Video job record tracing system and method based on order information association

Abstract

The invention discloses a video job record tracing system and method based on order information association, and relates to the technical field of information management and video data processing, the system comprises an order information acquisition module used for acquiring order information data containing order identifiers and timestamps, and a video matching module used for matching the order information data with the timestamps. The matching module is used for matching a corresponding video job record fragment based on a timestamp range in order information, the association storage module is used for establishing an association mapping relation between an order identifier and the matched video job record fragment and storing the association mapping relation in a database, and the tracing query module is used for tracing the video job record fragment. The extraction module is used for extracting an associated video record from a database according to an input order query condition and generating a visual tracing interface; according to the video job record tracing system and method based on order information association, the order query video is traced based on the order association video.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information management and video data processing, and particularly relates to a video operation record tracing system and method based on order information association. BACKGROUND

[0002] In the fields of modern manufacturing, logistics and operation management, with the continuous improvement of informatization and visualization level, video recording of operation site has gradually become an important means to guarantee operation quality, support responsibility tracing and strengthen process supervision. The existing technology usually relies on fixed or mobile camera equipment to perform uninterrupted video collection on the whole operation process, and stores the shot video in a local server or a cloud platform. Although these video recordings cover the key links of the operation process, there are still significant deficiencies in the data organization and retrieval method.

[0003] Specifically, the current video recording system often lacks effective association mechanism with order information. The naming or storage structure of video data is usually managed based on time stamp or equipment number, while order information exists in independent business systems, and the two cannot form a corresponding relationship in data level or index. Therefore, when tracing the operation process of a specific order, the management personnel can only rely on the time line clue of the order, manual experience or vague operation record to manually search in a large amount of video data. This not only consumes time and effort, but also has the risk of inaccurate positioning, missing key information or false detection, and cannot meet the efficient, accurate and systematic tracing requirements. In addition, the existing system also lacks unified event labeling and process identification capability, and it is difficult to automatically segment, identify or mark the video clips, further increasing the complexity of the query and limiting the deep application of video data in quality management, risk control and process optimization. SUMMARY

[0004] The purpose of the present application is to provide a video operation record tracing system and method based on order information association, which traces order query video based on order associated video.

[0005] To achieve the above purpose, the present application provides the following technical scheme: a video operation record tracing system based on order information association, the system comprising: an order information acquisition module for acquiring order information data containing order identifier and time stamp; a video matching module for matching corresponding video operation record segments based on the time stamp range in the order information; The association storage module is configured to associate and store the order identifier and the matched video job record segment in the database, including extracting the last digit of the order number, extracting the last digit of each associated video segment ID, calculating the absolute value of the two digits, if the absolute value is not more than the set threshold, the video is taken as the main associated video segment, if the absolute value is higher than the set threshold, the video is taken as the secondary associated video segment; The trace query module is configured to extract the associated video record from the database according to the input order query condition and generate a visual trace interface, in which the main associated video segment is preferentially displayed, and the main associated video segment is preferentially displayed over the secondary associated video segment, including recording each browsed video segment after the user completes the order trace query and loads the video segment list, counting the number of browsed segments, reading the total number of order segments, calculating the browsing coverage, if the coverage is lower than the set coverage threshold, preferentially displaying the high-matching-degree segment that has not been browsed, if the coverage is greater than or equal to the set coverage threshold, preferentially displaying the browsed segment.

[0006] Preferably, the order information acquisition module acquires order information data containing an order identifier and a timestamp, including: The order information acquisition module acquires order information data containing an order identifier and a timestamp, including:

[0007] Preferably, the order information acquisition module acquires order information data containing an order identifier and a timestamp, and the specific formula for calculating the order information integrity ratio is: R = B / C; Wherein, R represents the order information integrity ratio, B represents the actual number of filled fields in each order, and C represents the number of fields that each order should contain.

[0008] Preferably, the video matching module matches the corresponding video job record segment based on the timestamp range in the order information, including dividing each video into a 10-second window, calculating the image change frame distance in each window, calculating the image change rate, setting the image change rate threshold, if the image change rate is higher than the set image change rate threshold, judging the video as an active action segment, finding all segments with image change rate higher than the set image change rate threshold within the order time range as target segments, and outputting the corresponding segment start and end time, video ID and speed score.

[0009] Preferably, the video matching module matches the corresponding video job record segment based on the timestamp range in the order information, and the specific formula for calculating the image change rate is: s = g / t; Among them, s represents the image change rate, g represents the frame distance of image change in the video, and t represents the length of the time period in which the change occurs.

[0010] Preferably, the traceability query module extracts the associated video records from the database according to the input order query conditions and generates a specific formula for calculating the coverage in the visual traceability interface: A=X / Y; Where A represents coverage, X represents the number of views, and Y represents the total number of order fragments.

[0011] Preferably, the video matching module calculates the frame distance of image change in the video based on the matching of the video operation record segment corresponding to the timestamp range in the order information. The specific method includes comparing the image grayscale values ​​of each pair of adjacent frames in the video at the same position pixel by pixel, calculating the difference in grayscale values, and taking its absolute value, summing up these difference values ​​of all pixel positions to obtain the image change intensity between the two frames, repeatedly applying this calculation to the entire video frame sequence, and accumulating the changes between all adjacent frames to obtain the total amount of image change in the entire video within a time window, which is the frame distance of image change in the video.

[0012] Preferably, the visual tracing interface includes a video preview window, an order details area, a key frame snapshot area and a timeline label navigation component.

[0013] Preferably, the database adopts a primary key-secondary key structure, stores the primary associated video clips and order numbers in a primary table, stores the secondary associated video clips in a secondary table, and sets association weights for each.

[0014] A video operation record tracing method based on order information association adopts the video operation record tracing system based on order information association, the method comprising: S1. Obtain order information data including order identifier and timestamp; S2. Matching corresponding video operation record segments based on the timestamp range in the order information; S3. Establish an association mapping relationship between the order identifier and the matching video job record segment and store it in the database; S4. Extract related video records from the database according to the input order query conditions and generate a visual tracing interface.

[0015] It can be seen from the above technical solution that the present invention has the following beneficial effects: The video operation record tracing system and method based on order information association, through an order information acquisition module, order information data containing an order identifier and a timestamp are acquired, a video matching module matches corresponding video operation record segments based on the timestamp range in the order information, an association storage module establishes an association mapping relationship between the order identifier and the matched video operation record segments and stores them into a database, and a tracing query module extracts associated video records from the database according to input order query conditions and generates a visual tracing interface, effectively breaking through the data barrier between the business system and the video system, realizing fast positioning and reverse tracing from the order to the video, being able to quickly extract corresponding video segments from the database based on structured query conditions, significantly shortening the search time, reducing the risk of human error, being able to intuitively display video content related to the order, and being able to superimpose event labels, time axis navigation and other functions, improving the operation efficiency and management experience of users, realizing comprehensive tracing of key processes, operators and product quality, and being suitable for quality audit, abnormal analysis and after-sales tracing scenes, improving production transparency and risk control capability, and having strong universality and generalizability. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The system module connection diagram of the present application is shown in the figure. Figure 2 The method flowchart of the present application is shown in the figure. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application. Figure 1 As shown in the figure, the present application provides a video operation record tracing system based on order information association, characterized in that the system comprises: An order information acquisition module for acquiring order information data containing an order identifier and a timestamp; A video matching module for matching corresponding video operation record segments based on the timestamp range in the order information; An association storage module for establishing an association mapping relationship between the order identifier and the matched video operation record segments and storing them into a database, including extracting the last digit of the order number, extracting the last digit of each associated video segment ID, calculating the absolute value of the two digits, if the absolute value does not exceed the set threshold, taking the video as the main associated video segment, and if the absolute value is higher than the set threshold, taking the video as the secondary associated video segment; The trace query module is used for extracting associated video records from the database according to the input order query condition and generating a visual trace interface, in which the main associated video segment is preferentially displayed, and the main associated video segment is preferentially displayed compared with the secondary associated video segment, including recording each browsed video segment after the user completes the order trace query and loads the video segment list, counting the browsed quantity, reading the total segment quantity of the order, calculating the browsing coverage rate, preferentially displaying the high matching degree segment which has not been browsed if the coverage rate is lower than the set coverage rate threshold, and preferentially displaying the browsed segment if the coverage rate is greater than or equal to the set coverage rate threshold.

[0018] The system takes the order as the core information source, and collects data containing the order identifier and the corresponding timestamp through the order information acquisition module. The video matching module takes the timestamp as the matching basis to retrieve and filter the segments in the video job record that match the time period, and realizes preliminary association. Then, the associated storage module further refines the matching result, compares the last digit of the order number and the video segment ID, calculates the absolute value to judge the matching strength. If the absolute value is less than or equal to the threshold, it is marked as the main associated video segment; if it is greater than the threshold, it is stored as a secondary associated segment, so as to build a structured database mapping. The trace query module calls the database data after the user inputs the order information, displays the video record in the visual interface, and counts the proportion of the browsed segment quantity to the total segment quantity, and adjusts the display strategy through the set coverage rate threshold: when the coverage rate is insufficient, the system recommends the video which has not been browsed and has high matching degree, improves the traceability effect; when the coverage rate meets the standard, the user's visited segment is preferentially displayed, which improves the query efficiency and user experience.

[0019] The present application improves the traceability efficiency and matching accuracy of the job record by accurately matching and classifying the storage between the order identifier and the video record in multiple dimensions. The numerical similarity of the order number and the video ID tail number is used as a supplementary matching basis to enhance the rationality and reliability of the association result. At the same time, the visual query interface dynamically adjusts the display strategy to ensure that the user can quickly obtain the key record segment, especially when the coverage rate does not meet the standard, the user is preferentially guided to view the missed video, which is helpful for comprehensive traceability. In addition, recording the user's click behavior and calculating the browsing coverage rate effectively improves the interactivity and intelligent recommendation ability of the system, and optimizes the overall user experience.

[0020] The order information acquisition module acquires order information data containing order identifiers and timestamps, including: The number of fields contained in each order is set, the number of fields actually filled in each order is checked, the order information integrity ratio is calculated, if the order information integrity ratio is greater than or equal to the set threshold, the order information is transmitted to the video matching module, if the order information integrity ratio is less than the set threshold, it is marked as serious information blocking, and after supplementing or manual annotation, it is processed again.

[0021] In this implementation, the order information acquisition module not only collects basic order data but also implements an information integrity verification mechanism to ensure the quality of data passed to subsequent modules. The specific process is as follows: First, the system presets a "standard field count" that defines the standard number of fields that each valid order should contain, such as order number, customer name, start time, end time, and job type. Upon receiving each order record, the system automatically counts the number of fields actually filled in, compares this value with the standard value, and calculates an "order information completeness ratio," which is the ratio of the actual number of fields filled in to the total number of fields set. A completeness threshold (the default value can be 80%) is set to determine whether the order data is sufficiently complete for automatic video matching. When the ratio is greater than or equal to the threshold, the order data is essentially complete and the system passes it to the video matching module for further processing. If it is lower than the threshold, the system marks the order as "severely obscured" and suspends further processing until the missing fields are added or corrected through manual review and annotation. Matching can then resume. The threshold setting can be dynamically adjusted based on historical processing success rates. An initial recommended value is generally between 70% and 90%, determined by the fault tolerance requirements of the business scenario. This design effectively improves the stability of overall data processing and avoids matching errors or system anomalies caused by incomplete information.

[0022] By introducing a verification mechanism for order information integrity ratios, the present invention establishes a quality threshold at the initial stage of the system, effectively blocking low-quality or incomplete order data from flowing into downstream modules, avoiding ineffective waste of computing resources and mismatches. This strategy improves the system's robustness and data processing efficiency, while also leaving processing channels for subsequent data completion or manual annotation, enhancing the overall system's fault tolerance and intelligence level. Furthermore, as a key parameter, the integrity threshold offers excellent flexibility and configurability, allowing users to dynamically adjust it based on their industry application characteristics, facilitating its application in a variety of business scenarios.

[0023] The order information acquisition module obtains the order information data including the order identifier and timestamp and calculates the order information integrity ratio using the formula: R = B / C; Here, R represents the order information completeness ratio, B represents the number of fields actually filled in each order, and C represents the number of fields that each order should include.

[0024] The embodiment further defines the calculation formula of the integrity ratio on the basis of the aforementioned order information integrity assessment, to ensure that the system implementation has unified and repeatable standards. After receiving an order information, the system first loads the total number C of standard fields that the order should contain from a predefined configuration file or database, such as order number, customer information, start time, end time, job content, operator, etc. Then the system counts the number of non-empty fields filled in the order record, denoted as B. The integrity ratio R is calculated by the formula R = B / C, which is a precise real value (value range 0 to 1) reflecting the completeness of the current order data. The formula design is simple and clear, facilitating fast system implementation and subsequent module calling. The integrity ratio R is then compared with the set threshold to determine whether it has enough information to support video matching operations. This processing flow can be used as the first data quality inspection checkpoint after the order information enters the system, effectively ensuring the stability and consistency of the system.

[0025] By introducing the explicit ratio formula R = B / C, the system realizes a standardized and quantitative order information integrity assessment mechanism. This method has simple calculation logic and high execution efficiency, especially suitable for high-concurrency and high-data-volume business scenarios, avoiding errors introduced by subjective judgment. At the same time, this mechanism can be extended as part of the data quality monitoring system, performing real-time screening and statistical analysis on the input data set, providing a more stable data foundation for system operation. The explicit index calculation method also facilitates cross-module collaboration and problem positioning, allowing for tracing to specific field missing conditions when processing abnormalities occur, improving system maintainability.

[0026] The video matching module matches the corresponding video job record segment based on the timestamp range in the order information, including dividing each video into 10-second windows, calculating the image change frame distance in each window, calculating the image change rate, setting an image change rate threshold, and if the image change rate is higher than the set image change rate threshold, determining that the video is an active action segment. All segments with image change rates higher than the set image change rate threshold within the order time range are found as target segments, and the corresponding segment start and end times, video ID, and speed score are output.

[0027] In this embodiment, the video matching module divides the entire video into several equal-length units by constructing a time-domain sliding window mechanism, and sets the time length of each window to 10 seconds to ensure capturing fine-grained image changes within a short period. In each 10-second window, the system extracts consecutive image frames and calculates the inter-frame change distance, i.e., the sum of pixel differences or image histogram differences between adjacent image frames, to quantify the dynamic degree of video images. Subsequently, the system divides the total amount of frame distance changes per unit time by the number of frames to obtain the "image change rate" indicator. The system presets an image change rate threshold (which can be initially set to an average change frame distance exceeding a certain value, e.g., Δ>15 per second) to distinguish between static monitoring segments and "active segments" with obvious operations or actions. When the image change rate is higher than the threshold, the system determines that the video in the window is an action-intensive segment. On this basis, the system further filters all video segments within the timestamp range provided by the order that meet the condition of image change rate higher than the threshold, and selects them as target candidate segments. The output results include the start time, video ID, and "speed score" based on the rate of each segment, which can be used for subsequent sorting and display optimization to improve the accuracy and practicality of matching videos.

[0028] By introducing the image change rate analysis mechanism, this embodiment can quickly identify key action segments related to order processing behavior in massive video data, significantly improving the accuracy and context relevance of video matching. Compared with the timestamp matching method alone, this method increases the ability to distinguish video dynamic features, effectively filters irrelevant static images, and reduces user browsing interference. The introduction of the image change rate threshold provides an automated filtering standard, avoiding subjective judgment and improving system robustness. At the same time, the segmented output speed score also provides a quantitative basis for subsequent visualization display, which can support high-matching priority recommendation strategies, improve user traceability efficiency, and enhance system intelligence level.

[0029] The video matching module calculates the image change rate based on the timestamp range in the order information and the corresponding video job record segment. The specific formula is: s = g / t where s represents the image change rate, g represents the image change frame distance in the video, and t represents the length of the time period in which the change occurs.

[0030] In this embodiment, when the video matching module performs image change analysis on each divided 10-second video window, the formula s = g / t is further used to calculate the image change rate. Among them, the variable g represents the image change frame distance in the window, that is, the sum of the image difference values between all consecutive frames in the statistical window, which can be obtained by pixel-level difference, image histogram difference or score model based on structural similarity (SSIM). t represents the length of the time period in which the change occurs, that is, the fixed 10-second time length of the window in this embodiment, or the dynamic assignment according to the real change duration between frames in the deformation mode. The s value obtained by the formula can reflect the intensity of the image change, that is, the dynamic change intensity of the video per unit time. This rate is used for comparison with the image change rate threshold, as a basis for determining whether a certain time period belongs to an "active action segment". The system screens all video segments with s values higher than the set threshold as target segments, provides high-quality candidate content for the subsequent tracing module, and outputs the speed score as a matching degree sorting reference.

[0031] The introduction of this formula provides a standardized calculation path for the system's image dynamic analysis, making the image change rate quantifiable, repeatable, and comparable, significantly improving the accuracy of video screening and the consistency of system processing. After using the s = g / t model, the system can flexibly adjust the collection method or value method of g and t according to the rhythm of different operation processes, ensuring that the algorithm has strong adaptability and high processing precision. At the same time, this mechanism facilitates batch processing and high-concurrency operation, and is suitable for deployment in real-time video analysis environments, improving overall operation efficiency and performance.

[0032] The tracing query module extracts associated video records from the database according to the input order query condition and generates a visual tracing interface. The coverage rate is calculated according to the formula: A = X / Y; Where A represents the coverage rate, X represents the number of browsed segments, and Y represents the total number of order segments.

[0033] In this embodiment, the traceability query module records and evaluates the video browsing interaction of each user entering the order traceability interface in real time. The system defines the total number of all associated video segments corresponding to the order as Y, and tracks the user's click, play or preview behavior for each segment in real time in the database, and counts the number of segments X that have been actively accessed by the user. The system calculates the coverage A according to the formula A = X / Y, which is used as a quantitative indicator of the current user browsing depth. The coverage reflects the completeness of the user's order video data traceability, and is one of the key judgment bases for the system's intelligent recommendation strategy. Once calculated, the value will be compared with the preset coverage threshold. If A is lower than the threshold, the system will preferentially display high-matching video segments that the user has not yet accessed in the recommendation area, encouraging further traceability. Conversely, if A reaches or exceeds the threshold, the system will recommend more clicked segments, improving browsing efficiency and interaction experience.

[0034] The present embodiment calculates the coverage by a standardized formula, allowing user behavior to be quantitatively tracked, improving the system's understanding of user traceability behavior and response intelligence. Using the A = X / Y model, the system can accurately monitor the traceability progress, effectively avoiding the risk of incomplete evidence caused by missing segments. At the same time, the coverage value is linked to the recommendation strategy, allowing flexible switching between the "not seen first" and "traceback reinforcement" display strategies, improving the comprehensiveness and efficiency of user traceability operations. This method also facilitates statistical analysis and user behavior research for system administrators, providing a basis for subsequent system optimization and strategy iteration.

[0035] The video matching module matches the corresponding video job record segment based on the timestamp range in the order information. The specific method for calculating the image change frame distance in the video includes comparing the image gray values of each pair of adjacent frames in the video pixel by pixel, calculating the difference in gray values, taking the absolute value, and adding up all the difference values of the pixel positions to obtain the image change intensity between the two frames. This calculation is repeated for the entire video frame sequence, and the change amount between all adjacent frames is accumulated to obtain the total image change amount of the entire video within a time window, which is the image change frame distance in the video.

[0036] In this embodiment, the calculation of image change frame distance is an important basis for identifying video dynamics, and the core is to quantify the degree of visual change between adjacent image frames. The system first extracts the video segment frame by frame, and selects a pair of consecutive frames for processing. For each pair of adjacent frames, the system performs pixel-by-pixel gray difference analysis, that is, for each pixel position, the gray value (usually an integer between 0 and 255) is extracted, and then the absolute value of the gray difference between the same pixel in two frames is calculated. The system sums up these absolute values in the entire image range to obtain the "image change intensity" of the frame pair. Then, the system accumulates the change intensity between all adjacent frames within a set time window (such as 10 seconds) to obtain the image change frame distance g in that time period. This value is a key input parameter for subsequent image change rate calculation, reflecting the activity level of the video scene change in that time period, and facilitating subsequent judgment of whether it is an effective action segment.

[0037] This image change frame distance calculation method has high precision and high interpretability, and can capture subtle image changes in the video in detail, especially suitable for action recognition needs in industrial monitoring, warehouse operation, production operation and other scenarios. Through the pixel-by-pixel gray difference method, the system can avoid misjudgment caused by color change, noise interference or scene switching, and improve the accuracy of active segment identification. At the same time, this method does not rely on complex models and is suitable for efficient operation on edge devices and resource-constrained platforms, achieving fast response. The frame distance value as a quantitative indicator is also convenient for subsequent rule threshold setting and algorithm expansion, and has good universality and compatibility.

[0038] The visual trace interface includes a video preview window, an order details area, a key frame snapshot area, and a time axis label navigation component. In this embodiment, the visual trace interface is constructed as a multi-region, multi-module collaborative information display platform for intuitively displaying corresponding video operation records and related information when a user queries an order. The main interface is the "video preview window", which supports continuous playback, pause, drag and speed operation of the matching segment, and provides visual dynamic playback function; below or beside it is the "order details area" for displaying the basic data of the current order, such as order number, customer name, time range, operation type, etc. Field, to ensure that the user has a clear grasp of the order information during the trace process. The key frame snapshot area is composed of a sequence of static thumbnail images based on pre-extracted representative frames, which facilitates quick browsing and positioning of key segments by users; the generation of key frames can be based on image change rate analysis, action recognition or model selection technology. The time axis label navigation component constructs a logical index structure of the video segment, labels key events, user browsing records, system recommended segments, etc. as graphical labels, and users can quickly jump to the target segment position by clicking the label, thereby constructing an efficient "non-linear" trace experience.

[0039] The interface design significantly improves the visualization and interaction efficiency of order traceability operation. The video preview window provides an immersive job review environment, and the key frame snapshot area greatly reduces the positioning time of the user and lowers the video review threshold. The timeline label navigation mechanism visualizes the user interaction behavior and system recognition logic, forming a path-based browsing track, while improving the search flexibility and operation convenience. The information in the order details area and the video area is displayed together, which helps users maintain information integrity and context association during the viewing process, enhancing the overall system user experience and task efficiency.

[0040] The database adopts a primary-secondary key structure, storing the primary associated video segments and order numbers in the primary table, storing the secondary associated video segments in the secondary table, and setting the associated weights respectively. In this embodiment, in order to efficiently manage and store the multi-level association relationship between the video and the order, the system adopts a data modeling method of primary-secondary key structure. In the database design, the system constructs two logically associated data tables: the primary table and the secondary table. The primary table takes the order number as the primary key field (PrimaryKey), and establishes a one-to-many primary association relationship with the video segments that have a high matching degree (i.e. the image change frame distance and the order tail number difference are lower than the threshold), stores the start and end time, video ID and "primary association weight" calculated by the system for such segments; the secondary table takes the combination of order number and video segment ID as the secondary key (Composite Key), and is used to store "secondary associated segments" with relatively low matching degree but still with reference value, and set a lower "secondary association weight" for them separately. The primary-secondary table structure ensures data integrity through primary key and foreign key association, while supporting fast indexing and batch querying, so that the system can retrieve primary segments and secondary segments respectively during query, recommendation and display, and realize differentiated data utilization strategy.

[0041] The data management method of this primary-secondary table structure improves the flexibility and efficiency of the system in storing structured video association data. Through primary-secondary separation management, the system can preferentially access the primary table when processing high-precision queries or data analysis tasks, reducing the computational burden; while in the need to cover comprehensive scenarios, it combines the secondary table to provide more rich background information, ensuring data integrity. The introduction of the association weight mechanism further strengthens the quantitative management of the importance of segments by the system, facilitating subsequent sorting and display, recommendation logic and statistical analysis function calls, enhancing the system's response ability and customization level to user needs.

[0042] The order query conditions include order number, order creation time, job type, operator number and multiple fields. In the embodiment, the system supports multi-field combined query to improve the search accuracy and flexibility when performing order trace operation. When the user inputs the order query conditions in the visual trace interface, the user can quickly locate the unique order record by inputting the "order number"; the user can also use the "order creation time" as a time filtering condition to filter all orders within a specified time period, which is suitable for batch query and historical trace scenarios. The "job type" field is used to distinguish different business processes, such as sorting, boxing, and code scanning, which helps the system accurately match the corresponding video segments and key frame feature templates. The "operator number" field introduces the personnel dimension, so that the system can trace the job records by person, which is especially suitable for multi-person collaboration or responsibility attribution analysis scenarios. During the system query execution process, all input fields will be converted into SQL or equivalent semantic query conditions, which are passed to the database layer for matching and filtering, and combined with the timestamp and video association rules to retrieve the corresponding video segments and related data for display.

[0043] The embodiment significantly improves the query flexibility and scenario adaptability of the order trace system by expanding the dimensions of the query conditions. Different fields can be used individually or in combination to support users to accurately locate from the order number, and to support fuzzy search from multiple dimensions such as time, person, and job type, effectively covering various business needs from abnormal troubleshooting to process auditing. At the same time, the structured field input simplifies the operation process, improves the user experience and system response efficiency, and provides strong support for large-scale order data management and accurate traceability.

[0044] As shown in Figure 2 , a video job record trace method based on order information association is also provided, which uses the video job record trace system based on order information association. The method comprises: S1, obtaining order information data containing order identifier and timestamp; S2, matching the corresponding video job record segment based on the timestamp range in the order information; S3, establishing an association mapping relationship between the order identifier and the matched video job record segment and storing it in the database; S4, extracting the associated video record from the database according to the input order query condition and generating a visual trace interface.

[0045] This method, centered on order information, completes the entire process from data collection to video association and display through four steps. In step S1, the system uses the order information acquisition module to collect structured order data containing unique identifiers (such as order numbers) and timestamps (creation time, operation time, etc.), and verifies its integrity to filter out incomplete or erroneous records. In step S2, a timestamp range matching algorithm is used to filter all video clips covering that time range from the video database. Combined with image change rate analysis, active action clips are identified as candidate matches. In step S3, the system maps the matching results to the orders, using a primary-secondary key structure to record primary and secondary associated clips, assigning association weights, and constructing a hierarchical data mapping model. In step S4, the system receives user-entered query criteria (such as order number, job type, operator, etc.), extracts matching videos from the database, and constructs an interactive visual traceability interface using an integrated video preview window, order details area, keyframe snapshot area, and timeline navigation components, supporting quick browsing, jumps, and information comparison.

[0046] This method, through a step-by-step process design, makes the order-to-video traceability logic clearer and more standardized, with good operability and system feasibility. Integrating key technical modules such as dynamic image analysis, data integrity assessment, multi-dimensional query screening, and visual display, it significantly improves the efficiency and accuracy of order anomaly tracing, responsibility determination, and process review. The primary-secondary association and weighting mechanism enhances the expressive power of the system's data model, facilitating the subsequent expansion of statistical analysis, intelligent recommendation, and other functions. The entire method is applicable to multiple scenarios, including logistics, manufacturing, e-commerce, and quality control, and contributes to the construction of an intelligent, data-driven process management system.

[0047] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A video operation record tracing system based on order information association, characterized in that: The system comprises: An order information acquisition module is used to obtain order information data including an order identifier and a timestamp; A video matching module is used to match corresponding video job record segments based on the timestamp range in the order information; An association storage module is used to establish an association mapping relationship between the order identifier and the matching video operation record segment and store it in a database, including extracting the last digit of the order number, extracting the last digit of each associated video segment ID, calculating the absolute value of the two numbers, and if the absolute value does not exceed a set threshold, the video is used as the primary associated video segment; if the absolute value is higher than the set threshold, the video is used as the secondary associated video segment; The tracing query module is used to extract related video records from the database according to the input order query conditions and generate a visual tracing interface. The main related video clips are displayed preferentially in the interface, and the main related video clips are displayed preferentially over the secondary related video clips. After the user completes the order tracing query and loads the video clip list, the module records the video clips browsed each time, counts the number of views, reads the total number of order clips, calculates the browsing coverage rate, and if the coverage rate is lower than the set coverage rate threshold, gives priority to displaying the unbrowsed high-matching clips; if the coverage rate is greater than or equal to the set coverage rate threshold, gives priority to displaying the viewed clips.

2. The video operation record tracing system based on order information association according to claim 1 is characterized by: The order information acquisition module acquires order information data including an order identifier and a timestamp, including: Set the number of fields that each order should contain, check the actual number of fields filled in each order, calculate the order information completeness ratio, and if the order information completeness ratio is greater than or equal to the set threshold, pass the order information to the video matching module. If the order information completeness ratio is less than the set threshold, mark it as severely blocked information and wait for supplementation or manual marking before processing.

3. The video operation record tracing system based on order information association according to claim 2 is characterized by: The specific formula for calculating the order information integrity ratio from the order information data containing the order identifier and timestamp obtained by the order information acquisition module is: R = B / C; Here, R represents the order information completeness ratio, B represents the number of fields actually filled in each order, and C represents the number of fields that each order should include.

4. The video operation record tracing system based on order information association according to claim 1 is characterized by: The video matching module matches the corresponding video operation record segments based on the timestamp range in the order information, including dividing each video into 10-second windows, calculating the frame distance of image change in the video within each window, calculating the image change rate, setting the image change rate threshold, and if the image change rate is higher than the set image change rate threshold, judging the video as an active action segment, searching for all segments with image change rates higher than the set image change rate threshold within the order time range as target segments, and outputting the corresponding segment start and end time, video ID, and speed score.

5. The video operation record tracing system based on order information association according to claim 4 is characterized by: The video matching module calculates the image change rate in the video operation record segment corresponding to the timestamp range in the order information using the following formula: s = g / t; Among them, s represents the image change rate, g represents the frame distance of image change in the video, and t represents the length of the time period in which the change occurs.

6. The video operation record tracing system based on order information association according to claim 1 is characterized by: The traceability query module extracts the associated video records from the database according to the input order query conditions and generates a visual traceability interface. The specific formula for calculating coverage is: A=X / Y; Where A represents coverage, X represents the number of views, and Y represents the total number of order fragments.

7. The video operation record tracing system based on order information association according to claim 1 is characterized by: The video matching module calculates the frame distance of image change in the video in the corresponding video operation record segment based on the timestamp range in the order information. The specific method includes comparing the image grayscale values ​​of each pair of adjacent frames in the video at the same position pixel by pixel, calculating the difference in grayscale values, and taking its absolute value, summing up these difference values ​​at all pixel positions to obtain the image change intensity between the two frames, repeatedly applying this calculation to the entire video frame sequence, and accumulating the changes between all adjacent frames to obtain the total amount of image change in the entire video within a time window, which is the frame distance of image change in the video.

8. The video operation record tracing system based on order information association according to claim 1 is characterized by: The visual tracing interface includes a video preview window, an order details area, a key frame snapshot area, and a timeline label navigation component.

9. The video operation record tracing system based on order information association according to claim 1 is characterized by: The database adopts a primary key-secondary key structure, stores primary associated video clips and order numbers in a primary table, stores secondary associated video clips in a secondary table, and sets association weights for each.

10. A video operation record tracing method based on order information association, using the video operation record tracing system based on order information association according to any one of claims 1 to 9, characterized in that: The method comprises: S1. Obtain order information data including order identifier and timestamp; S2. Matching corresponding video operation record segments based on the timestamp range in the order information; S3. Establish an association mapping relationship between the order identifier and the matching video job record segment and store it in the database; S4. Extract related video records from the database according to the input order query conditions and generate a visual tracing interface.

Images