[0030] Preset the index value ranges of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi respectively, and set each index value range of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi to correspond to a conductor damage score and a porcelain bottle damage score respectively;

[0031] The wire damage assessment index Sai and porcelain bottle damage assessment index Sbi of each test sample are extracted and matched with the corresponding index value ranges of each group to obtain the wire damage score and porcelain bottle damage score of each test sample, and the two groups of scores of each test sample are added together to obtain the production quality score of each test sample.

[0032] In some embodiments, the overall quality of the porcelain bottle binders produced in the current batch is evaluated based on the comparison results, specifically:

[0033] A reference threshold score for the production quality score is preset, and the production quality score of each test sample is compared with the preset reference threshold score. If the production quality score of one or more groups of test samples is greater than the preset reference threshold score, the test samples with scores greater than the reference threshold score are marked as unqualified samples;

[0034] The number of samples with unqualified quality is counted as the unqualified number, and the proportion of the unqualified number in the test evaluation number k is calculated as the unqualified proportion;

[0035] If the production quality scores of all tested samples are less than the preset reference threshold scores, a re-inspection signal is triggered and sent to the manager's mobile terminal. After receiving the triggered re-inspection signal and confirming it, the manager performs steps 2 and 3 again. If the manager receives the triggered re-inspection signal and denies it, the current batch of porcelain bottle binders will be marked as a batch with overall qualified quality.

[0036] In some embodiments, if the overall quality evaluation result of the current batch of porcelain bottle binders is qualified, the specific execution steps are:

[0037] If the overall quality assessment results show that the current batch of porcelain bottle binders is a qualified batch as a whole, the information in each test sample and the corresponding image will be transmitted to the verification workshop for further review by the technical staff in the verification workshop. If the review results show that there are samples with unqualified quality, the optimization signal will be triggered, and the management personnel will be notified to check the collection and analysis process of step two, and optimize step two based on the results of the inspection; the optimization includes adjusting the parameters of the camera during the image collection process, and optimizing the parameters and influencing weight factors during the analysis process; after the optimization is completed, step two will be executed again.

[0038] In some embodiments, if the overall quality assessment result of the current batch of porcelain bottle binders is that there are samples with unqualified quality, the specific execution steps are:

[0039] If the overall quality assessment results show that there are samples with unqualified quality, the corresponding unqualified ratio will be extracted, and the unqualified ratio will be used as the defective ratio of the current batch of porcelain bottle binders. The unqualified quality samples in each test sample will be transported to the inspection workshop, and further reviewed by the technical staff in the inspection workshop. If the review result is correct, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time. If the review result is incorrect, the management personnel will be notified to check the collection and analysis process of step two first, and step two will be optimized based on the inspection results, and the unqualified quality samples will be corrected based on the review results. If there are still unqualified quality samples after the correction is completed, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time.

[0040] After the maintenance signal is triggered, the technicians in the inspection workshop will search for the reasons for the failure of the samples. After the search is completed, the search results will be sent to the mobile terminal of the manager. The reasons for failure include material quality and equipment abnormality.

[0041] If the search result shows that the reason for failure is equipment abnormality, the production line corresponding to the current batch of porcelain bottle binders will be stopped, and the outlier value will be selected. The largest maintenance personnel go to the location of the corresponding production line to check and maintain the equipment on the production line.

[0042] In some embodiments, the different values ​​of each maintenance personnel are obtained. The specific process is:

[0043] With the location of the production line corresponding to the current production of porcelain bottle binding devices as the center of the circle, set the distance as the radius to draw a circle, and select all maintenance personnel within the circle as candidates;

[0044] Obtain the current location of each candidate, and calculate the distance between the location of the production line and the location of the production line as the distance value of each candidate, and record it as Lc;

[0045] Further obtaining the work log of each candidate, and extracting the working years and total maintenance times of each candidate from the work log of each candidate;

[0046] The working years of each candidate are marked as Le; at the same time, the duration of each maintenance is extracted from the total number of maintenance times of each candidate; the duration is calculated from the time when the personnel arrives at the location as the starting time point, and the time point when the investigation and maintenance are completed is the end time point. The time difference between the starting time point and the end time point is calculated to obtain the duration of personnel maintenance;

[0047] The average time taken by each candidate is calculated to obtain the average maintenance time of each candidate, which is recorded as Lm.

[0048] According to the formula , weighted calculation is performed on the distance value Lc, working years Le and average maintenance time Lm of each candidate to obtain the difference value of each candidate ;in as well as They are the influencing weight factors of distance value Lc, working years Le and average maintenance time Lm respectively.

[0049] In some embodiments, the number X of porcelain bottle binders produced in the current batch is analyzed to obtain the number of samples of porcelain bottle binders produced in the current batch, specifically:

[0050] Preset the quantity value ranges of X, set each group of quantity value ranges to correspond to a sample quantity, match the quantity X of the current batch of porcelain bottle binders produced with the preset quantity value ranges, and obtain the sample quantity of the current batch of porcelain bottle binders produced.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention not only improves the detection efficiency and accuracy through automated image acquisition and analysis, but also can comprehensively evaluate the use quality of the product, including the damage of the wire and the porcelain bottle, and further analyzes and quantifies and evaluates the degree of damage to the product, and optimizes and adjusts the analysis and production process according to the evaluation results, thus solving the problem that the prior art can only detect the geometric dimensions of the product but cannot evaluate the damage of the product in actual use;

[0053] When a production quality problem is detected, the present invention can quickly feedback and trigger corresponding signaling to promptly solve the problem in production. At the same time, by calculating the outlier value of the maintenance personnel, the most suitable maintenance personnel can be reasonably selected, thereby improving the efficiency of the maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0055] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION

[0056] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0057] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0058] See also Figure 1 As shown, the production quality verification method of porcelain bottle binding device based on image analysis includes:

[0059] Step 1: Obtain the number of porcelain bottle binders produced in the current batch, represent the number with X, preset the value ranges of each number of X, set each group of value ranges of the number to correspond to a number of samples, match the number X of porcelain bottle binders produced in the current batch with the preset value ranges of each number, obtain the number of samples of porcelain bottle binders produced in the current batch, use the number of samples as the inspection and evaluation number of porcelain bottle binders produced in the current batch, and represent the inspection and evaluation number with number i, where i=1,2,......,k, and k is the total number of inspection and evaluation numbers;

[0060] Step 2: Based on the inspection and evaluation quantity of the porcelain bottle binders produced in the current batch, randomly select the porcelain bottle binders of the corresponding inspection and evaluation quantity from the porcelain bottle binders produced in the current batch as inspection samples, and test the inspection samples to simulate the compression action in the actual use process, and collect image information for analysis, thereby obtaining the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each inspection sample;

[0061] Specifically:

[0062] Prepare the tools needed for testing, including standard-sized wires and porcelain bottles, and the quantity should match the samples to be tested;

[0063] The initial image information of the application wire and the porcelain bottle is collected by the image sensor, and the extracted test samples are operated to simulate the pressing action in the actual use process;

[0064] After a single clamping action is completed, the operation is stopped, and the time point of the stop operation is marked as a single analysis time point, and the image information of the wire and the porcelain bottle at the single analysis time point is collected and integrated into a single image data packet; the collected image information is respectively the image information of the contact area between the wire and the clamping block of the porcelain bottle binding device, and the image information of the contact area between the porcelain bottle and the left and right clamping arms;

[0065] After the collection is completed, continue to operate the extracted test sample to repeat the compaction action for P times; P>3, P is the number of repetitions set by the management personnel, and the specific value is set by the management personnel; after the P times of compaction action are completed, stop the operation, mark the time point of stopping the operation as the repeated analysis time point, collect the image information of the wire and the porcelain bottle at the end point of the repeated analysis time, and integrate them into a repeated image data packet;

[0066] Analyze the single image data package and repeated image data package of each test sample to obtain the production quality score of each test sample;

[0067] Extract the single image data package of each test sample, analyze the image information of the wire in the single image data package corresponding to each test sample, highlight the boundary of the wire surface in the image information, and segment the wire surface from the background to identify the indentation damage on the wire surface; this is achieved through a trained classifier;

[0068] According to the recognition result, the position of the indentation on the conductor surface is located as the indentation area of ​​the conductor surface, and the number of pixels in the indentation area is calculated. The actual size is converted based on the resolution of the image to obtain the indentation estimation value Hai corresponding to the conductor image information of each test sample;

[0069] Before the test, feature points are marked on the original image of the test wire, and the position coordinates of the feature points of the wire before and after a single compression action are determined; for each pair of feature points corresponding to each other before and after compression, the displacement vector between them is obtained, and the size of the displacement vector is calculated according to the obtained displacement vector; that is, the calculation is performed using the Euclidean distance formula of the displacement;

[0070] The displacement sizes of all feature points are sorted from large to small, and the maximum displacement among all feature points is determined to obtain the displacement estimation Hbi of the wire image information corresponding to each test sample;

[0071] Similarly, the repeated image data packets of each test sample are extracted, and the image information of the wire in the repeated image data packets corresponding to each test sample is analyzed to obtain the total indentation value Vai and the total displacement value Vbi of the wire image information corresponding to each test sample;

[0072] Substitute the indentation estimate Hai and displacement estimate Hbi corresponding to the single image data packet of each test sample, and the total indentation value Vai and total displacement value Vbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the wire damage assessment index Sai of each test sample; as well as They represent the highest allowable reference values ​​set for the indentation estimate Hai, displacement estimate Hbi, indentation total value Vai and displacement total value Vbi respectively; as well as They are the influencing weight factors of the indentation estimate Hai, displacement estimate Hbi, the total indentation value Vai and the total displacement value Vbi;

[0073] Taking into account the damage after single and repeated compression, the wire damage assessment index Sai is obtained through weighted calculation, which more comprehensively reflects the quality of the porcelain bottle binding device.

[0074] Extracting a single image data packet of each test sample, analyzing the image information of the porcelain bottle in the single image data packet corresponding to each test sample, identifying scratch damage and notch damage on the surface of the porcelain bottle, and locating the position of the scratch damage and notch damage in the porcelain bottle as the scratch area and notch area of ​​the porcelain bottle according to the identification result;

[0075] Calculate the number of pixels in the scratch area and the notch area, convert the actual size based on the image resolution, and obtain the scratch estimation Wai and notch estimation Wbi of the porcelain bottle image information corresponding to each test sample;

[0076] Similarly, the repeated image data packets of each test sample are extracted, and the image information of the porcelain bottle in the repeated image data packets corresponding to each test sample is analyzed to obtain the total scratch value Cai and the total notch value Cbi of the porcelain bottle image information corresponding to each test sample;

[0077] Substitute the scratch estimate Wai and gap estimate Wbi corresponding to the single image data packet of each test sample, and the total scratch value Cai and total gap value Cbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the porcelain bottle damage assessment index Sbi of each test sample; as well as They represent the highest allowable reference values ​​set for the scratch estimate Wai, the notch estimate Wbi, the total scratch value Cai and the total notch value Cbi respectively; as well as They are the impact weight factors of scratch valuation Wai, gap valuation Wbi, total scratch value Cai and total gap value Cbi;

[0078] The above analysis process requires image preprocessing, which includes operations such as image denoising and contrast enhancement to improve image quality and make features easier to identify.

[0079] Preset the index value ranges of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi respectively, and set each index value range of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi to correspond to a conductor damage score and a porcelain bottle damage score respectively;

[0080] Extract the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, and match them with the corresponding index value ranges of each group, to obtain the wire damage score and porcelain bottle damage score of each test sample, and add up the two groups of scores of each test sample to obtain the production quality score of each test sample;

[0081] Step 3: Based on the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, the production quality score of each test sample is obtained;

[0082] Specifically:

[0083] Preset the index value ranges of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi respectively, and set each index value range of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi to correspond to a conductor damage score and a porcelain bottle damage score respectively;

[0084] Extract the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, and match them with the corresponding index value ranges of each group, to obtain the wire damage score and porcelain bottle damage score of each test sample, and add up the two groups of scores of each test sample to obtain the production quality score of each test sample;

[0085] Step 4: preset a reference threshold score for the production quality score, extract the production quality score of each test sample and compare it with the preset reference threshold score, and evaluate the overall quality of the porcelain bottle binders produced in the current batch based on the comparison results;

[0086] Specifically:

[0087] According to the quality requirements of the current batch of porcelain bottle binders, a reference threshold score for the production quality score is preset, and the production quality score of each test sample is compared with the preset reference threshold score. If the production quality score of one or more groups of test samples is greater than the preset reference threshold score, the test samples with scores greater than the reference threshold score are marked as unqualified samples;

[0088] The number of samples with unqualified quality is counted as the unqualified number, and the proportion of the unqualified number in the test evaluation number k is calculated as the unqualified proportion;

[0089] If the production quality scores of all tested samples are less than the preset reference threshold scores, a re-inspection signal is triggered and sent to the manager's mobile terminal. After receiving and confirming the triggered re-inspection signal, the manager performs steps 2 and 3 again. If the manager denies the triggered re-inspection signal, the current batch of porcelain bottle binders is marked as a batch with overall qualified quality.

[0090] It should be noted that the administrator can also actively trigger the re-check signaling and execute step 2 and step 3 again when the re-check signaling is not triggered. Step 2 and step 3 can be executed multiple times, which is specifically set by the administrator.

[0091] Step 5: Based on the overall quality evaluation results of the current batch of porcelain bottle binders, the evaluation results include the overall evaluation results being qualified and the existence of samples with unqualified quality, and corresponding steps are performed according to the overall quality evaluation results of the current batch of porcelain bottle binders;

[0092] Specifically:

[0093] If the overall quality assessment result shows that the current batch of porcelain bottle binders is an overall qualified batch, the corresponding image information of each test sample will be transmitted to the verification workshop, and further reviewed by the technical personnel in the verification workshop. If the review result shows that there are samples with unqualified quality, the optimization signaling will be triggered, and the management personnel will be notified to check the collection and analysis process of step 2, and optimize step 2 based on the inspection results; the optimization includes adjusting the parameters of the camera during the image collection process, and optimizing the parameters and influencing weight factors during the analysis process; after the optimization is completed, step 2 will be executed again;

[0094] If the overall quality assessment results show that there are samples with unqualified quality, the corresponding unqualified ratio will be extracted, and the unqualified ratio will be used as the defective ratio of the current batch of porcelain bottle binders. The unqualified quality samples in each test sample will be transported to the inspection workshop, and further reviewed by the technical staff in the inspection workshop. If the review result is correct, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time. If the review result is incorrect, the management personnel will be notified to check the collection and analysis process of step two first, and step two will be optimized based on the inspection results, and the unqualified quality samples will be corrected based on the review results. If there are still unqualified quality samples after the correction is completed, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time.

[0095] After the maintenance signal is triggered, the technicians in the inspection workshop will search for the reasons for the failure of the samples. After the search is completed, the search results will be sent to the mobile terminal of the manager. The reasons for failure include material quality and equipment abnormality.

[0096] Technicians can quickly retrieve the reasons for failure by checking the components of the porcelain bottle tie-down device and comparing them with qualified samples to find differences;

[0097] If the search result shows that the reason for failure is equipment abnormality, the production line corresponding to the current batch of porcelain bottle binders will be stopped, and the outlier value will be selected. The largest maintenance personnel go to the location of the corresponding production line to check and maintain the equipment on the production line;

[0098] Get the different values ​​of each maintenance personnel The specific process is:

[0099] With the location of the production line corresponding to the current production of porcelain bottle binding devices as the center of the circle, set the distance as the radius to draw a circle, and select all maintenance personnel within the circle as candidates;

[0100] Obtain the current location of each candidate, and calculate the distance between the location of the production line and the location of the production line as the distance value of each candidate, and record it as Lc;

[0101] Further obtaining the work log of each candidate, and extracting the working years and total maintenance times of each candidate from the work log of each candidate;

[0102] The working years of each candidate are marked as Le; at the same time, the duration of each maintenance is extracted from the total number of maintenance times of each candidate; the duration is calculated from the time when the personnel arrives at the location as the starting time point, and the time point when the investigation and maintenance are completed is the end time point. The time difference between the starting time point and the end time point is calculated to obtain the duration of personnel maintenance;

[0103] The average time taken by each candidate is calculated to obtain the average maintenance time of each candidate, which is recorded as Lm.

[0104] According to the formula , weighted calculation is performed on the distance value Lc, working years Le and average maintenance time Lm of each candidate to obtain the difference value of each candidate ;in as well as They are the influencing weight factors of the distance value Lc, the service life Le and the average maintenance time Lm;

[0105] The above formulas are obtained by collecting a large amount of data and performing software simulation, and a formula close to the actual value is selected. The influencing weight factor and specific coefficient value in the formula are set by technical personnel in this field according to actual conditions, and can be adjusted and modified later.

[0106] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. Quality verification method for porcelain bottle tying device production based on image analysis, It is characterized in that include: Step 1: Obtain the number of porcelain bottle binders produced in the current batch, and represent the number as X. Analyze the number X of porcelain bottle binders produced in the current batch, and obtain the number of samples drawn from the current batch of porcelain bottle binders. The number of samples drawn is used as the inspection and evaluation number of porcelain bottle binders produced in the current batch. The inspection and evaluation number is represented by a number i, where i=1,2,......,k, and k is the total number of inspection and evaluation numbers. Step 2: Based on the inspection and evaluation quantity of the porcelain bottle binders produced in the current batch, randomly select the porcelain bottle binders of the corresponding inspection and evaluation quantity from the porcelain bottle binders produced in the current batch as inspection samples, and test the inspection samples to simulate the compression action in the actual use process, and collect image information for analysis, thereby obtaining the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each inspection sample; The wire damage assessment index Sai of each test sample is obtained, specifically: Extract the single image data package of each test sample, analyze the image information of the wire in the single image data package corresponding to each test sample, highlight the boundary of the wire surface in the image information, and segment the wire surface from the background to identify the indentation damage on the wire surface; According to the recognition result, the position of the indentation on the conductor surface is located as the indentation area of ​​the conductor surface, and the number of pixels in the indentation area is calculated. The actual size is converted based on the resolution of the image to obtain the indentation estimation value Hai corresponding to the conductor image information of each test sample; Before the test, feature points are marked on the original image of the test wire, and the position coordinates of the feature points of the wire before and after a single compression action are determined; for each pair of feature points corresponding to each other before and after compression, the displacement vector between them is obtained, and the size of the displacement vector is calculated according to the obtained displacement vector; The displacement sizes of all feature points are sorted from large to small, and the maximum displacement among all feature points is determined to obtain the displacement estimation Hbi of the wire image information corresponding to each test sample; Similarly, the repeated image data packets of each test sample are extracted, and the image information of the wire in the repeated image data packets corresponding to each test sample is analyzed to obtain the total indentation value Vai and the total displacement value Vbi of the wire image information corresponding to each test sample; Substitute the indentation estimate Hai and displacement estimate Hbi corresponding to the single image data packet of each test sample, and the total indentation value Vai and total displacement value Vbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the wire damage assessment index Sai of each test sample; as well as They represent the highest allowable reference values ​​set for the indentation estimate Hai, displacement estimate Hbi, indentation total value Vai and displacement total value Vbi respectively; as well as They are the influencing weight factors of the indentation estimate Hai, displacement estimate Hbi, the total indentation value Vai and the total displacement value Vbi; The porcelain bottle damage assessment index Sbi of each test sample is obtained, specifically: Extracting a single image data packet of each test sample, analyzing the image information of the porcelain bottle in the single image data packet corresponding to each test sample, identifying scratch damage and notch damage on the surface of the porcelain bottle, and locating the position of the scratch damage and notch damage in the porcelain bottle as the scratch area and notch area of ​​the porcelain bottle according to the identification result; Calculate the number of pixels in the scratch area and the notch area, convert the actual size based on the image resolution, and obtain the scratch estimation Wai and notch estimation Wbi of the porcelain bottle image information corresponding to each test sample; Similarly, the repeated image data packets of each test sample are extracted, and the image information of the porcelain bottle in the repeated image data packets corresponding to each test sample is analyzed to obtain the total scratch value Cai and the total notch value Cbi of the porcelain bottle image information corresponding to each test sample; Substitute the scratch estimate Wai and gap estimate Wbi corresponding to the single image data packet of each test sample, and the total scratch value Cai and total gap value Cbi corresponding to the repeated image data packet of each test sample into the formula , weighted calculation is performed to obtain the porcelain bottle damage assessment index Sbi of each test sample; as well as They represent the highest allowable reference values ​​set for the scratch estimate Wai, the notch estimate Wbi, the total scratch value Cai and the total notch value Cbi respectively; as well as They are the impact weight factors of scratch valuation Wai, gap valuation Wbi, total scratch value Cai and total gap value Cbi; Step 3: Based on the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, the production quality score of each test sample is obtained; The production quality score of each test sample is obtained, specifically: Preset the index value ranges of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi respectively, and set each index value range of the conductor damage assessment index Sai and the porcelain bottle damage assessment index Sbi to correspond to a conductor damage score and a porcelain bottle damage score respectively; Extract the wire damage assessment index Sai and the porcelain bottle damage assessment index Sbi of each test sample, and match them with the corresponding index value ranges of each group, to obtain the wire damage score and porcelain bottle damage score of each test sample, and add up the two groups of scores of each test sample to obtain the production quality score of each test sample; Step 4: preset a reference threshold score for the production quality score, extract the production quality score of each test sample and compare it with the preset reference threshold score, and evaluate the overall quality of the porcelain bottle binders produced in the current batch based on the comparison results; Step 5: Based on the overall quality assessment results of the current batch of porcelain bottle binders, the assessment results include the overall assessment results being qualified and the existence of samples with unqualified quality. Execute corresponding steps according to the overall quality assessment results of the current batch of porcelain bottle binders.

2. According to the method for verifying the production quality of porcelain bottle tying device based on image analysis in claim 1, It is characterized in that The test samples are tested as follows: Collect the initial image information of the application wire and porcelain bottle, and operate the extracted test samples to simulate the compression action during actual use; After a single clamping action is completed, the operation is stopped, and the time point of the stop operation is marked as a single analysis time point, and the image information of the wire and the porcelain bottle at the single analysis time point is collected and integrated into a single image data packet; the collected image information is respectively the image information of the contact area between the wire and the clamping block of the porcelain bottle binding device, and the image information of the contact area between the porcelain bottle and the left and right clamping arms; After the acquisition is completed, continue to operate the extracted test sample to repeat the compacting action P times; P>3; stop the operation after the P compacting actions are completed, mark the time point of stopping the operation as the repeated analysis time point, collect the image information of the wire and the porcelain bottle at the repeated analysis time endpoint and integrate them into a repeated image data packet.

3. According to the method for verifying the production quality of porcelain bottle tying device based on image analysis in claim 1, It is characterized in that Based on the comparison results, the overall quality of the current batch of porcelain bottle binders is evaluated, specifically: The production quality score of each test sample is compared with the preset reference threshold score. If the production quality score of one or more groups of test samples is greater than the preset reference threshold score, the test samples with scores greater than the reference threshold score are marked as unqualified samples; The number of samples with unqualified quality is counted as the unqualified number, and the proportion of the unqualified number in the test evaluation number k is calculated as the unqualified proportion; If the production quality scores of all tested samples are less than the preset reference threshold scores, a re-inspection signal is triggered and sent to the manager's mobile terminal. After receiving the triggered re-inspection signal and confirming it, the manager performs steps 2 and 3 again. If the manager receives the triggered re-inspection signal and denies it, the current batch of porcelain bottle binders will be marked as a batch with overall qualified quality.

4. According to the method for verifying the production quality of porcelain bottle tying device based on image analysis in claim 1, It is characterized in that If the overall quality assessment result of the current batch of porcelain bottle binders is qualified, the specific implementation steps are as follows: If the overall quality assessment result shows that the current batch of porcelain bottle binders is an overall qualified batch, the corresponding image information of each test sample will be transmitted to the verification workshop, and further reviewed by the technical staff in the verification workshop. If the review result shows that there are unqualified samples, the optimization signaling will be triggered, and the management personnel will be notified to check the collection and analysis process of step two, and optimize step two based on the inspection results; the optimization includes adjusting the parameters of the camera during the image collection process, and optimizing the parameters and influencing weight factors during the analysis process; After the optimization is completed, execute step 2 again.

5. According to the method for verifying the production quality of porcelain bottle tying device based on image analysis in claim 1, It is characterized in that If the overall quality assessment result of the current batch of porcelain bottle binders is that there are samples with unqualified quality, the specific implementation steps are as follows: If the overall quality assessment results show that there are samples with unqualified quality, the corresponding unqualified ratio will be extracted, and the unqualified ratio will be used as the defective ratio of the current batch of porcelain bottle binders. The unqualified quality samples in each test sample will be transported to the inspection workshop, and further reviewed by the technical staff in the inspection workshop. If the review result is correct, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time. If the review result is incorrect, the management personnel will be notified to check the collection and analysis process of step two first, and step two will be optimized based on the inspection results, and the unqualified quality samples will be corrected based on the review results. If there are still unqualified quality samples after the correction is completed, the current batch of porcelain bottle binders will be inspected as a whole, and the maintenance signaling will be triggered at the same time. After the maintenance signal is triggered, the technicians in the inspection workshop will search for the reasons for the failure of the samples. After the search is completed, the search results will be sent to the mobile terminal of the manager. The reasons for failure include material quality and equipment abnormality. If the search result shows that the reason for failure is equipment abnormality, the production line corresponding to the current batch of porcelain bottle binders will be stopped, and the outlier value will be selected. The largest maintenance personnel go to the location of the corresponding production line to check and maintain the equipment on the production line.

6. According to the method for verifying the production quality of porcelain bottle tying devices based on image analysis according to claim 5, It is characterized in that Get the different values ​​of each maintenance personnel The specific process is: With the location of the production line corresponding to the current production of porcelain bottle binding devices as the center of the circle, set the distance as the radius to draw a circle, and select all maintenance personnel within the circle as candidates; Obtain the current location of each candidate, and calculate the distance between the location of the production line and the location of the production line as the distance value of each candidate, and record it as Lc; Further obtaining the work log of each candidate, and extracting the working years and total maintenance times of each candidate from the work log of each candidate; The working years of each candidate are marked as Le; at the same time, the duration of each maintenance is extracted from the total number of maintenance times of each candidate; The duration is calculated from the time when the personnel arrive at the location as the starting time point, and the time when the troubleshooting and maintenance are completed as the ending time point. The time difference between the starting time point and the ending time point is calculated to obtain the duration of personnel maintenance; The average time taken by each candidate is calculated to obtain the average maintenance time of each candidate, which is recorded as Lm. According to the formula , weighted calculation is performed on the distance value Lc, working years Le and average maintenance time Lm of each candidate to obtain the difference value of each candidate ;in as well as They are the influencing weight factors of distance value Lc, working years Le and average maintenance time Lm respectively.

7. According to the method for verifying the production quality of porcelain bottle tying device based on image analysis in claim 1, It is characterized in that The number X of porcelain bottle binders produced in the current batch is analyzed to obtain the number of samples drawn from the current batch of porcelain bottle binders, specifically: Preset the quantity value ranges of X, set each group of quantity value ranges to correspond to a sample quantity, match the quantity X of the current batch of porcelain bottle binders produced with the preset quantity value ranges, and obtain the sample quantity of the current batch of porcelain bottle binders produced.