Dynamic tracking method for missing automobile parts

Abstract

The invention discloses a dynamic tracking method for automobile part missing, and relates to the technical field of data tracking, and the method comprises the steps: 1, identity information is confirmed, part information is obtained, the part information comprises an identification code and a maintenance record, the identification code is obtained, whether the identification code is activated or not is judged, and if the identification code is not activated, the maintenance record is not activated; if yes, accessory activation is directly carried out, and identity binding is carried out; if the identification code is activated, the maintenance record is uploaded or downloaded, and the accessory information is output; step 2, by obtaining accessory information, when an accessory bound with the identity information is missing, missing accessory tracking is carried out; marking the missing accessory, and when the information of the missing accessory is uploaded, locking the uploading position of the information of the accessory; when the missing accessory information is not uploaded, at least five times of maintenance records before accessory missing are obtained, and the relation between the maintenance records and accessory missing is monitored by analyzing the maintenance records.

Description

Technical Field

[0001] The present invention relates to the technical field of data tracking, and in particular to a dynamic tracking method for missing automobile parts. Background Art

[0002] RFID (Radio Frequency Identification) technology is an advanced automatic identification technology that uses wireless radio frequency to perform non-contact two-way data transmission to achieve automatic identification of objects and information exchange. With the advancement of technology and the reduction of costs, RFID technology has begun to be widely used in supply chain management and inventory control, especially in the automotive industry. Its application scope has extended from the production stage to after-sales maintenance services, providing comprehensive support for the entire life cycle of automotive parts.

[0003] Although RFID technology has shown great potential in improving the efficiency of auto parts tracking, there are still some limitations in practical applications. First, although traditional RFID tags can provide real-time location information, when it comes to complex tracking processes, such as the need to record detailed historical maintenance records and usage of each part, the existing system is often too cumbersome and lacks a unified and efficient parts tracking management system. This leads to scattered information and difficulty in forming a complete parts archive, making it inconvenient for users to view parts information. Secondly, in order to ensure the effectiveness of parts tracking, the cooperation of maintenance personnel and the supervision of car owners are usually required. This process not only increases the workload, but also may affect the accuracy of tracking results due to human factors. In addition, due to the lack of transparency, it is difficult for car owners to have a clear understanding of the true condition of the parts. Especially during the parts replacement process, criminals may take advantage of this and replace the original parts by replacing the labels, thereby engaging in illegal reselling activities. This situation not only harms the rights and interests of consumers, but also creates an unfair competition environment for regular parts suppliers.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a dynamic tracking method for missing auto parts to solve the technical defects mentioned above. The present invention establishes a corresponding parts tracking system and binds it with the parts information. When the parts are missing, the system can be reported in time to track the parts, and the maintenance records are saved at the same time. The abnormal conditions in the maintenance content are alerted. When the corresponding operation is performed at the maintenance point, the owner is notified through the mobile terminal. At the same time, a convolutional neural network is used to extract the features of the image, and automobile image recognition and classification are performed. Regression analysis is used to establish the relationship between the loss of parts and the maintenance records, and a loss coefficient is generated; the expected service life is calculated by combining the service life of the auto parts in the database and the loss coefficient, and the expected service life is compared and judged with the replacement time of the parts in the relevant area. When the expected service life is equal to the replacement time of the parts in the relevant area, a warning is issued to the relevant area or store for reselling parts.

[0006] The purpose of the present invention can be achieved by the following technical solution: A dynamic tracking method for missing automobile parts, comprising the following steps: Step 1: Obtaining the accessory information of the automobile accessory and confirming the user identity information, the accessory information includes an identification code and a maintenance record, obtaining the activation state of the identification code and performing a pre-configured operation according to the activation state; wherein, if the activation state of the identification code is inactive, directly activating the accessory and binding the user identity; if the activation state of the identification code is activated, uploading or downloading the maintenance record and outputting the accessory information; Step 2: When the accessory bound to the user identity information is missing, the missing accessory is tracked according to the accessory information; the missing accessory tracking includes: marking the missing accessory, and if the accessory information of the missing accessory is uploaded, locking the upload location of the accessory information of the missing accessory; if the accessory information of the missing accessory is not uploaded, obtaining at least five maintenance records of the missing accessory before the missing accessory is lost, associating the maintenance records with multiple groups of background user data obtained in advance, and combining the relationship between the maintenance records and the missing accessories, establishing a maintenance record and missing accessories feedback mechanism, and outputting the accessory information of the missing accessories; Step 3: Obtain the automobile image obtained by pre-scanning, and establish a model relationship between the loss and maintenance record of the automobile accessories in combination with the accessory information, and generate a loss coefficient based on the model relationship; Step 4: Calculate the estimated service life of the auto parts based on the recorded service life of the auto parts in the database and the loss coefficient, compare the estimated service life with the replacement time of the missing parts in the relevant area, and determine if the estimated service life of the missing parts is equal to the replacement time, triggering an alert mechanism.

[0007] Preferably, in step 1, confirming the user identity information includes: First, the account information provided by the user is verified and compared with the pre-stored user database. After successful identity authentication, access rights are assigned.

[0008] Preferably, in step 1, obtaining the activation state of the identification code and performing a pre-configured operation according to the activation state includes: Obtain the unique identification code on the missing accessory, read the identification code, and upload the read data; perform preliminary verification on the uploaded identification code, and after the data verification passes, query the activation status of the identification code, find the record corresponding to the uploaded identification code, and check the activation status field in the database record; According to the query results, determine whether the identification code has been activated: When the status field of the identification code shows activated, a corresponding message is returned to show that the identification code has been activated; When the status field of the identification code shows inactive, a corresponding message is returned, indicating that the identification code has not been activated, and the accessory is bound to the identity.

[0009] Preferably, in the step 2, if the accessory information of the missing accessory is uploaded, locking the upload location of the accessory information of the missing accessory includes: When it is confirmed that an accessory is missing, the missing accessory will be marked as "missing" in the system. By logging into the identity account bound to the missing accessory, the missing accessory will be marked through the identity account and the status of the accessory will be updated in the database. If the missing accessory information is obtained again, the geographic location of the missing accessory will be captured.

[0010] Preferably, in step 2, analyzing the maintenance records and monitoring the relationship between the maintenance records and the missing parts, and establishing a maintenance record and missing parts feedback mechanism include: S201: extracting abnormal situation features, including parts maintenance frequency, maintenance type, maintenance location change and maintenance time interval, and pre-processing the abnormal situation features; S202: Extract the five maintenance records of the accessory before the last upload from the database, as well as multiple sets of user data obtained from the background, establish a maintenance record and missing accessories feedback mechanism, and monitor the proportion of abnormal situations in the maintenance records in the missing accessories. The formula is as follows: ;

[0011] Where p is the proportion of abnormal situations when accessories are missing, Ai and Bi are the characteristic vectors of abnormal situations A and B respectively; S203: Calculate the probability of abnormal situations of the five maintenance records before the brake pad is missing, sort the probabilities of the abnormal situations, and alert the top g abnormal situations.

[0012] Preferably, in the step three, obtaining the automobile image obtained by pre-scanning, combining the accessory information, and establishing a model relationship between the wear and tear of the automobile accessories and the maintenance record, including: Scanning the exterior of the car at multiple angles to obtain a car image, wherein the car image includes at least a front view, a side view, and a rear view; Correcting and reducing noise on the car image, and cutting out the background; The pre-trained CNN model is used to extract the features of the car image, and the regression analysis method is used to establish the relationship between the user's parts loss and maintenance records. The linear regression model is used to establish the model relationship between the car parts loss and maintenance records. The formula is: ;

[0013] Where y is the degree of wear and tear of the car; beta0 is the intercept term; beta1, beta2, ..., betan are regression coefficients, indicating the impact of each feature on the degree of wear and tear; x1, x2, ..., xn are maintenance records; and e is the error term.

[0014] Preferably, in the step three, generating the loss coefficient based on the model relationship includes: By obtaining the loss degree of different user groups, combined with the total number of user groups, the current user group and the loss value, the loss coefficient f of the car is calculated according to the formula: ;

[0015] Where, f is the loss coefficient; m is the total number of user groups; i is the number of the current user group; y0, y1, y2...yn are the loss degrees of different user groups; z is the loss value, ranging from 1 <z<1.5。

[0016] Preferably, in the step 4, the estimated service life of the auto parts is calculated by combining the recorded service life of the auto parts in the database and the loss coefficient, including: The service life of the auto parts is taken as the difference between unit 1 and the loss coefficient, and multiplied by the recorded service life of the auto parts in the database to calculate the expected service life of each auto part. The formula is: T=(1-f)*S Where T is the expected service life of the auto parts; S is the recorded service life of the auto parts in the database.

[0017] Preferably, in step 4, if the estimated service life is equal to the replacement time, a warning mechanism is triggered, including: According to the comparison results, the error range is set to ±0.05, and the following judgment is made: If the estimated service life is not equal to the replacement time or the error range is outside ±0.05, the warning mechanism will not be triggered; If the estimated service life is equal to the replacement time or the error range is within ±0.05, the warning mechanism will be triggered, and the reselling of accessories in the relevant area or store will be recorded and output.

[0018] The beneficial effects of the present invention are as follows: (1) The present invention establishes a corresponding parts tracking system and binds it to the parts information. When a part is missing, it can report to the system in time and track the part. At the same time, the maintenance record is saved and an abnormal situation in the maintenance content is warned. When the corresponding operation is performed at the maintenance point, the car owner is notified through the mobile terminal.

[0019] (2) The present invention uses a convolutional neural network to extract image features and perform automobile image recognition and classification. It uses regression analysis to establish the relationship between accessory wear and maintenance records and generate a wear coefficient. It combines the service life of the automobile accessory in the database and the wear coefficient to calculate the expected service life, and compares the expected service life with the replacement time of the accessory in the relevant area and makes a judgment. When the expected service life is equal to the replacement time of the accessory in the relevant area, a warning is issued to the relevant area or store for reselling accessories. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described below in conjunction with the accompanying drawings; Figure 1 It is a reference diagram of the method of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1 As shown, this embodiment provides a dynamic tracking method for missing automobile parts, comprising the following steps: Step 1: Confirm the identity information and obtain the accessory information. The accessory information includes the identification code and maintenance record. By obtaining the identification code, determine whether the identification code is activated. If the identification code is not activated, directly activate the accessory and bind the identity; if the identification code is activated, upload or download the maintenance record and output the accessory information; The identity information confirmation process is as follows: The car owner initiates an identity authentication request through a dedicated application or App. After the system receives the request, it first verifies the account information provided by the user and compares it with the pre-stored user database to confirm the validity of the account. After the identity authentication is successful, the system will further check whether the user has the authority to access specific accessory information; Different user roles have different access rights, such as car owners, maintenance personnel, and managers, which is similar to most common management systems on the market, so I will not go into details here; Identification code acquisition and whether to activate it: Use the camera function of a smartphone or tablet to scan and obtain the unique identification code on the brake pad, read the identification code, and upload the read data to the API interface of the central server or cloud platform; Perform preliminary verification on the uploaded identification code on the server side to ensure the integrity and format correctness of the data, check whether the identification code conforms to the predetermined format, ensure that the identification code is unique in the system and there is no duplication, and verify whether the identification code exists in a known database. After the data verification is passed, query the activation status of the identification code, access the database, find the record corresponding to the uploaded identification code, and check the activation status field in the database record, which usually indicates whether the identification code has been activated; According to the query results, determine whether the identification code has been activated: When the status field of the identification code shows activated, a corresponding message is returned to show that the identification code has been activated; When the status field of the identification code shows that it is not activated, a corresponding message is returned, indicating that the identification code has not been activated, and further operation instructions are provided to bind the accessory to the identity; A new record is created in the database, the identification code is associated with the user identity information, and the activation status field of the identification code is updated to "activated", indicating that the accessory has been bound to the user identity and activated. The server returns the binding and activation results through the API, the application displays the corresponding prompt information on the user interface, and the server notifies the user of the binding and activation results through email or SMS; Step 2: By obtaining the accessory information, when the accessory bound to the identity information is missing, the missing accessory is tracked; the missing accessory is marked, and when the missing accessory information is uploaded, the location where the accessory information is uploaded is locked; when the missing accessory information is not uploaded, at least five maintenance records before the missing accessory is obtained, and the relationship between the maintenance record and the missing accessory is monitored by analyzing the maintenance record, a feedback mechanism between the maintenance record and the missing accessory is established, and the accessory information is output; When missing accessory information is uploaded: When it is confirmed that a component is missing, the component is marked as "missing" in the system; By logging into the identity account associated with the missing part, the missing part is marked through the identity account and the status of the part is updated in the database; if someone tries to rescan and upload the brake pad information, the system will capture the new geographic location, for example, longitude 116.4074, latitude 39.9042; the system automatically sends an alert to the administrator and local police, and provides the last known location information of the part; When the missing accessory information is not uploaded: Extract the brake pad information from the database. When the missing parts information is not uploaded, for example, in the maintenance records of the owners of the missing parts, the proportion of abnormal situations such as frequent cross-provincial maintenance and short maintenance intervals is too high; Extract at least five maintenance records before the last upload of the part from the database, use these maintenance records and multiple sets of user data obtained from the backend for analysis to identify the relationship between the maintenance records and the missing parts, and establish a maintenance record and missing parts feedback mechanism; Record the relationship between missing parts and establish a maintenance record and missing parts feedback mechanism: S201: Extract abnormal situation features, including parts maintenance frequency, maintenance type, maintenance location change and maintenance time interval; perform data cleaning based on abnormal situation features; remove duplicate records: check and delete duplicate maintenance records in the database; handle missing values: for missing data points, you can choose to fill (such as using the mean, median or mode), delete records with missing values, or use interpolation methods; outlier detection: identify and handle outliers, such as impossible maintenance time intervals or unreasonable maintenance frequencies; S202: Extract the five maintenance records of the accessory before the last upload from the database, as well as multiple sets of user data obtained from the background, establish a maintenance record and missing parts feedback mechanism, and monitor the proportion of abnormal situations in the maintenance records to missing parts. The formula is as follows: ;

[0023] In the formula, p (abnormal situation) is the proportion of abnormal situations when accessories are missing, Ai and Bi are the characteristic vectors of abnormal situations A and B respectively; Example: A brake pad is missing. The five maintenance records before the last upload of the brake pad are extracted from the database. The five maintenance records before the missing brake pad are as follows:

[0024] At the same time, the previous five data of brake pad loss of multiple groups of users obtained from the background are obtained, and each information fragment in the maintenance location, maintenance time and maintenance content is obtained; an effective maintenance record and parts missing feedback mechanism is established, and natural language processing technology is used to use text summary algorithms, such as TF-IDF-based methods or deep learning models, to identify and extract the main content of each maintenance. For example, a sequence labeling model can be trained to identify key entities in the maintenance report (such as fault type, replacement parts, etc.), and then a concise maintenance description is generated based on these entities, and a simple SQL aggregation function is used to perform quantitative statistics, such as calculating the number of times a certain type of maintenance operation occurs: ```sql SELECTxlxm,COUNT(*)ASrepair_count FROMQCWX WHEREljbh='Brake pad number' GROUPBYxlxm ORDER BY repair_countDESC; ``` S203: calculating the probability of abnormal situations in the five maintenance records before the brake pad is missing, and ranking the probabilities of the abnormal situations, alerting the top g abnormal situations, and notifying the car owner through the mobile terminal when the maintenance point performs corresponding operations; By conducting an in-depth analysis of the top three abnormal situations and taking corresponding warning measures, the prevention and treatment efficiency of brake pad problems can be significantly improved. After collecting the latest five maintenance records and calculating the probability of abnormal situations, assuming that they are ranked according to probability, the top three abnormal situations are: the brake pad wears too fast, which may be due to driving habits (such as frequent emergency braking), the use of inferior brake pad materials, or the car is overloaded; scratches or damage on the surface of the brake disc, uneven friction between the brake pad and the brake disc, or hard impurities in the brake pad; the presence of air in the brake system leads to poor braking effect, and the air in the brake system may cause insufficient brake fluid pressure, thereby affecting the braking effect; after the maintenance is completed, the owner checks for the missing corresponding accessories; Step 3: By acquiring the accessory information and scanning the car from multiple angles, using convolutional neural networks to extract image features, and performing car image recognition and classification, using regression analysis to establish the relationship between accessory wear and maintenance records, and generate a wear coefficient; combining the service life of the auto accessory in the database and the wear coefficient, calculate the expected service life, and compare and judge the expected service life with the replacement time of the accessory in the relevant area. When the expected service life is equal to the replacement time of the accessory in the relevant area, be alert to the reselling of accessories in the relevant area or store; Capture clear car images from multiple angles, keep them stable during the scanning process, and pre-process the car images. Through feature extraction and image recognition, the extracted image features are combined with maintenance records, service life and other data to form a data set. Use regression analysis methods to establish the relationship between parts loss and maintenance records, and calculate the loss coefficient of each part. The process of multi-angle scanning of a car is as follows: Under good lighting conditions, use a camera to clearly capture the image of the car, use a fixed scanning platform or lifting device to ensure that the car remains stable during the scanning process, obtain comprehensive information about the car's appearance, and scan from multiple angles: Front view: Take at least three images from the front of the car, making sure to cover the entire front; Rear view: Take at least three images from behind the car, making sure to cover the entire rear; Side view: Take at least three images from each side of the car, making sure to cover the entire side; The car image preprocessing process is: Correct the captured car image to remove distortion and noise; improve the contrast and brightness of the car image to make it clearer; cut out unnecessary background and keep only the car part; Take multiple car pictures at different angles and positions, use OpenCV to detect the edges in each car image, calculate the camera's intrinsic matrix and distortion coefficient, use the original image to correct the distortion, select the appropriate filtering method according to the noise type, use the filtering function provided by OpenCV, adjust the brightness through linear transformation or Gamma correction, locate the car through color threshold, edge detection or machine learning methods, determine the cropping boundary based on the car's position, and perform the cropping operation. Cutting off irrelevant background areas can make the car image analysis more focused on the target; The regression analysis process is: The pre-trained CNN model is used to extract the features of the car image, and the regression analysis method is used to establish the relationship between the user's parts loss and maintenance records. The linear regression model is used to establish the relationship between the car loss and maintenance records. The formula is: ;

[0025] In the formula, y is the degree of wear and tear of the car; beta0 is the intercept term; beta1, beta2...betan are regression coefficients, indicating the impact of each feature on the degree of wear and tear; x1, x2,..., xn are maintenance records; e is the error term; The calculation process for the estimated useful life is as follows: By obtaining the loss degree of different user groups, combined with the total number of user groups, the current user group and the loss value, the loss coefficient f of the car is calculated according to the formula: ;

[0026] Where, f is the loss coefficient; m is the total number of user groups; i is the number of the current user group; y0, y1, y2...yn are the loss degrees of different user groups; z is the loss value, ranging from 1 <z<1.5; The expected service life is calculated by taking the service life of the auto part as the difference between unit 1 and the loss coefficient and multiplying it by the service life of the auto part in the database. The formula is: T=(1-f)*S In the formula, T is the expected service life; S is the service life of the auto parts in the database; Compare the expected service life with the replacement time of the accessory in the relevant area and determine: By obtaining the replacement time G, setting the error range to ±C years, the expected service life T is compared with the replacement time G to determine whether they are equal, that is, whether the error is within the ±C year range: When the expected service life T is equal to the replacement time G, be alert to the reselling of parts in the relevant areas or stores and investigate the repair shops; When the expected service life T is equal to the replacement time G, no alert will be issued for the reselling of accessories in the relevant area or store at this time; Example: The service life S of the brake pad obtained from the database is 2 years. The loss coefficient calculated through regression analysis is 0.42. The expected service life T is calculated according to the formula, T=1.16 / year. The average replacement time of the brake pad in the relevant area obtained from the database is 1.2 years, that is, the replacement time G is 1.2 years; According to the comparison results, an error range of ±0.05 is set and the following judgment is made: When the estimated service life T is equal to the replacement time G or the error range is within ±0.05, the warning mechanism is triggered; When the estimated service life T is not equal to the replacement time G or the error range is outside ±0.05, the warning mechanism will not be triggered; Since the expected service life T is equal to the replacement time G, we need to be vigilant against the reselling of parts in the relevant area. Through applications or text messages, we remind users that the brake pads are about to reach the replacement time, recommend timely inspection and replacement, strengthen monitoring of the parts market in the area, prevent illegal reselling of brake pads, report warning information to local law enforcement agencies, assist in cracking down on illegal activities, and check repair shops to ensure that they are not involved in illegal reselling of parts.

[0027] In the application, the several formulas involved are all calculated by removing the dimensions and taking their numerical values. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The formula is set by technical personnel in this field according to actual conditions.

[0028] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0029] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0030] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A dynamic tracking method for missing automobile parts, characterized in that: The method includes: Step 1: Obtaining the accessory information of the automobile accessory and confirming the user identity information, the accessory information includes an identification code and a maintenance record, obtaining the activation state of the identification code and performing a pre-configured operation according to the activation state; wherein, if the activation state of the identification code is inactive, directly activating the accessory and binding the user identity; if the activation state of the identification code is activated, uploading or downloading the maintenance record and outputting the accessory information; Step 2: When the accessory bound to the user identity information is missing, the missing accessory is tracked according to the accessory information; the missing accessory tracking includes: marking the missing accessory, and if the accessory information of the missing accessory is uploaded, locking the upload location of the accessory information of the missing accessory; if the accessory information of the missing accessory is not uploaded, obtaining at least five maintenance records of the missing accessory before the missing accessory is lost, associating the maintenance records with multiple groups of background user data obtained in advance, and combining the relationship between the maintenance records and the missing accessories, establishing a maintenance record and missing accessories feedback mechanism, and outputting the accessory information of the missing accessories; Step 3: Obtain the automobile image obtained by pre-scanning, and establish a model relationship between the loss and maintenance record of the automobile accessories in combination with the accessory information, and generate a loss coefficient based on the model relationship; Step 4: Calculate the estimated service life of the auto parts based on the recorded service life of the auto parts in the database and the loss coefficient, compare the estimated service life with the replacement time of the missing parts in the relevant area, and determine if the estimated service life of the missing parts is equal to the replacement time, triggering an alert mechanism.

2. A method for dynamically tracking missing auto parts according to claim 1, characterized in that: In the step 1, confirming the user identity information includes: First, the account information provided by the user is verified and compared with the pre-stored user database. After successful identity authentication, access rights are assigned.

3. A dynamic tracking method for missing automobile parts according to claim 2, characterized in that: In the step 1, obtaining the activation status of the identification code and performing a pre-configured operation according to the activation status includes: Obtain the unique identification code on the missing accessory, read the identification code, and upload the read data; perform preliminary verification on the uploaded identification code, and after the data verification passes, query the activation status of the identification code, find the record corresponding to the uploaded identification code, and check the activation status field in the database record; According to the query results, determine whether the identification code has been activated: When the status field of the identification code shows activated, a corresponding message is returned to show that the identification code has been activated; When the status field of the identification code shows inactive, a corresponding message is returned, indicating that the identification code has not been activated, and the accessory is bound to the identity.

4. The method for dynamically tracking missing automobile parts according to claim 1, characterized in that: In the step 2, if the accessory information of the missing accessory is uploaded, locking the location where the accessory information of the missing accessory is uploaded includes: When it is confirmed that an accessory is missing, the missing accessory will be marked as "missing" in the system. By logging into the identity account bound to the missing accessory, the missing accessory will be marked through the identity account and the status of the accessory will be updated in the database. If the missing accessory information is obtained again, the geographic location of the missing accessory will be captured.

5. The method for dynamically tracking missing auto parts according to claim 1, characterized in that: In the step 2, the maintenance records are analyzed and the relationship between the maintenance records and the missing parts is monitored, and a maintenance record and missing parts feedback mechanism is established, including: S201: extracting abnormal situation features, including parts maintenance frequency, maintenance type, maintenance location change and maintenance time interval, and pre-processing the abnormal situation features; S202: Extract the five maintenance records of the accessory before the last upload from the database, as well as multiple sets of user data obtained from the background, establish a maintenance record and missing accessories feedback mechanism, and monitor the proportion of abnormal situations in the maintenance records in the missing accessories. The formula is as follows: ； Where p is the proportion of abnormal situations when accessories are missing, Ai and Bi are the characteristic vectors of abnormal situations A and B respectively; S203: Calculate the probability of abnormal situations of the five maintenance records before the brake pad is missing, sort the probabilities of the abnormal situations, and alert the top g abnormal situations.

6. The method for dynamically tracking missing automobile parts according to claim 1, characterized in that: In the step 3, the automobile image obtained by pre-scanning is obtained, and the model relationship between the loss and maintenance record of the automobile accessories is established in combination with the accessory information, including: Scanning the exterior of the car at multiple angles to obtain a car image, wherein the car image includes at least a front view, a side view, and a rear view; Correcting and reducing noise on the car image, and cutting out the background; The pre-trained CNN model is used to extract the features of the car image, and the regression analysis method is used to establish the relationship between the user's parts loss and maintenance records. The linear regression model is used to establish the model relationship between the car parts loss and maintenance records. The formula is: ； Where y is the degree of wear and tear of the car; beta0 is the intercept term; beta1, beta2, ..., betan are regression coefficients, indicating the impact of each feature on the degree of wear and tear; x1, x2, ..., xn are maintenance records; and e is the error term.

7. A method for dynamically tracking missing auto parts according to claim 6, characterized in that: In the step three, generating a loss coefficient based on the model relationship includes: By obtaining the loss degree of different user groups, combined with the total number of user groups, the current user group and the loss value, the loss coefficient f of the car is calculated according to the formula: ； Where, f is the loss coefficient; m is the total number of user groups; i is the number of the current user group; y0, y1, y2...yn are the loss degrees of different user groups; z is the loss value, ranging from 1 <z<1.5。 8. A method for dynamically tracking missing auto parts according to claim 7, characterized in that: In the step 4, the estimated service life of the auto parts is calculated by combining the recorded service life of the auto parts in the database and the loss coefficient, including: The service life of the auto parts is taken as the difference between unit 1 and the loss coefficient, and multiplied by the recorded service life of the auto parts in the database to calculate the expected service life of each auto part. The formula is: T=(1-f)*S Where T is the expected service life of the auto parts; f is the loss coefficient; S is the recorded service life of the auto parts in the database.

9. The method for dynamically tracking missing automobile parts according to claim 1, characterized in that: In the step 4, if the estimated service life is equal to the replacement time, a warning mechanism is triggered, including: Based on the comparison results of the expected service life and replacement time, the error range is set to ±0.05 and the following judgment is made: If the estimated service life is not equal to the replacement time or the error range is outside ±0.05, the warning mechanism will not be triggered; If the estimated service life is equal to the replacement time or the error range is within ±0.05, the warning mechanism will be triggered, and the reselling of accessories in the relevant area or store will be recorded and output.

Images