Standardized management method, device, equipment and storage medium

Abstract

The present application relates to the technical field of data processing, and more particularly to a standardized management method, device and equipment and storage medium, which gets rid of the subjective limitations of traditional manual on-site inspection, realizes objective quantification and accurate determination of compliance audit through the combination of a clear multi-step data preprocessing process and a specific algorithm, and ensures the sustainability, traceability and reproducibility of standardized management with the help of a hierarchical classification periodic reminder mechanism and a full-dimensional data management library; the method not only solves the core technical problems such as lack of standard landing monitoring, isolated execution, low efficiency, subjective and fuzzy audit results and the like in the existing management mode, but also significantly reduces the labor cost and time cost of standardized management through process optimization and algorithm empowerment, effectively shortens the single batch audit time, improves the accuracy of the audit results, and improves the fairness and accuracy of the audit results, providing systematic technical support for brand image standardization and operation efficiency optimization.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a standardized management method, apparatus, equipment, and storage medium. Background Technology

[0002] The standardized management of storefronts, operating vehicles, and service counters in the industry is currently facing multiple challenges. The existing management model relies too heavily on periodic manual on-site inspections, and the evaluation standards are mostly subjective descriptions such as cleanliness and standardization, lacking objective and quantifiable data support. This leads to doubts about the fairness of the assessment results and fails to provide customers with clear directions for improvement.

[0003] Furthermore, the management process only extends to policy issuance and final result acceptance, lacking effective monitoring and timely correction of the entire standard implementation process. During the cycle from standard issuance to completion of transformation, issues such as information attenuation and execution deviations are prone to occur. Moreover, the standardized management system and core business systems are isolated from each other, making data entry an additional burden and resulting in insufficient data authenticity and freshness.

[0004] Furthermore, the existing logistics management system has a complex interface logic, cumbersome operation, and poor mobile terminal adaptability, making it difficult to meet the needs of front-line customers for high-frequency and rapid operations, further exacerbating the difficulty of standardization.

[0005] However, concrete problems such as inconsistent visual elements of storefronts, unauthorized modifications and damage to vehicle paint, and non-standard hardware and services at service counters have long been unable to be systematically resolved due to the lack of an efficient collection, review, and feedback mechanism, which seriously affects brand image and operational efficiency.

[0006] It is evident that existing technologies still need improvement and enhancement. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, the present invention aims to provide a standardized management method that constructs a closed-loop system covering the entire process from information acquisition, processing, and review to feedback monitoring, thereby achieving objective quantification and accurate judgment of compliance review.

[0008] The first aspect of this invention provides a standardized management method, comprising: acquiring information to be reviewed, the information to be reviewed including at least one of store information, service desk information, and vehicle information; sequentially performing data preprocessing and compliance review processing on the information to be reviewed, generating corresponding review results and rectification suggestions corresponding to the review results, the review results including at least one of store review results, service desk review results, and vehicle review results, and the rectification suggestions including at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions; retrieving a preset periodic reminder mechanism, determining a reminder period based on the review results and the periodic reminder mechanism, integrating the review results, rectification suggestions, and reminder period to generate review feedback information; storing the review feedback information in a pre-built data management database, and monitoring the review feedback information according to the reminder period.

[0009] Optionally, in a first implementation of the first aspect of the present invention, the step of obtaining the information to be reviewed, which includes at least one of store information, service desk information, and vehicle information, includes: obtaining customer-uploaded data to be reviewed and login account permission information, wherein the data to be reviewed includes at least one of store information, service desk information, and vehicle information; associating the basic information of the corresponding end-service station based on the login account permission information, wherein the basic information includes the customer name and a unique customer code; and associating and integrating the data to be reviewed with the basic information to generate the information to be reviewed.

[0010] Optionally, in a second implementation of the first aspect of the present invention, the step of sequentially performing data preprocessing and compliance review processing on the information to be reviewed to generate corresponding review results and rectification suggestions corresponding to the review results includes: when the information to be reviewed includes store information or service desk information, compressing the information to be reviewed to obtain compressed review information; using a Gaussian filtering algorithm to denoise the compressed review information, and extracting key visual features from the denoised compressed review information using image segmentation technology to obtain structured feature data to be reviewed; retrieving a preset standardized parameter library corresponding to the information to be reviewed, and using a convolutional neural network model to compare the structured feature data to be reviewed and its corresponding standardized parameter library to output a similarity score; using a target detection algorithm to identify defects in the denoised compressed review information to generate defect detection results; determining the review result based on the similarity score and the defect detection results, and determining rectification suggestions based on the defect detection results.

[0011] Optionally, in a third implementation of the first aspect of the present invention, the step of compressing the review information to obtain compressed review information when the information to be reviewed includes store information or service desk information includes: when the information to be reviewed includes store information or service desk information, determining the information type of the information to be reviewed, wherein the information type includes image type and video type; if the information to be reviewed is image type and includes multiple images, then performing uniform resolution adjustment on the information to be reviewed, and compressing the information to be reviewed after uniform resolution adjustment according to preset compression parameters to obtain compressed review information; if the information to be reviewed is video type, then performing cropping processing on the information to be reviewed according to a preset duration threshold, and performing keyframe extraction processing on the cropped information to generate compressed review information.

[0012] Optionally, in the fourth implementation of the first aspect of the present invention, the step of sequentially performing data preprocessing and compliance review processing on the information to be reviewed, and generating corresponding review results and rectification suggestions corresponding to the review results, includes: when the information to be reviewed includes vehicle information, compressing the vehicle information to obtain compressed vehicle information; using a Gaussian filtering algorithm to denoise the compressed vehicle information, and extracting key visual features of the vehicle from the denoised compressed vehicle information using image segmentation technology, wherein the key visual features of the vehicle include paint area, color parameters, and vehicle body integrity; collecting vehicle type information corresponding to the license plate, integrating the key visual features of the vehicle and the vehicle type information to obtain vehicle structured feature data; retrieving a preset vehicle standardization parameter library, using a convolutional neural network model to compare and calculate the vehicle structured feature data and the vehicle standardization parameter library, and outputting a fit score; using a target detection algorithm to identify defects in the denoised compressed vehicle information, generating vehicle defect detection results; determining the vehicle review result based on the fit score and the vehicle defect detection results, and determining vehicle rectification suggestions based on the vehicle defect detection results.

[0013] Optionally, in the fifth implementation of the first aspect of the present invention, the step of retrieving a preset periodic reminder mechanism, determining a reminder period based on the audit result and the periodic reminder mechanism, and integrating the audit result, rectification suggestions, and reminder period to generate audit feedback information includes: retrieving a preset periodic reminder mechanism to determine the corresponding reminder period and reminder node based on the audit result; when the audit result is qualified, the reminder period is 6 months, and the reminder node corresponding to the reminder period is 15 days before the expiration date; when the audit result is unqualified, the corresponding reminder period is determined based on the defect detection result, and the reminder node corresponding to the reminder period is 1 day before the expiration date; and integrating the audit result, rectification suggestions, reminder period, and reminder node to generate audit feedback information.

[0014] Optionally, in the sixth implementation of the first aspect of the present invention, the step of storing the review feedback information in a pre-built data management library and monitoring the review feedback information according to the reminder period includes: associating and integrating the review feedback information, the structured feature data to be reviewed, and the compressed review information after noise reduction, generating a review record package, and storing the review record package in the pre-built data management library; monitoring the flow status of the review feedback information based on the reminder period, the flow status including pending rectification, in rectification, rectified, and overdue; if the flow status is overdue, generating alarm information.

[0015] A second aspect of the present invention provides a standardized management device, comprising: an acquisition module for acquiring information to be reviewed, the information to be reviewed including at least one of store information, service desk information, and vehicle information; an review module for sequentially performing data preprocessing and compliance review processing on the information to be reviewed, generating corresponding review results and rectification suggestions corresponding to the review results, the review results including at least one of store review results, service desk review results, and vehicle review results, and the rectification suggestions including at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions; a generation module for retrieving a preset periodic reminder mechanism, determining a reminder period based on the review results and the periodic reminder mechanism, integrating the review results, rectification suggestions, and reminder period to generate review feedback information; and a monitoring module for storing the review feedback information in a pre-built data management database and monitoring the review feedback information according to the reminder period.

[0016] A third aspect of the present invention provides a standardization management device, the standardization management device comprising: a memory and at least one processor, the memory storing instructions; the at least one processor invokes the instructions in the memory to cause the standardization management device to perform the various steps of the standardization management method described in any of the preceding claims.

[0017] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the standardized management method described in any of the preceding claims.

[0018] The technical solution of this invention overcomes the subjective limitations of traditional manual on-site inspections. By combining a clear multi-step data preprocessing workflow with specific algorithms, it achieves objective quantification and accurate judgment of compliance audits. With a tiered and categorized periodic reminder mechanism and a comprehensive data management library, it ensures the continuity, traceability, and reproducibility of standardized management. This method not only specifically addresses core technical problems in existing management models, such as incomplete standard implementation monitoring, isolated data, low execution efficiency, and subjective ambiguity in audit results, but also significantly reduces the human and time costs of standardized management through process optimization and algorithm empowerment. It effectively shortens the audit time for a single batch, improves the accuracy of audit results, and enhances the fairness and accuracy of audit results. It provides systematic technical support for brand image standardization and operational efficiency optimization, and is applicable to the standardized management needs of various service industries that rely on offline sites and mobile operation tools. Attached Figure Description

[0019] Figure 1 A first flowchart of a standardized management method provided in an embodiment of the present invention; Figure 2 A second flowchart of the standardized management method provided in the embodiments of the present invention; Figure 3 A third flowchart of the standardized management method provided in the embodiments of the present invention; Figure 4 A fourth flowchart of the standardized management method provided in this embodiment of the invention; Figure 5 A fifth flowchart of the standardized management method provided in the embodiments of the present invention; Figure 6 A sixth flowchart of the standardized management method provided in the embodiments of the present invention; Figure 7 A seventh flowchart of the standardized management method provided in the embodiments of the present invention; Figure 8 A schematic diagram of the structure of the standardized management device provided in the embodiments of the present invention; Figure 9 A schematic diagram of the structure of the standardized management equipment provided in an embodiment of the present invention. Detailed Implementation

[0020] This invention provides a standardized management method, apparatus, device, and storage medium. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the standardized management method in this invention includes: 101. Obtain information to be reviewed, wherein the information to be reviewed includes at least one of store information, service desk information, and vehicle information; In this embodiment, the information to be reviewed is obtained by the customer uploading corresponding data after entering the standardized management module through the enterprise's designated mobile application APP or its lightweight mini-program. While receiving the uploaded data, the system simultaneously collects the customer's login account permission information. This account permission information is pre-bound to the customer's end-service site through the enterprise's backend management system. The binding relationship includes unique identification information such as account ID, customer code, and site code, ensuring that each piece of information to be reviewed can clearly identify the subject and providing a clear subject basis for subsequent review and traceability.

[0022] 102. Perform data preprocessing and compliance audit processing on the information to be audited in sequence to generate corresponding audit results and rectification suggestions corresponding to the audit results. The audit results include at least one of store audit results, service desk audit results, and vehicle audit results. The rectification suggestions include at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions. In this embodiment, data preprocessing is a standardized optimization operation performed on different types of information to be reviewed, which effectively improves the quality and standardization of the information to be reviewed. By removing invalid and interfering information and extracting core feature data, it provides a reliable data foundation for compliance review and avoids review errors caused by problems with the quality of the original data.

[0023] 103. Retrieve the preset periodic reminder mechanism, determine the reminder period based on the audit results and the periodic reminder mechanism, integrate the audit results, rectification suggestions and reminder period, and generate audit feedback information; In this embodiment, the preset periodic reminder mechanism is a system of reminder rules pre-configured based on different audit results and different defect levels. This mechanism clarifies the core content such as the reminder period duration, number of reminder nodes, reminder method, reminder recipients, and information content templates corresponding to various audit results. The introduction of the periodic reminder mechanism breaks the static mode of traditional standardized management and ensures the long-term implementation of standardized requirements. The differentiated reminder period settings based on audit results and defect levels ensure the relevance and rationality of reminders, avoiding information interference caused by excessive reminders or rectification omissions caused by insufficient reminders.

[0024] 104. Store the audit feedback information in a pre-built data management database, and monitor the audit feedback information according to the reminder cycle; In this embodiment, the pre-built data management library is a structured database specifically designed to store data for the entire standardized management process. It adopts a MySQL database architecture and possesses core functions such as data classification, storage, retrieval, traceability, security backup, access control, and data encryption. The data management library's storage function enables centralized archiving of data for the entire standardized management process, ensuring data integrity, security, and relevance. This provides solid data support for subsequent data statistical analysis, compliance audit issue tracing, and responsibility identification. Real-time monitoring based on reminder cycles effectively avoids delays or oversights in rectification, ensuring that audit results are effectively implemented and improving the execution of standardized management.

[0025] The standardized management method disclosed in this invention overcomes the subjective limitations of traditional manual on-site inspections. Through a clear multi-step data preprocessing workflow combined with specific algorithms, it achieves objective quantification and precise judgment of compliance audits. Utilizing a tiered and categorized periodic reminder mechanism and a comprehensive data management library, it ensures the continuity, traceability, and reproducibility of standardized management. This method not only specifically addresses core technical issues in existing management models, such as incomplete standard implementation monitoring, isolated data, low execution efficiency, and subjective ambiguity in audit results, but also significantly reduces the human and time costs of standardized management through process optimization and algorithm empowerment. It effectively shortens the audit time for each batch, improves the accuracy of audit results, and enhances the fairness and accuracy of audit results. This provides systematic technical support for brand image standardization and operational efficiency optimization, and is applicable to the standardized management needs of various service industries that rely on offline sites and mobile operation tools.

[0026] Please see Figure 2In this embodiment of the invention, the step of obtaining the information to be reviewed includes at least one of store information, service desk information, and vehicle information, including: 201. Obtain the data to be reviewed uploaded by the customer and the login account permission information, wherein the data to be reviewed includes at least one of store information, service desk information, and vehicle information; In this embodiment, the data types of the data uploaded by the customer for review include four categories: static images, dynamic videos, text descriptions, and numerical parameters. For example, the data to be reviewed includes store information such as front and side views of the storefront, installation location diagrams, overall layout videos of the service counter, detailed diagrams of functional areas, and material placement lists, complete paint images of vehicles, clear images of license plates, and vehicle type and size parameters. The login account permission information refers to the relevant permission configuration data of the account used by the customer when logging into the standardized management module, and this account is uniformly assigned by the enterprise.

[0027] 202. Based on the login account permission information, associate the basic information of the corresponding end-service site, the basic information including the customer name and unique customer code; In this embodiment, the system pre-establishes a mapping relationship between login account permission information and basic information of end-service sites. This mapping relationship is stored in key-value pairs, where the key is the account ID or permission identifier, and the value is the corresponding basic information of the end-service site. This ensures that each legitimate account uniquely corresponds to one end-service site, avoiding data ownership confusion. When the login account permission information is obtained, the corresponding basic information of the end-service site is automatically retrieved and displayed through this mapping relationship. Here, the customer name is the official registered name of the end-service site, and the unique customer code is the exclusive identification code assigned by the enterprise to each end-service site. Through the automatic association and verification mechanism between account permissions and site basic information, data processing efficiency is improved, the accuracy and uniqueness of the subject of the information to be reviewed are strengthened, and a clear identification basis is provided for the accurate push of subsequent review results and the location of responsibility issues.

[0028] 203. Associate and integrate the data to be reviewed with the basic information to generate information to be reviewed; In this embodiment, the integration of the data to be reviewed with the basic information ensures that each piece of data has a clear subject identifier and complete associated information, providing a clear and reliable basis for accurate feedback of subsequent review results, determination of responsibility, and tracing of issues.

[0029] Please see Figure 3 In this embodiment of the invention, the step of sequentially performing data preprocessing and compliance review processing on the information to be reviewed, and generating corresponding review results and rectification suggestions corresponding to the review results, includes: 301. When the information to be reviewed includes store information or service desk information, the information to be reviewed is compressed to obtain compressed review information. In this embodiment, the compression processing targets include image data and video data. Through compression processing, the data storage and transmission pressure can be significantly reduced, and the operating efficiency of the entire processing flow can be improved.

[0030] 302. The compressed review information is denoised using a Gaussian filtering algorithm, and key visual features are extracted from the denoised compressed review information using image segmentation technology to obtain the structured feature data to be reviewed. In this embodiment, the Gaussian filtering algorithm is a commonly used linear image denoising algorithm. In this step, a 3×3 filter kernel is used with a standard deviation of 0.8. The image data in the compressed review information is denoised through convolution operation. This can effectively filter out image noise and blur caused by insufficient ambient light, low pixel count of the shooting device, and shooting shake, thereby improving the clarity and quality of the image and providing a reliable guarantee for the accurate extraction of key visual features in the subsequent process. Image segmentation technology uses the semantic segmentation U-Net model, which can accurately divide the denoised image into regions according to semantics, distinguishing different regions such as the entrance, service counter, and background environment; For store information, the key visual features extracted include the location coordinates of the logo area on the storefront, logo size parameters, logo color RGB values, overall storefront size parameters, installation height, storefront material characteristics, and storefront surface integrity. For service counter information, the key visual features extracted include the counter's length, width, and height dimensions, counter style characteristics, backdrop design style and color parameters, functional area layout coordinates, material placement coordinates, material type identification, promotional material expiration date identification, and countertop cleanliness characteristics. The structured feature data to be reviewed is standardized data formed by quantifying, organizing and encoding the extracted key visual features. It is stored in JSON format, and each feature parameter contains specific numerical units and precision information, which facilitates comparison and calculation with the standardized parameter library.

[0031] 303. Retrieve the preset standardized parameter library corresponding to the information to be reviewed, and use the convolutional neural network model to compare the structured feature data to be reviewed with its corresponding standardized parameter library to output a similarity score. In this embodiment, the preset standardized parameter library is pre-built based on industry standards, enterprise internal management specifications, actual operating experience and brand image requirements, and corresponds one-to-one with the types of information to be reviewed, that is, store information corresponds to the store standardized parameter library, and service desk information corresponds to the service desk standardized parameter library. The standardized parameter library includes specific quantitative standards for various key visual features, such as standard numerical ranges, allowable deviation thresholds, standard style templates, and standard color RGB value ranges. For example, the standard style template for a storefront logo has a standard color RGB value range of R255G0B0±5, a standard size range of 50cm-60cm in length and 30cm-40cm in width, and an allowable deviation threshold of ±2cm. The standard dimensions for a service counter are 200cm in length, 60cm in width, and 100cm in height, with an allowable deviation threshold of ±3cm, as well as standard layout coordinates for functional areas. The convolutional neural network model uses the ResNet50 model. This model compares the structured feature data to be reviewed with the corresponding standards in the standardized parameter library dimension by dimension. The comparison results of each feature dimension are weighted according to preset weights. The weights are set according to the importance of the features, with core features having higher weights and secondary features having lower weights. Finally, a similarity score is output based on the weighted calculation results. The score ranges from zero to one hundred. The higher the score, the more the information to be reviewed meets the standardization requirements. A score of 90 or above indicates that both core and secondary features meet the standard requirements. A score of 80 to 89 indicates that the core features meet the requirements and there are slight deviations in the secondary features. A score below 80 indicates that there are deviations in the core features or multiple secondary features.

[0032] 304. Use a target detection algorithm to identify defects in the denoised compressed audit information and generate defect detection results; In this embodiment, the target detection algorithm uses the YOLOv5 model. During the detection process, the confidence threshold is set to 0.7 and the IOU threshold is set to 0.5 to ensure the accuracy and reliability of the detection results. For store information, the scope of defect identification includes issues such as mixed use of logos on the storefront, logo size deviation, logo color mismatch, damaged storefront, faded storefront, obstruction around the storefront, non-standard installation location of the storefront, and inconsistent storefront materials; For information about the service counter, the scope of defect identification includes issues such as incorrect counter size and style, non-standard background design, chaotic functional zoning, random placement of materials, expired promotional materials, dirty and messy countertops, non-compliant material types, and non-standard employee attire. During defect detection, the YOLOv5 model automatically identifies defect areas in images, marks defect location coordinates, determines defect types, and categorizes defects into three levels—minor, moderate, and major—based on their impact on standardization implementation. Minor defects have minimal impact on brand image and operational functions and can be quickly rectified, such as slight dust on countertops or minor misplacement of materials. Moderate defects have some impact on brand image and operational functions and require rectification within a specified timeframe, such as expired promotional materials or slight fading of storefront signs. Major defects severely impact brand image and operational functions and require immediate rectification, such as misuse of logos, severe damage to storefront signs, or significantly incorrect counter dimensions. The defect detection results are a detailed record of the identified defects, including the defect type, defect location coordinates, defect severity, defect description, and defect impact range, ensuring that management and customers can clearly understand the specific problems.

[0033] 305. Determine the audit result based on the similarity score and defect detection results, and determine rectification suggestions based on the defect detection results; In this embodiment, the audit result is determined using a dual judgment standard, which combines the similarity score and the defect detection results for comprehensive judgment. Specifically, the judgment rule is as follows: if the similarity score is not lower than 90 and there are no major defects, no moderate defects, or only one minor defect, the audit is judged as qualified; if the similarity score is between 80 and 89 and there are only minor defects, and the number of minor defects does not exceed 3, the audit is judged as qualified but requires optimization; if the similarity score is lower than 80 or there are major defects, or the number of moderate defects exceeds 2 or the number of minor defects exceeds 3, the audit is judged as unqualified. The determination of rectification suggestions is based on the specific defects recorded in the defect detection results. Combining the corresponding standards in the standardized parameter library, the severity of the defect, and the difficulty of rectification, a customized improvement suggestion is formulated for each defect. The rectification suggestion must include core content such as rectification goals, rectification steps, rectification standards, rectification timeframe, required materials, and rectification acceptance criteria to ensure the accuracy, operability, and verifiability of the suggestion. For example, for a major defect of mixed logo usage, the rectification suggestion clearly states the rectification goal as replacing it with the latest standard logo. The rectification steps include contacting the company's designated supplier to obtain the standard logo, removing the old logo, cleaning the installation area, installing the standard logo, and taking images of the rectified logo. The rectification standards are that the logo style, color, and size meet the requirements of the store's standardized parameter library. The rectification timeframe is 3 working days. The rectification acceptance criteria are that the rectified image, after system review, has a similarity score of no less than 90 points and no defects.

[0034] Please see Figure 4In this embodiment of the invention, when the information to be reviewed includes store information or service desk information, the review information is compressed to obtain compressed review information, including: 401. When the information to be reviewed includes store information or service desk information, determine the information type of the information to be reviewed, wherein the information type includes image type and video type; In this embodiment, image information refers to data existing in the form of one or more static images, with file format JPG or PNG; video information refers to data existing in the form of dynamic video, with file format MP4; the system quickly and accurately identifies the information type by reading the file header information and file format suffix of the data, providing a basis for the selection of subsequent compression processing methods.

[0035] 402. If the information to be reviewed is an image and includes multiple images, then a uniform resolution adjustment is performed on the information to be reviewed, and the information to be reviewed after uniform resolution adjustment is compressed according to preset compression parameters to obtain compressed review information. In this embodiment, when the information to be reviewed is an image and contains multiple images, a unified resolution adjustment operation is first performed to adjust it to the enterprise's preset standard resolution of 1920×1080 pixels. During the resolution adjustment process, a bilinear interpolation algorithm is used. This algorithm can effectively preserve image details while adjusting the resolution, avoid image distortion and deformation, and ensure that the aspect ratio of the adjusted image remains consistent with the original image. After the resolution is uniformly adjusted, the images are compressed according to preset compression parameters. The compression algorithm used is JPEG compression, and the preset compression parameters include a compression ratio of 5:1 and a compression quality factor of 80. The size of a single compressed image is strictly controlled within 500KB. After multiple images are compressed, they are named and organized according to image type and shooting order to form compression review information for easy subsequent unified processing. The preset compression parameters ensure the stability and uniformity of the compression effect, significantly reducing data volume and storage and transmission pressure while maximizing the protection of key information required for review, thus improving data processing efficiency and storage utilization.

[0036] 403. If the information to be reviewed is a video, then the information to be reviewed is cropped according to a preset duration threshold, and keyframe extraction is performed on the cropped information to generate compressed review information. In this embodiment, the preset duration threshold is the maximum video retention time set by the system based on review requirements, actual operating scenarios, and data processing efficiency, which is sixty seconds. The cropping process refers to the process of cutting out videos that exceed the preset duration threshold of sixty seconds by using a timeline cropping method to retain the core content of the first sixty seconds of the video. For videos that are less than sixty seconds long, they are left as is and no cropping process is performed. After cropping, keyframe extraction is performed on the cropped video. The keyframe extraction uses the inter-frame difference method, specifically extracting three keyframes: the first 3 seconds, the middle 30 seconds, and the last 3 seconds of the video. Each keyframe corresponds to a specific time point in the video. The extracted keyframe images are processed according to the aforementioned image compression standard, i.e., uniformly adjusted to a resolution of 1920×1080 pixels and compressed to within 500KB using the JPEG compression algorithm. Finally, compressed review information consisting of three keyframe compressed images is generated. This process retains the core visual information required for review while greatly reducing the data size, and enables subsequent operations such as noise reduction, feature extraction, and review comparison to be carried out effectively.

[0037] Please see Figure 5 In this embodiment of the invention, the step of sequentially performing data preprocessing and compliance review processing on the information to be reviewed, and generating corresponding review results and rectification suggestions corresponding to the review results, includes: 501. When the information to be reviewed includes vehicle information, the vehicle information is compressed to obtain compressed vehicle information; In this embodiment, the compression processing targets include image data and video data in vehicle information. Text and numerical data are only subjected to format standardization processing. The compression processing method for image data and video data is consistent with the compression processing method for store information and service desk information.

[0038] 502. The compressed vehicle information is denoised using a Gaussian filtering algorithm, and key visual features of the vehicle are extracted from the denoised compressed vehicle information using image segmentation technology. The key visual features of the vehicle include paint area, color parameters, and vehicle body integrity. In this embodiment, the specific parameter settings of the Gaussian filtering algorithm are consistent with those for the noise reduction processing of store and service counter information. The image segmentation technology focuses on extracting three core key visual features: paint area, color parameters, and vehicle body integrity. At the same time, it also extracts auxiliary features such as license plate area features and vehicle size parameters. Among them, the paint area refers to the specific area on the vehicle that is painted or has advertising signs affixed according to the company's standard requirements. The extracted content includes the location coordinates, range, and size of the paint area. The color parameters refer to the color values ​​of the paint area, which are quantized using the RGB color space. The RGB values ​​of each pixel are extracted and the average value is calculated as the standard color parameters of the paint area. The vehicle body integrity refers to the condition of the vehicle body surface and paint area. The extracted content includes whether there are defects such as damage, peeling, curling, and dirt, as well as the location and range of defects. After quantification, sorting, and encoding, the extracted key visual features form standardized feature data, which is stored in JSON format for easy comparison and review later.

[0039] 503. Collect vehicle type information corresponding to the license plate, integrate the key visual features of the vehicle and the vehicle type information to obtain vehicle structured feature data; In this embodiment, vehicle type information is collected by recognizing license plate information and combining it with the company's pre-set vehicle type registration database. The license plate information is extracted from the denoised compressed vehicle information, and the license plate area is located using the YOLOv5 object detection algorithm. Then, the license plate characters are recognized using the OCR optical character recognition algorithm. The company's pre-set vehicle type registration database stores the license plate information, vehicle type information, and vehicle ownership information of all compliant operating vehicles of cooperative customers. The vehicle types include vans, trucks, tricycles, etc. After the system recognizes the license plate characters, it searches the vehicle type registration database and matches the corresponding vehicle type information according to the license plate characters to ensure the accuracy of the vehicle type information. The integration process involves associating and binding the extracted key visual features of the vehicle with the collected vehicle type information, and then standardizing and organizing the data according to a pre-defined tree-like data structure. The data structure includes three main categories: basic vehicle information, core feature information, and auxiliary feature information. Among them, the basic vehicle information includes license plate number, vehicle type, and vehicle ownership, while the core feature information includes paint area, color parameters, and vehicle body integrity, and the auxiliary feature information includes vehicle size and vehicle status. The final result is a complete and standardized structured vehicle feature data that contains both feature data and type information, providing comprehensive data support for subsequent comparison and verification.

[0040] 504. Retrieve the preset vehicle standardized parameter library, use a convolutional neural network model to compare and calculate the vehicle structured feature data and the vehicle standardized parameter library, and output the matching score. In this embodiment, the preset vehicle standardization parameter library is a dedicated database built according to the enterprise's vehicle standardization management requirements, brand image specifications, industry standards and actual operational needs, forming a unified standardization system with the store and service desk standardization parameter libraries; The vehicle standardization parameter library includes various quantitative parameters related to vehicle standardization, including standard templates for vehicle painting, standard color RGB value ranges, logo position and size standards, compliance standards for different vehicle types, standards for judging vehicle body integrity, and standard ranges for painting areas. For example, the standard color RGB value range for truck painting is R0G128B255±5, the standard logo position is the middle area on the side of the vehicle, the standard size is 80cm long and 50cm wide, the allowable deviation threshold is ±3cm, and the standard for vehicle body integrity is no damage, no peeling, no curling, and no dirt. The convolutional neural network model uses the ResNet50 model, which is consistent with the model used for information verification at stores and service counters. This model compares the vehicle's structured feature data with the corresponding standards in the vehicle's standardized parameter library dimension by dimension. Each feature dimension is assigned a weight according to its importance. Core features such as paint color and logo position have higher weights, while auxiliary features such as vehicle size have lower weights. The comparison results of each dimension are weighted and calculated to finally output a matching score. The score ranges from zero to one hundred points, which intuitively reflects the degree of standardization compliance of the vehicle information. The higher the score, the higher the degree of compliance.

[0041] 505. Use a target detection algorithm to identify defects in the denoised compressed vehicle information and generate vehicle defect detection results; In this embodiment, the target detection algorithm adopts the YOLOv5 model, which is consistent with the defect identification model of stores and service counters. The scope of defect identification is specifically set for the characteristics of vehicle information, including vehicle paint peeling, curling, unauthorized advertising, color deviation, dirt, damage, scratches on the vehicle body, unclear, obscured, or missing license plates, non-compliant vehicle types, mixed use of logos, size deviation, position deviation, and other issues. During the defect identification process, the model automatically locates the defect area, marks the defect location coordinates, determines the defect type, and classifies defects into three levels—minor, moderate, and major—based on their impact on brand image, operational safety, and standardization implementation. Minor defects, such as slight dust on the vehicle body or slight color difference in the paint, have minimal impact on operations and the brand. Moderate defects, such as slight localized paint peeling or slightly blurred license plates, have some impact on operations and the brand. Major defects, such as large-area paint peeling, mixed logo usage, or non-compliant vehicle types, severely impact operations and brand image. Vehicle defect detection results are detailed records of identified defects, including defect type, defect location coordinates, defect severity, defect description, defect impact range, and rectification priority. This ensures the detail and accuracy of defect information, providing sufficient basis for subsequent review results and rectification recommendations.

[0042] 506. Determine the vehicle review result based on the fit score and vehicle defect detection results, and determine vehicle rectification suggestions based on the vehicle defect detection results; In this embodiment, the determination of vehicle review results adopts a dual standard that combines the fit score and the defect detection results. The specific judgment rules are consistent with the store and service desk information review rules, so as to achieve the uniformity of the review standards of the entire system. Vehicle rectification recommendations are tailored to specific defects identified in defect detection results. They are developed based on the requirements of the vehicle's standardized parameter library, the severity of the defect, the difficulty of rectification, and the priority of rectification. Each recommendation includes core elements such as rectification goals, steps, standards, timelines, required materials, acceptance criteria, and responsible party, ensuring operability and verifiability. For example, for a moderate defect of paint peeling, the rectification recommendation specifies the goal as repairing the peeling area to meet standardized requirements. The steps include cleaning the peeling area, sanding the base layer, contacting a designated supplier to obtain standard paint materials, spraying according to standard processes, allowing it to air dry, and taking images of the rectified area. The rectification standards are that the paint color and material meet the requirements of the vehicle's standardized parameter library, with no peeling or bubbles. The rectification timeline is 7 working days, and the acceptance criteria are that the system verifies the absence of the defect in the rectified images.

[0043] Please see Figure 6 In this embodiment of the invention, the step of retrieving a preset periodic reminder mechanism, determining the reminder period based on the audit results and the periodic reminder mechanism, and integrating the audit results, rectification suggestions, and reminder period to generate audit feedback information includes: 601. Retrieve the preset periodic reminder mechanism to determine the corresponding reminder period and reminder node based on the audit results; In this embodiment, the preset periodic reminder mechanism clarifies the core content such as the reminder period duration, number of reminder nodes, reminder method, reminder recipients, and information content templates corresponding to different review results. It is a key mechanism to ensure the continuity of standardized management.

[0044] 602. When the audit result is qualified, the reminder period is 6 months, and the reminder node corresponding to the reminder period is the 15th day before the expiration of the validity period; In this embodiment, when the audit result is qualified, it indicates that the information to be audited fully meets the standardization requirements. The six-month reminder cycle is determined based on the normal service life of key visual features such as storefronts and vehicle paint, the outdoor environmental wear and tear patterns, and the company's standardized maintenance cycle. The reminder node is set to the fifteenth day before the expiration date, allowing customers ample preparation time, including arranging for personnel to photograph new review data, upload and submit it, and wait for review, thus avoiding overdue updates due to time constraints.

[0045] 603. When the audit result is unqualified, a corresponding reminder period is determined based on the defect detection result, and the reminder node corresponding to the reminder period is the first day before the expiration date; In this embodiment, when the audit result is unqualified, the determination of the reminder period needs to be combined with the severity of the defect in the defect detection result. The severity of the defect is divided into three levels: minor, moderate and severe. Different levels correspond to different rectification difficulties and required time. Therefore, a differentiated reminder period is set. For minor defects, which are easy to rectify and quick to fix, the reminder period is three days; for moderate defects, which require a certain amount of time and effort to rectify, the reminder period is five days; and for severe defects, which are difficult to rectify and may involve supplier collaboration, the reminder period is seven days. The reminder deadline is set to the day before the expiration date to ultimately urge customers to complete the rectification and prevent them from forgetting or delaying the rectification.

[0046] 604. Integrate the audit results, rectification suggestions, reminder cycles, and reminder points to generate audit feedback information; In this embodiment, the integration operation involves organizing key information such as the specific level of the audit result, the judgment basis, the details of the defects, the detailed content of the rectification suggestions, the duration of the reminder period, the specific date of the reminder node, the reminder method, and the recipients according to a preset unified format, forming audit feedback information that is clearly structured, logically coherent, and complete in elements. The integrated audit feedback information centrally presents all the key information required by the client and the management, avoiding problems such as inconvenience in obtaining information and misunderstanding caused by information dispersion, and improving the efficiency and accuracy of information transmission.

[0047] Please see Figure 7 In this embodiment of the invention, storing the review feedback information in a pre-built data management database and monitoring the review feedback information according to the reminder period includes: 701. Associate and integrate the audit feedback information, the structured feature data to be audited, and the compressed audit information after noise reduction, generate an audit record package, and store the audit record package in a pre-built data management library; In this embodiment, the association and integration operation binds three types of core data: audit feedback information, structured feature data to be audited, and compressed audit information after noise reduction, to form an audit record package containing key data of the entire audit process, ensuring the integrity and relevance of the data. Each piece of data in the audit record package is uniquely identified by dimensions such as timestamp, audit object customer code, and audit module type, providing complete data support for subsequent compliance audits, problem tracing, responsibility identification, and data analysis.

[0048] 702. Monitor the flow status of the review feedback information based on the reminder cycle, wherein the flow status includes pending rectification, in the process of rectification, rectified, and overdue; In this embodiment, monitoring based on the reminder cycle means that the system tracks the execution status, i.e. the flow status, of the review feedback information in real time from the date the review feedback information is generated, according to the determined reminder cycle and reminder node, so as to ensure full control over the standardized execution process; The workflow status is divided into four states: pending rectification, in progress, rectified, and overdue. Pending rectification means the review feedback information has been successfully pushed to the customer, who has viewed it but has not yet taken any rectification or data update actions; this can be determined by checking the customer's viewing records and operation logs. In progress means the customer has submitted supplementary data after rectification, but the system has not yet completed the review of this supplementary data; this can be determined by checking the data upload records and review process status. Rectified means the customer's submitted rectification data has been approved by the system, or the information update corresponding to the qualified result has been completed and approved; this can be determined by the final review result. Overdue means that the customer has not completed the rectification or information update actions after the set reminder period has expired; this can be determined by comparing timestamps with operation records.

[0049] 703. If the circulation status is overdue, an alarm message is generated; In this embodiment, when the system detects that the status of the review feedback information has changed to overdue, it will automatically trigger an alarm mechanism and generate alarm information containing key information. The alarm information includes core information such as customer name, unique customer code, review target module, review result, overdue duration, description of incomplete items, and corresponding reminder cycle and node. The overdue alarm mechanism effectively solves the problem of overdue items being difficult to detect in a timely manner in traditional management, ensuring that managers can quickly intervene in overdue or unrectified items, take targeted measures to urge customers to complete the relevant operations, and avoid the loss of control in standardized management.

[0050] The standardization management method in the embodiments of the present invention has been described above. The standardization management device in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 8 One embodiment of the standardized management device in this invention includes: The acquisition module 801 is used to acquire information to be reviewed, which includes at least one of store information, service desk information, and vehicle information. The audit module 802 is used to sequentially perform data preprocessing and compliance audit processing on the information to be audited, and generate corresponding audit results and rectification suggestions corresponding to the audit results. The audit results include at least one of store audit results, service desk audit results, and vehicle audit results, and the rectification suggestions include at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions. The generation module 803 is used to retrieve a preset periodic reminder mechanism, determine the reminder period based on the audit results and the periodic reminder mechanism, integrate the audit results, rectification suggestions and reminder period, and generate audit feedback information; The monitoring module 804 is used to store the review feedback information in a pre-built data management library and monitor the review feedback information according to the reminder period.

[0051] Based on the same ideas as the methods in the above embodiments, the apparatus provided by the present invention can implement the methods in the above embodiments.

[0052] The above combination Figure 8 The standardized management device in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The standardized management equipment in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0053] Figure 9 This is a schematic diagram of a standardized management device 900 provided in an embodiment of the present invention. The standardized management device 900 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 910 (e.g., one or more processors) and a memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) for storing application programs 933 or data 932. The memory 920 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the standardized management device 900. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute the series of instruction operations in the storage media 930 on the standardized management device 900 to implement the steps of the standardized management method provided in the above-described method embodiments.

[0054] The standardized management device 900 may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 9 The standardized management equipment structure shown does not constitute a limitation on the standardized management equipment. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0055] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a standardized management method.

[0056] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0057] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0058] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A standardized management method, characterized in that, include: Obtain information to be reviewed, which includes at least one of store information, service desk information, and vehicle information; The information to be reviewed is subjected to data preprocessing and compliance review processing in sequence, generating corresponding review results and rectification suggestions corresponding to the review results. The review results include at least one of store review results, service desk review results, and vehicle review results, and the rectification suggestions include at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions. The system retrieves a preset periodic reminder mechanism, determines the reminder period based on the audit results and the periodic reminder mechanism, and integrates the audit results, rectification suggestions, and reminder period to generate audit feedback information. The review feedback information is stored in a pre-built data management library, and the review feedback information is monitored according to the reminder cycle.

2. The standardized management method according to claim 1, characterized in that, The step of obtaining the information to be reviewed includes at least one of store information, service desk information, and vehicle information, including: Obtain customer-uploaded data pending review and login account permission information, wherein the data pending review includes at least one of store information, service desk information, and vehicle information; Based on the login account permission information, the corresponding basic information of the end-service site is associated with the basic information, which includes the customer name and unique customer code; The data to be reviewed is linked and integrated with the basic information to generate information to be reviewed.

3. The standardized management method according to claim 1, characterized in that, The process of performing data preprocessing and compliance auditing on the information to be audited in sequence, generating corresponding audit results and rectification suggestions corresponding to the audit results, includes: When the information to be reviewed includes store information or service desk information, the information to be reviewed is compressed to obtain compressed review information. The compressed review information is denoised using a Gaussian filtering algorithm, and key visual features are extracted from the denoised compressed review information using image segmentation technology to obtain the structured feature data to be reviewed. Retrieve a pre-defined standardized parameter library corresponding to the information to be reviewed, and use a convolutional neural network model to compare the structured feature data to be reviewed with its corresponding standardized parameter library to output a similarity score. A target detection algorithm is used to identify defects in the denoised compressed audit information and generate defect detection results. The audit results are determined based on the similarity score and the defect detection results, and rectification suggestions are determined based on the defect detection results.

4. The standardized management method according to claim 3, characterized in that, When the information to be reviewed includes store information or service desk information, the review information is compressed to obtain compressed review information, including: When the information to be reviewed includes store information or service desk information, the information type of the information to be reviewed is determined, and the information type includes image type and video type. If the information to be reviewed is an image and includes multiple images, then a uniform resolution adjustment is performed on the information to be reviewed, and the information to be reviewed after uniform resolution adjustment is compressed according to preset compression parameters to obtain compressed review information. If the information to be reviewed is a video, then the information to be reviewed is cropped according to a preset duration threshold, and keyframe extraction is performed on the cropped information to generate compressed review information.

5. The standardized management method according to claim 1, characterized in that, The process of performing data preprocessing and compliance auditing on the information to be audited in sequence, generating corresponding audit results and rectification suggestions corresponding to the audit results, includes: When the information to be reviewed includes vehicle information, the vehicle information is compressed to obtain compressed vehicle information. The Gaussian filtering algorithm is used to denoise the compressed vehicle information, and the key visual features of the vehicle are extracted from the denoised compressed vehicle information by image segmentation technology. The key visual features of the vehicle include paint area, color parameters and vehicle body integrity. Collect vehicle type information corresponding to license plates, integrate the key visual features of the vehicle and vehicle type information to obtain vehicle structured feature data; The system retrieves a pre-defined standardized vehicle parameter library, uses a convolutional neural network model to compare and calculate the vehicle's structured feature data with the standardized vehicle parameter library, and outputs a matching score. A target detection algorithm is used to identify defects in the denoised compressed vehicle information to generate vehicle defect detection results. The vehicle review result is determined based on the fit score and the vehicle defect detection results, and vehicle rectification suggestions are determined based on the vehicle defect detection results.

6. The standardized management method according to claim 3, characterized in that, The preset periodic reminder mechanism is invoked, and the reminder period is determined based on the audit results and the periodic reminder mechanism. The audit results, rectification suggestions, and reminder period are integrated to generate audit feedback information, including: The preset periodic reminder mechanism is invoked to determine the corresponding reminder period and reminder node based on the audit results; When the audit result is qualified, the reminder period is 6 months, and the reminder node corresponding to the reminder period is 15 days before the expiration date; When the audit result is unqualified, a corresponding reminder period is determined based on the defect detection result, and the reminder node corresponding to the reminder period is the first day before the expiration of the validity period; The audit results, rectification suggestions, reminder cycles, and reminder points are integrated to generate audit feedback information.

7. The standardized management method according to claim 3, characterized in that, The step of storing the review feedback information in a pre-built data management database and monitoring the review feedback information according to the reminder cycle includes: The system integrates and correlates audit feedback information, structured feature data to be audited, and compressed audit information after noise reduction to generate an audit record package, which is then stored in a pre-built data management library. Based on the aforementioned reminder cycle, the flow status of the review feedback information is monitored, and the flow status includes pending rectification, in the process of rectification, rectified, and overdue. If the circulation status is overdue, an alarm message will be generated.

8. A standardized management device, characterized in that, include: The acquisition module is used to acquire information to be reviewed, which includes at least one of store information, service desk information, and vehicle information. The review module is used to sequentially perform data preprocessing and compliance review processing on the information to be reviewed, and generate corresponding review results and rectification suggestions corresponding to the review results. The review results include at least one of store review results, service desk review results, and vehicle review results, and the rectification suggestions include at least one of store rectification suggestions, service desk rectification suggestions, and vehicle rectification suggestions. The generation module is used to retrieve a preset periodic reminder mechanism, determine the reminder period based on the audit results and the periodic reminder mechanism, and integrate the audit results, rectification suggestions and reminder period to generate audit feedback information; The monitoring module is used to store the review feedback information in a pre-built data management database and monitor the review feedback information according to the reminder period.

9. A standardized management device, characterized in that, The standardized management device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the standardization management device to perform the steps of the standardization management method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the standardized management method as described in any one of claims 1-7.

Images