A Data Analysis-Based Detection and Authentication Method

Abstract

The present invention relates to the technical field of detection and certification, and specifically to a detection and certification method based on data analysis. The detection and certification method specifically comprises the following steps: receiving a verification request for a product through a detection and certification unit, extracting product-related information based on the verification request, and marking it as product information, wherein the product information includes product ID data, production date, maintenance data, and product parts data. The present invention defines the possibility of product damage by the time of production of the product, and performs component quality inspection based on the operating damage of the internal parts of the product. The inspection results are combined with the maintenance status of the product after leaving the factory, so as to comprehensively analyze and calculate the fault status of the product, judge the quality of the product based on the fault status, and thus judge whether the detection and certification is passed, thereby increasing the accuracy of data analysis, improving the efficiency of detection and certification, and increasing the monitoring of product quality.

Description

Technical Field

[0001] This invention relates to the field of testing and certification technology, specifically to a testing and certification method based on data analysis. Background Technology

[0002] Testing is the use of specified methods to inspect and test the technical performance indicators of an object. It is applicable to quality assessment in various industries. Certification refers to the certification by a certification body that a product, service, or management system conforms to relevant technical specifications.

[0003] Currently, some testing and certification organizations use a relatively intelligent testing device to scan the equipment data for quality inspection when testing and certifying certain mechanical structures. However, this approach can only analyze the operating parameters of the mechanical equipment and cannot analyze the damage of internal components. Furthermore, it cannot combine data from various aspects for comprehensive analysis, resulting in inaccurate analysis results.

[0004] Therefore, we propose a detection and authentication method based on data analysis. Summary of the Invention

[0005] The purpose of this invention is to provide a detection and authentication method based on data analysis.

[0006] The objective of this invention can be achieved through the following technical solution: a detection and authentication method based on data analysis, which specifically includes the following steps:

[0007] Step 1: The security verification unit verifies the user's account. It matches the account and password data entered by the customer during login with the user data and password data stored in the cloud server to verify the security of the account and password. After successful verification, the user is redirected to the detection and authentication unit.

[0008] Step 2: Receive the product verification request through the testing and certification unit, extract relevant product information based on the verification request, and label it as product information. Product information includes product ID data, manufacturing date, maintenance data, and product parts data. Product parts data includes the product's component number and the image data corresponding to the component number. Maintenance data includes the maintenance start time, maintenance end time, and replacement parts.

[0009] Step 3: Perform part replacement analysis on the product by combining the data obtained in Step 2 with the original parts number and original parts image data stored in the cloud server, and generate the corresponding part not replaced signal or part replaced signal.

[0010] Step 4: Based on whether the parts have been replaced, conduct internal parts wear analysis to obtain damage signals and damaged part numbers;

[0011] Step 5: Analyze the product's maintenance status using maintenance data. If no records are found in the maintenance data, it is determined that the product has not undergone maintenance, generating a no-maintenance signal. Calculate the difference between the maintenance start time and the manufacturing date in the maintenance data to obtain several initial maintenance interval differences. Select the smallest initial maintenance interval difference and mark it as the initial maintenance interval difference. Calculate the difference between two adjacent maintenance start times and maintenance end times to obtain several maintenance durations. Calculate the difference between the next maintenance start time and the previous maintenance end time to obtain several maintenance interval durations. Based on the maintenance duration and maintenance interval durations, perform data processing to obtain fault change assignments. Maintenance interval serial number;

[0012] Step Six: Extract signals for long factory interval, normal factory interval, no part replacement, part replacement, slight wear, severe wear, and no maintenance. When all signals for normal factory interval, no part replacement, slight wear, and no maintenance are detected simultaneously, the product is deemed to be of high quality, and a test certification pass signal is generated.

[0013] Step 7: If the following signals cannot be detected simultaneously: normal factory interval signal, no part replacement signal, slight wear signal, and no maintenance signal, then the product is determined to be abnormal. An abnormal signal is generated, and a fault evaluation value is calculated based on the abnormal signal. ;

[0014] Step 8: Calculate the fault evaluation value Compare with the fault evaluation threshold MN, when When the value is ≥MN, the product is judged to have a serious fault and low product quality, and an authentication error signal is generated. When the value is less than MN, the product is considered to have few faults and normal product quality. A certification pass signal is generated, and the certification error signal or certification pass signal is transmitted to the detection and certification unit for display and a prompt signal is issued.

[0015] Furthermore, the input account data is matched with the user data. If the two match, it is determined that the input account data exists and a password verification signal is generated. Otherwise, it is determined that the input account does not exist and a re-login signal is generated.

[0016] The system identifies password verification signals and re-login signals. When a re-login signal is detected, the user is automatically redirected to the login page. When a password verification signal is detected, the system extracts the corresponding input password data and matches it with the user's password data. If the matching results are inconsistent, the password is determined to be incorrect, and a password error signal is generated. If the matching results are consistent, the account password is determined to be correct, and a password correct signal is generated.

[0017] The system identifies both correct and incorrect password signals. When a correct password signal is detected, the user's login password is determined to be correct, a redirection signal is generated, and the user is automatically redirected to the detection and authentication unit based on the redirection signal. When an incorrect password signal is detected, the user's login password is determined to be incorrect, and the user is automatically redirected to the login interface to log in with their account and password again.

[0018] Furthermore, if the user has not been redirected to the detection and authentication platform within a set time period, a security protection signal is generated. Based on this signal, the user's account security is protected. The specific protection process involves detecting and sorting password characters that do not match the entered password data, and marking them as such. The values ​​of i and j are both positive integers. Let $\frac{i}{j}$ represent the password errors in the i-th verification of the j-th order, and $\frac{j}{j}$ be one or more sets, representing the set of password errors entered by the user at the current time. The value of n is a positive integer;

[0019] Extract and mark instances where the same user enters an incorrect password before successful login each time they log in within the cloud server. The value of 'a' is a positive integer. This represents the user's password being incorrect during the i-th login attempt (password in the j-th order), based on the customer's login settings record set. It identifies and labels the probability of each character being entered incorrectly each time a user logs in and enters their password. ;

[0020] Match the incorrect passwords in error set A with the probability of each character being incorrect, find the error probability of each character in error set A, sum the error probabilities of each character in error set A, and calculate the conversion evaluation coefficient of the user's incorrect password input at the current time. The specific calculation formula is as follows: ,in, This represents the evaluation conversion factor corresponding to each password character;

[0021] Transformation evaluation coefficient Compare with the conversion evaluation threshold FY, when the conversion evaluation coefficient If the conversion evaluation coefficient is less than the conversion evaluation threshold FY, the user's input error probability is considered to be within a safe range, and the account is considered safe. When the conversion evaluation coefficient equals the conversion evaluation threshold FY, it is determined that the probability of user input errors poses a security risk, and the account verification is abnormal. If the error rate exceeds the conversion evaluation threshold FY, the user's input error probability is determined to be outside the safe range, and the account is at risk of being stolen.

[0022] Transformation evaluation coefficient Substituting the conversion evaluation threshold FY into the formula for calculating the conversion evaluation coefficient, the number of verifications corresponding to the password verification is derived in reverse. After the number of verifications is completed, the user is redirected to the lock screen, which prohibits login of the account password.

[0023] Furthermore, based on the chronological order, the difference between two adjacent repair times is calculated. A Cartesian coordinate system is established, and the repair time differences are marked according to the chronological order. Connecting each adjacent coordinate point with a straight line forms a repair time difference graph. When the repair time difference graph shows an upward trend, it is determined that the product's repair time is gradually increasing, and the faults are gradually becoming more severe, generating a fault growth trend signal. When the repair time difference graph shows a downward trend, it is determined that the product's repair time is gradually decreasing, and the faults are gradually decreasing, generating a fault decline trend signal. When the repair time difference graph shows a fluctuating trend, it is determined that the product's repair time is constantly fluctuating, and the faults are uncontrollable, generating a fault fluctuation trend signal. The fault growth trend signal, fault decline trend signal, and fault fluctuation trend signal are then uniformly converted, assigned values, and tagged as fault change values. r=1, 2, 3, when r=1, This represents the assignment of a fault growth trend signal; when r=2, This represents the assignment of the fault decline trend signal; when r=3, This represents the assignment of a fault fluctuation trend signal;

[0024] Several maintenance intervals are sorted according to time sequence, and the sequence corresponding to the maintenance interval duration is marked as the maintenance interval sequence number.

[0025] Furthermore, the impact of abnormal signals on long and normal factory interval signals is assigned and labeled as the factory interval impact factor. The value of e is 1 or 2. When e = 1, This represents the influence factor corresponding to the long factory interval signal. When e=2, This is represented as the influencing factor corresponding to the normal signal of the factory exit interval, and the factory exit interval time difference is also extracted.

[0026] The number of damaged component numbers corresponding to the damage signals is counted and marked as the damage count value.

[0027] Furthermore, based on the preset fault evaluation calculation formula:

[0028]

[0029] Calculate the fault evaluation value ,in, The values ​​are: CJ (factory delivery interval impact factor), Pg (time difference between factory delivery intervals), and Pg (number of damaged items). The transformation influence factor is represented by the number of damaged values, and Wx represents the difference in initial maintenance intervals. This is represented as the transformation factor of the initial maintenance interval difference. This represents the assignment of values ​​for fault changes. This represents the transformation impact factor assigned to fault changes. This represents the maintenance interval duration, where c takes the value of a positive integer. This is represented as the maintenance interval sequence number. It is represented as the deviation correction factor for the fault evaluation value.

[0030] The beneficial effects of this invention are:

[0031] (1) This invention verifies the existence of a user's account by comparing the user's account before authentication. It also performs character matching on the password corresponding to the user's account and calculates the character error rate of the incorrect password entered by the user after character matching, thereby locking the account and increasing account security;

[0032] (2) This invention defines the potential damage of a product by the product’s manufacturing time, and conducts component quality testing based on the operational damage of the product’s internal parts. The test results are combined with the product’s maintenance status after it leaves the factory, thereby comprehensively analyzing and calculating the product’s fault status. The quality of the product is determined based on the fault status, thereby determining whether it has passed the testing and certification. This increases the accuracy of data analysis, improves the efficiency of testing and certification, and enhances the monitoring of product quality. Detailed Implementation

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0034] This invention relates to a detection and certification method based on data analysis, which specifically includes the following steps:

[0035] The security verification unit performs security verification on the user's account and collects relevant information entered by the customer during login in real time, and marks it as input information. Input information includes input account data and input password data. Input account data refers to the account ID entered by the user on the login interface, and input password data refers to the password corresponding to the account ID entered by the user on the login interface.

[0036] Retrieve user information stored in the cloud server. User information includes user data and user password data. User data refers to the user's account ID registered on the Internet, and user password data refers to the login password corresponding to the user's account ID registered on the Internet.

[0037] The system matches the entered account data with the user data. If the two match, the system determines that the entered account data exists and generates a password verification signal. Otherwise, the system determines that the entered account does not exist and generates a re-login signal.

[0038] The system identifies password verification signals and re-login signals. When a re-login signal is detected, the user is automatically redirected to the login page. When a password verification signal is detected, the system extracts the corresponding input password data and user password data and matches them. If the matching results are inconsistent, the password is determined to be incorrect, and a password error signal is generated. If the matching results are consistent, the account password is determined to be correct, and a password correct signal is generated.

[0039] The system identifies both password error and password correct signals. When a password correct signal is detected, the user's login password is determined to be correct, a redirection signal is generated, and the user is automatically redirected to the detection and authentication unit based on the redirection signal. When a password error signal is detected, the user's login password is determined to be incorrect, and the user is automatically redirected to the login interface to log in with their account and password again.

[0040] If the user has not been redirected to the authentication platform within a set time period, a security protection signal is generated. Based on this signal, the user's account security is protected. The specific protection process involves detecting and sorting password characters that do not match the entered password data, and marking them as such. The values ​​of i and j are both positive integers. Let $\frac{i}{j}$ represent the password errors in the i-th verification of the j-th order, and $\frac{j}{j}$ be one or more sets, representing the set of password errors entered by the user at the current time. The value of n is a positive integer;

[0041] Extract and mark instances where the same user enters an incorrect password before successful login each time they log in within the cloud server. The value of 'a' is a positive integer. This represents the user's password being incorrect during the i-th login attempt (password in the j-th order), based on the customer's login settings record set. It identifies and labels the probability of each character being entered incorrectly each time a user logs in and enters their password. ;

[0042] Match the incorrect passwords in error set A with the probability of each character being incorrect, find the error probability of each character in error set A, sum the error probabilities of each character in error set A, and calculate the conversion evaluation coefficient of the user's incorrect password input at the current time. The specific calculation formula is as follows: ,in, This represents the evaluation conversion factor corresponding to each password character;

[0043] Transformation evaluation coefficient Compare with the conversion evaluation threshold FY, when the conversion evaluation coefficient If the conversion evaluation coefficient is less than the conversion evaluation threshold FY, the user's input error probability is considered to be within a safe range, and the account is considered safe. When the conversion evaluation coefficient equals the conversion evaluation threshold FY, it is determined that the probability of user input errors poses a security risk, and the account verification is abnormal. If the error rate exceeds the conversion evaluation threshold FY, the user's input error probability is determined to be outside the safe range, and the account is at risk of being stolen.

[0044] Transformation evaluation coefficient Substituting the conversion evaluation threshold FY into the formula for calculating the conversion evaluation coefficient, the number of verifications corresponding to the password verification is derived in reverse. After the number of verifications is completed, the user is redirected to the lock screen, which prevents login to the account password.

[0045] The testing and certification unit receives product verification requests and extracts relevant product information based on the verification requests, which is then labeled as product information. Product information includes product ID data, manufacturing date, repair data, and product parts data. Product parts data includes the product's component number and the corresponding image data. Repair data includes the repair start time, repair end time, and repair replacement parts. Among these, repair replacement parts refer to the component numbers replaced during repair, product ID data refers to the product's manufacturing number, and manufacturing date refers to the manufacturing time recorded by the merchant.

[0046] Extract the corresponding manufacturing date based on the product ID data, calculate the difference between the manufacturing date and the current detection time. If the calculated manufacturing interval time difference is greater than the manufacturing interval threshold, a long manufacturing interval signal is generated. If the calculated manufacturing interval time difference is less than or equal to the manufacturing interval threshold, a normal manufacturing interval signal is generated.

[0047] The cloud server stores the original factory part numbers and original factory part image data corresponding to the product ID data transmitted by the manufacturer's terminal. The part number corresponding to the product ID data is matched with the original factory part number. When the matching result is consistent, it is determined that the part inside the product has not been replaced, and a part not replaced signal is generated. When the matching result is inconsistent, it is determined that the part inside the product has been replaced, and a part replaced signal is generated.

[0048] Extract and identify signals indicating whether a component has been replaced or not. If a replacement signal is detected, identify the replaced component number and mark it as such. If a non-replacement signal is detected, extract the corresponding image data and original equipment manufacturer (OEM) component image data, and perform component wear analysis. Specifically:

[0049] The image data and corresponding original parts image data are processed into three-dimensional imaging. Overlapping imaging is performed in a virtual rectangular coordinate system with the center of the parts as the base point. The distance between each edge coordinate of the same part and the center point of the parts after overlapping imaging is calculated. The distance data corresponding to the original parts image data and the distance data corresponding to the image data are calculated to obtain several edge difference values. These edge difference values ​​are compared with the wear distance threshold. If the edge difference value is greater than or equal to the wear distance threshold, the parts are judged to be severely worn and a severe wear signal is generated. If the edge difference value is less than the wear distance threshold, the parts are judged to be slightly worn and a slight wear signal is generated. The number of severe wear signals is counted. When the number of severe wear signals is greater than or equal to the safe wear number threshold, the parts are judged to be damaged and a damage signal is generated. Each part number is identified and marked as the damaged part number based on the damage signal.

[0050] Extract the maintenance data corresponding to the product ID data. If there is no record in the maintenance data, it is determined that the product has not been maintained and a no-maintenance signal is generated. Calculate the difference between the maintenance start time and the manufacturing date in the maintenance data to calculate several initial maintenance interval differences. Select the smallest initial maintenance interval difference and mark it as the initial maintenance interval difference.

[0051] The difference between two adjacent maintenance start time points and maintenance end time points is calculated to obtain several maintenance durations. The difference between the maintenance start time point of the next time sorted and the maintenance end time point of the previous time is calculated to obtain several maintenance interval durations.

[0052] The time difference between two adjacent repair times is calculated based on their chronological order. A Cartesian coordinate system is established, and the time differences are marked on the coordinates according to their chronological order. Connecting each adjacent coordinate point with a straight line forms a time difference graph. When the time difference graph shows an upward trend, it indicates that the product's repair time is gradually increasing and the faults are becoming more severe, generating a fault growth trend signal. When the time difference graph shows a downward trend, it indicates that the product's repair time is gradually decreasing and the faults are becoming less severe, generating a fault decline trend signal. When the time difference graph shows a fluctuating trend, it indicates that the product's repair time is constantly fluctuating and the faults are uncontrollable, generating a fault fluctuation trend signal. The fault growth trend signal, fault decline trend signal, and fault fluctuation trend signal are then uniformly converted, assigned values, and tagged as fault change values. r=1, 2, 3, when r=1, This represents the assignment of a fault growth trend signal; when r=2, This represents the assignment of the fault decline trend signal; when r=3, This represents the assignment of a fault fluctuation trend signal;

[0053] The maintenance interval durations are sorted according to time sequence, and the sequence corresponding to the maintenance interval durations is marked as the maintenance interval sequence number;

[0054] Extract signals for long factory interval, normal factory interval, no part replacement, part replacement, slight wear, severe wear, and no maintenance. When all three signals are simultaneously detected, the product is deemed to be of high quality, and a test certification pass signal is generated. If none of these signals are simultaneously detected, the product is deemed to be abnormal, and an abnormal signal is generated.

[0055] The impact of abnormal signals on long and normal factory interval signals is assigned and labeled as the factory interval impact factor. The value of e is 1 or 2. When e = 1, This represents the influence factor corresponding to the long factory interval signal. When e=2, This is represented as the influencing factor corresponding to the normal signal of the factory exit interval, and the factory exit interval time difference is also extracted.

[0056] Count the number of damaged parts corresponding to the damage signals and mark them as the damage count;

[0057] Based on the preset fault evaluation calculation formula:

[0058]

[0059] Calculate the fault evaluation value ,in, The values ​​are: CJ (factory delivery interval impact factor), Pg (time difference between factory delivery intervals), and Pg (number of damaged items). The transformation influence factor is represented by the number of damaged values, and Wx represents the difference in initial maintenance intervals. This is represented as the transformation factor of the initial maintenance interval difference. This represents the assignment of values ​​for fault changes. This represents the transformation impact factor assigned to fault changes. This represents the maintenance interval duration, where c takes the value of a positive integer. This is represented as the maintenance interval sequence number. This is expressed as a deviation correction factor for the fault evaluation value;

[0060] Fault evaluation value Compare with the fault evaluation threshold MN, when the fault evaluation value When the fault evaluation value is greater than or equal to the fault evaluation threshold MN, the product is judged to have a serious fault and low product quality, and an authentication error signal is generated. When the value is less than the fault evaluation threshold MN, the product is judged to have few faults, the product quality is normal, and a certification pass signal is generated.

[0061] The authentication error signal or authentication pass signal is transmitted to the detection and authentication unit for display and a prompt signal is issued.

[0062] This invention verifies the existence of user accounts by comparing them with those before testing and certification. It also performs character matching on the passwords corresponding to the user accounts and calculates the character error rate for incorrect passwords entered after character matching, thereby locking the account and increasing account security. Furthermore, it identifies potential product damage by considering the product's manufacturing date and conducts component quality testing based on the operational damage status of internal parts. The test results are combined with post-manufacturing repair records to comprehensively analyze and calculate the product's fault conditions. Based on these fault conditions, the product's quality is determined, thus deciding whether it passes testing and certification. This increases the accuracy of data analysis, improves the efficiency of testing and certification, and enhances product quality monitoring.

[0063] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A detection and authentication method based on data analysis, characterized in that, The testing and certification method specifically includes the following steps: Step 1: The security verification unit verifies the user's account. It matches the account and password data entered by the customer during login with the user data and password data stored in the cloud server to verify the security of the account and password. After successful verification, the user is redirected to the detection and authentication unit. Step 2: Receive the product verification request through the testing and certification unit, extract relevant product information based on the verification request, and label it as product information. Product information includes product ID data, manufacturing date, maintenance data, and product parts data. Product parts data includes the product's component number and the image data corresponding to the component number. Maintenance data includes the maintenance start time, maintenance end time, and replacement parts. Step 3: Based on the manufacturing date, part number, and corresponding image data, perform data processing on internal parts to obtain the manufacturing interval time difference, long manufacturing interval signal, part not replaced signal, normal manufacturing interval signal, damage signal, and damaged part number. Based on the product's maintenance data, the interval between the first maintenance time and the manufacturing date is calculated and marked as the initial maintenance interval difference. Based on the maintenance start time and maintenance end time, the maintenance duration and the interval between each maintenance are calculated and marked as the maintenance interval duration. The difference between two adjacent maintenance times is calculated based on the chronological order. A Cartesian coordinate system is set up, and the maintenance time difference is marked on the coordinates according to the chronological order. Each adjacent coordinate point is connected by a straight line to form a maintenance time difference graph. When the maintenance time difference graph shows an upward trend, a fault growth trend signal is generated. When the maintenance time difference graph shows a downward trend, a fault decline trend signal is generated. When the maintenance time difference graph shows a fluctuating trend, a fault fluctuation trend signal is generated. The fault growth trend signal, fault decline trend signal, and fault fluctuation trend signal are uniformly transformed, assigned values, and calibrated as fault change values. r=1, 2, 3; The maintenance interval durations are sorted according to time sequence, and the sequence corresponding to the maintenance interval durations is marked as the maintenance interval sequence number; Step 4: Perform preliminary verification and fault evaluation calculations on the data results analyzed in Step 3, and calculate the fault evaluation value. Fault evaluation value Compare with the fault evaluation threshold MN; when When the value is ≥MN, the product is judged to have a serious fault and low product quality, and an authentication error signal is generated. When the value is less than MN, the product is judged to have few faults and normal product quality, and a certification pass signal is generated. The specific process of fault evaluation calculation is as follows: The long factory interval signal and the normal factory interval signal are assigned and labeled as the factory interval influence factor. The value of e is 1 or 2. Count the number of damaged parts corresponding to the damage signals and mark them as the damage count; Based on the preset fault evaluation calculation formula: Calculate the fault evaluation value ,in, The output is represented by the factory delivery interval impact factor, CJ represents the factory delivery interval time difference, and Pg represents the number of damaged items. The transformation influence factor is represented by the number of damaged values, and Wx represents the difference in initial maintenance intervals. This is expressed as the transformation influencing factor of the initial maintenance interval difference. This represents the assignment of values ​​for fault changes. This represents the transformation impact factor assigned to fault changes. This represents the maintenance interval duration, where c takes the value of a positive integer. This is represented as the maintenance interval sequence number. This is expressed as a deviation correction factor for the fault evaluation value; Step 5: Transmit the authentication error signal or authentication pass signal to the detection and authentication unit for display and issue a prompt signal.

2. The detection and authentication method based on data analysis according to claim 1, characterized in that, The cloud server stores the original factory part numbers corresponding to the product ID data transmitted from the manufacturer's terminal, as well as the original factory part image data.

3. The detection and authentication method based on data analysis according to claim 2, characterized in that, The data processing procedure for internal parts is as follows: Based on the manufacturing date and the current testing time, and combined with the difference calculation formula, the manufacturing interval time difference is calculated. When the manufacturing interval time difference is greater than the manufacturing interval threshold and less than or equal to the manufacturing interval threshold, a long manufacturing interval signal and a normal manufacturing interval signal are generated respectively. The part number is matched with the original part number. If the result is consistent, a part not replaced signal is generated, and the replaced part number is identified and marked as a replaced part. If the result is inconsistent, a part replaced signal is generated, and the corresponding image data and original part image data are extracted to perform part wear analysis to generate severe wear signal, slight wear signal, damage signal and damaged part number. The image data and the corresponding original parts image data are processed into three-dimensional imaging. Overlapping imaging is performed in a virtual space rectangular coordinate system with the center of the parts as the base point. The distance between the coordinates of each edge of the same part of the same overlapping imaging and the coordinates of the center point of the parts is calculated. The distance data corresponding to the original parts image data is calculated by comparing the distance data with the distance data corresponding to the image data. Several edge differences are compared with the wear distance threshold. If they are greater than or equal to the wear distance threshold, a severe wear signal is generated. If they are less than the wear distance threshold, a slight wear signal is generated. If the number of severe wear signals is greater than or equal to the safe wear number threshold, a damage signal is generated. Each part number is identified and marked as the damaged part number based on the damage signal. Extract signals for long factory interval, normal factory interval, no part replacement, part replacement, slight wear, severe wear, and no maintenance. When all three signals are detected simultaneously, a pass signal is generated. When none of these signals are detected simultaneously, an abnormal signal is generated. Fault evaluation calculations are performed based on abnormal signals to calculate fault evaluation values. .

Images