Traceable industrial worker integral whole process management method and system

Abstract

The invention is suitable for the technical field of manufacturing management, and provides a traceable industrial worker integral whole-process management method and system, and the method comprises the following steps: receiving an integral updating instruction which comprises a worker number, an integral increase and decrease type, and a verification picture; determining a basic integral according to an integral increase and decrease type, identifying the verification picture to determine environment information, calling an environment weight, and determining a work type weight and a historical performance weight based on a worker number; integral increase and decrease values are calculated based on the basic integral, the environment weight, the work type weight and the historical performance weight, and integral updating information is obtained; receiving a point exchange instruction, including exchanging a commodity, to obtain point exchange information; and generating a point record ledger based on the point updating information and the point exchange information, and analyzing the safety risk index of the worker based on the point record ledger. Accuracy and traceability of integral records can be ensured, and integral management is more comprehensive and objective by introducing multiple factors such as environment weight and work type weight.

Description

Technical Field

[0001] The present invention relates to the technical field of manufacturing management, and in particular to a traceable whole-process management method and system for industrial workers' points. Background Art

[0002] In modern industrial production, the management and incentive mechanism of industrial workers are of great significance to improving production efficiency and ensuring production safety. Traditional worker management methods often rely on manual records and evaluations, which is not only inefficient but also easily affected by human factors, resulting in management results that are not objective and fair. With the continuous development of information technology, digital and intelligent management methods have gradually become an industry trend.

[0003] However, there are still some problems in the field of point management for industrial workers. On the one hand, workers' point records often lack unified standards and processes, resulting in inaccurate and opaque point calculations. Factors such as workers' work performance, differences in job types, and working environment have not been fully considered, making it difficult for point management to fully reflect workers' actual contributions and risk levels. On the other hand, there is also a lack of effective supervision and records in the point redemption process, which is prone to problems such as point abuse and unfair redemption, affecting workers' enthusiasm and corporate management efficiency.

[0004] Therefore, there is a need to provide a traceable method and system for the entire process of industrial workers' points management, aiming to solve the above problems. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a traceable industrial worker points management method and system for the entire process, so as to solve the problems existing in the above-mentioned background technology.

[0006] The present invention is implemented as follows: a traceable industrial worker integral whole process management method, the method comprising the following steps: receiving a points update instruction, wherein the points update instruction includes a worker number, a points increase or decrease type, and a verification picture; Determine the basic points based on the points increase or decrease type, identify the verification image to determine the environmental information, retrieve the environmental weight, and determine the job type weight and historical performance weight based on the worker number; Calculate the points increase and decrease based on the basic points, environmental weight, job type weight and historical performance weight to obtain the points update information; Receiving a points redemption instruction, the points redemption instruction including redeeming goods, and obtaining points redemption information; A points record ledger is generated based on the points update information and the points redemption information, and the workers' safety risk index is analyzed based on the points record ledger.

[0007] Another object of the present invention is to provide a traceable industrial worker integral whole process management system, the system comprising: A worker points update module, used to receive a points update instruction, wherein the points update instruction includes a worker number, a points increase or decrease type, and a verification picture; The relevant weight determination module is used to determine the basic points according to the point increase or decrease type, identify the verification image to determine the environmental information, retrieve the environmental weight, and determine the job type weight and historical performance weight based on the worker number; The points update information module is used to calculate the points increase or decrease value based on the basic points, environmental weight, job type weight and historical performance weight to obtain the points update information; The points redemption information module is used to receive a points redemption instruction, wherein the points redemption instruction includes redeeming goods and obtaining points redemption information; The worker safety risk module is used to generate a points record ledger based on the points update information and the points redemption information, and analyze the worker's safety risk index based on the points record ledger.

[0008] Compared with the prior art, the present invention has the following beneficial effects: The present invention realizes the standardization and automation of point management by receiving point update instructions and point exchange instructions. The instructions include information such as worker number, point increase and decrease type, and verification picture, which ensures the accuracy and traceability of point records. In the point calculation process, not only the basic points are considered, but also multiple factors such as environmental weights, job type weights, and historical performance weights are introduced, making point management more comprehensive and objective, and able to more accurately reflect the actual contribution and risk level of workers. By recording and analyzing point information through digital means and generating a point record ledger, management efficiency is greatly improved. At the same time, the safety risk index of workers is analyzed based on the point record ledger, which provides strong support for the safety management of enterprises. Through the point exchange mechanism, workers can exchange goods with their own points. This positive incentive method helps to enhance workers' work enthusiasm and sense of belonging. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flowchart of a traceable method for managing the whole process of industrial workers’ points.

[0010] Figure 2 A flowchart for determining relevant weights in a traceable full-process management method for industrial workers' points.

[0011] Figure 3 The present invention is a flowchart for calculating the safety risk index in a traceable industrial worker integral whole-process management method.

[0012] Figure 4A flow chart for generating safety training information and work adjustment plans in a traceable integrated full-process management approach for industrial workers.

[0013] Figure 5 This is a structural diagram of a traceable industrial worker points management system for the entire process. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0015] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0016] like Figure 1 As shown, an embodiment of the present invention provides a traceable industrial worker points management method for the entire process, the method comprising the following steps: S100, receiving a points update instruction, wherein the points update instruction includes a worker number, a points increase or decrease type, and a verification picture; S200, determining basic points according to the point increase or decrease type, identifying the verification image to determine environmental information, retrieving environmental weights, and determining job type weights and historical performance weights based on worker numbers; S300, calculating the score increase or decrease value based on the basic score, the environment weight, the job type weight and the historical performance weight, and obtaining the score update information; S400, receiving a points redemption instruction, the points redemption instruction including redeeming goods, and obtaining points redemption information; S500, generating a points record ledger based on the points update information and the points redemption information, and analyzing the worker's safety risk index based on the points record ledger.

[0017] It should be noted that in the field of point management for industrial workers, workers' point records often lack unified standards and processes, resulting in inaccurate and opaque point calculations. Factors such as workers' work performance, differences in job types, and working environment have not been fully considered, making it difficult for point management to fully reflect workers' actual contributions and risk levels. In addition, the point exchange process also lacks effective supervision and records, which is prone to problems such as point abuse and unfair exchange, affecting workers' enthusiasm and the management efficiency of enterprises. The embodiments of the present invention are intended to solve the above problems.

[0018] In an embodiment of the present invention, a points management platform will be constructed, and points update instructions will be input through the points management platform. The points update instructions include the worker number, the points increase or decrease type and the verification picture. Each worker has a unique number, and the points increase or decrease type is set. Each points increase or decrease type corresponds to a basic point. Participation in safety training, discovery of safety hazards, compliance with safety operating procedures and other behaviors can increase points, while violations of safety regulations, safety accidents and other behaviors will deduct points. The verification picture refers to a picture that proves the occurrence of the corresponding behavior, thereby ensuring the accuracy and traceability of the points record.

[0019] After the points update instruction is input, the embodiment of the present invention determines the basic points according to the points increase or decrease type, identifies the verification picture to determine the environmental information, and retrieves the environmental weight. The environmental weight is a weight set according to the degree of danger of the working environment (high-risk environment weight> low-risk environment weight). The job type weight and historical performance weight are determined based on the worker number. The job type weight is a weight set according to the risk level of the job type (high-risk job type weight> low-risk job type weight). The historical performance weight is a weight set according to the worker's historical safety performance. This comprehensive consideration makes the points management more comprehensive and objective, and can more accurately reflect the workers' actual contributions and risk levels. Then, the points increase or decrease value can be calculated based on the basic points, environmental weight, job type weight and historical performance weight. The points increase or decrease value = basic points × job type weight × environmental weight × historical performance weight, and the points update information is obtained.

[0020] When workers need to redeem items, they enter the points redemption instruction, which includes redeeming goods, obtaining points redemption information, and generating a points record ledger based on the points update information and points redemption information. This positive incentive method helps to improve workers' work enthusiasm and sense of belonging. At the same time, the transparent points record and redemption process also helps to enhance workers' trust and satisfaction. Finally, the workers' safety risk index will be analyzed based on the points record ledger, providing strong support for the company's safety management.

[0021] like Figure 2 As shown, as a preferred embodiment of the present invention, the steps of identifying the verification picture to determine the environmental information, retrieving the environmental weight, and determining the job type weight and the historical performance weight based on the worker number specifically include: S201, extracting the background environment of the verification image, and inputting the background environment into an environment type library for feature matching; S202, outputting the matching environment information, and determining the environment weight according to the environment information; S203, determining the type of work information and the performance information of the most recent set time period according to the worker number, determining the type of work weight according to the type of work information, and determining the historical performance weight according to the performance information.

[0022] In an embodiment of the present invention, the background environment of the verification image is automatically extracted and input into the environment type library for feature matching. The environment type library is established in advance and contains several environment types, each of which corresponds to environment characteristics and environment weights. After feature matching, the matching environment information is output and the corresponding environment weight is determined. In addition, the type of work information and the performance information of the most recent set time period are automatically determined based on the worker number, the type of work weight is determined based on the type of work information, and the historical performance weight is determined based on the performance information, for example, the most recent set time period is the most recent month.

[0023] As a preferred embodiment of the present invention, the step of extracting the background environment of the verification image and inputting the background environment into the environment type library for feature matching specifically includes: The verification images are processed using a multi-scale and multi-modal visual feature fusion method to obtain basic visual characteristics containing multi-scale information and multi-modal visual information; Perform semantic segmentation on the verification image based on deep learning semantic analysis to obtain a semantic segmentation map containing high-level semantic information; The semantic segmentation map is encoded into a semantic feature vector, and the corresponding first fusion weights are assigned to the basic visual characteristics and the semantic feature vector according to the complexity of the current environment and the accuracy of historical matching, and weighted fusion is performed to obtain a multi-dimensional environmental feature set; Establish an environment type library, which contains information of multiple environment types. Each environment type corresponds to a set of feature descriptions and an environment weight value. The environment type library is used as a standard template. The feature set of each environment type in the environment type library is matched with the multi-dimensional environment feature set. During the matching process, a matching weight is assigned to each type of feature according to its historical performance accuracy and weighted calculation is performed to obtain a weighted similarity score corresponding to each environment type. Compare the weighted similarity scores of all environment types and select the environment type with the highest score as the best environment match result corresponding to the verification image.

[0024] In the embodiments of the present invention, traditional image features (such as color, texture, and edge) and semantic features based on deep learning (such as environmental semantic segmentation) are combined to make up for the limitations of a single feature pattern. Traditional features ensure the capture of low-level visual information, while deep semantic features enhance the understanding of high-level semantics of the environment. The fusion of the two makes the expression of environmental background information more comprehensive, significantly improving the accuracy and robustness of recognition.

[0025] In addition, for different feature modes, the weights are dynamically adjusted according to the accuracy of historical matching and the complexity of the environment to ensure that the features with better performance are used first in different application scenarios. This dynamic adjustment mechanism enables the system to adapt to various complex environments, reduce misjudgments caused by the failure of a single feature, and enhance overall stability and reliability. The pre-built environment type library contains a variety of typical environment categories and their feature descriptions to form a standardized comparison template. Each environment type in the library corresponds to a feature set and risk weight, providing a unified basis for subsequent integral calculations. The library ensures the standardization and consistency of the matching process, which is helpful for data comparison and analysis across time and location. Finally, the weighted similarity score is used to comprehensively analyze the matching of each feature mode, taking into account the actual role of different features in environmental identification, and avoiding the excessive influence of any single feature on the results. This flexible matching strategy improves the fault tolerance of environmental identification, so that even if some features are disturbed, the overall identification is still accurate.

[0026] As a preferred embodiment of the present invention, the verification image is processed by a multi-scale multi-modal visual feature fusion method to obtain basic visual characteristics including multi-scale information and multi-modal visual information, specifically comprising: Convert the input verification image into multiple color space channels and use each color channel as an independent input image; Using color histogram statistics, the number of color histogram bins is set according to the complexity of color information of the input image, and normalized to obtain the normalized color histogram vector corresponding to each color channel; Set the number of scale layers, perform Gaussian blur and smooth sampling on the verification images in sequence according to the number of scale layers, gradually reduce the image resolution to form a multi-layer image pyramid, and obtain an image set of multiple scale layers; Apply a multi-directional texture filter to each scale layer image to obtain a multi-directional filter response map of multiple scale layers; Apply a nonlinear activation function to the multi-directional filter response map of each scale layer to obtain an activated filter response map; The activated filter response map is divided into a number of local grid units, and the local statistical features are calculated in each local grid unit to obtain the local statistical feature vector corresponding to each scale and direction; The local statistical feature vectors corresponding to all scales and directions are concatenated in sequence to form a complete set of texture feature vectors; Apply the gradient operator to each scale layer image to calculate the gradient components in the horizontal and vertical directions, and obtain the gradient magnitude map and gradient direction map corresponding to each scale layer; The gradient direction is divided into a predetermined number of angle intervals, and the gradient amplitude map of each scale layer image and the gradient direction map in each angle interval region are counted to obtain the directional gradient histogram feature vector of each scale layer; The sum of the overall gradient strength of each scale layer is calculated according to the gradient amplitude map corresponding to each scale layer, and the weight value of each scale layer is obtained after normalization; The directional gradient histogram feature vector of each scale layer is weighted with the weight value of each scale layer, and the weighted results are spliced ​​to obtain comprehensive multi-scale edge features; The gradient magnitude map and gradient direction map of the current scale layer are divided into several local grid units. In each local grid unit, the sum of the gradient magnitudes of different gradient directions is counted to obtain the directional gradient histogram of all local units of the current scale layer. The directional gradient histograms of all local units of the current scale layer are weighted by the total edge strength index weight of the current scale layer to obtain the weighted edge feature vector of the current scale layer; The weighted edge feature vectors of different scale layers are concatenated to form an overall edge feature vector containing multi-scale information, and a complete multi-scale weighted edge feature vector is output.

[0027] On the verification image, the variance of the pixel grayscale value in each window is calculated in units of a fixed-size sliding window to form a local variance matrix; Statistically verify the probability distribution of the gray level of the image and obtain the overall entropy value; The local variance matrix is ​​normalized and fused with the overall entropy value to obtain the local importance weight matrix; The normalized color histogram vector corresponding to each color channel, the complete texture feature vector set, and the comprehensive multi-scale edge feature are assigned an initial second fusion weight for weighted fusion to obtain a current fused multi-modal visual feature vector; The self-supervised loss feedback mechanism is used to iteratively adjust the second fusion weight to obtain the final fusion weight vector and fusion feature vector.

[0028] In the embodiment of the present invention, three types of visual features, color, texture and edge, are combined, and each type of feature is processed by multi-scale, multi-directional and statistical coding, which can capture the rich information in the environmental image in detail. In particular, the multi-space conversion and adaptive binning strategy of the color channel effectively improve the expression ability of color details and adapt to the diverse and complex environmental color changes. Through spatial pyramid hierarchical coding combined with local weight weighting, it is possible to take into account both global and local features, retain the spatial structure information of the environmental image, thereby improving the discriminative power of features and the accuracy of environmental matching. The fusion process uses a self-supervised iterative mechanism to dynamically adjust the weights of each feature and output a more robust comprehensive visual feature. Finally, the spatial pyramid structure and local weights are combined to achieve spatially sensitive feature coding, ensuring that the features contain both global information and retain key local details.

[0029] As a preferred embodiment of the present invention, the steps of performing semantic segmentation on the verification image based on semantic analysis of deep learning and obtaining a semantic segmentation map containing high-level semantic information specifically include: The image of each scale layer is extracted with a deep convolutional neural network to obtain a multi-dimensional feature representation corresponding to each scale layer; Calculate the amount of information represented by the multidimensional features corresponding to each scale layer, and assign corresponding information weights according to the amount of information; The multi-dimensional feature representations corresponding to all scale layers are weighted and summed according to the corresponding information weights to obtain the fused multi-scale features; The fused multi-scale features are processed by the spatial attention mechanism and the channel attention mechanism respectively to obtain the spatial attention matrix and the channel attention matrix respectively; The spatial attention matrix and the channel attention matrix are fused by element-by-element multiplication, and the fused multi-scale features are weighted by the fusion result to obtain the final weighted feature map; The final weighted feature map is fed into a deep convolutional decoder network to convert the high-dimensional feature map into a semantic prediction map that matches the size of the verification image, and a preliminary semantic segmentation probability map is obtained. Based on the preliminary semantic segmentation probability map, self-supervision is used to assist in optimizing the learnable parameters. After the optimization is completed, the optimized semantic segmentation probability map is output; Calculate the gradient information of the optimized semantic segmentation probability map to obtain a gradient map; Input the gradient map into the pre-trained boundary refinement network to increase the boundary and obtain the boundary enhancement map; The boundary enhancement map is weighted according to the proportional coefficient and superimposed on the optimized semantic segmentation probability map to enhance the category confidence of the boundary area, and the semantic segmentation probability map after boundary refinement and fusion is obtained.

[0030] In an embodiment of the present invention, by performing multi-scale processing on the verification image and dynamically adjusting the weights of the features at each scale, the scheme can capture multi-level information from coarse to fine and from macro to micro. This multi-scale fusion effectively compensates for the shortcomings of single-scale information, allowing the model to understand the image content more comprehensively and improve the accuracy and robustness of semantic segmentation. In addition, the dual attention mechanism weights the features of spatial position and channel dimensions, effectively highlighting semantic key areas and important feature channels, and suppressing irrelevant or interfering information. This greatly improves the model's ability to perceive the target semantics and enhances its ability to distinguish subtle semantic differences in complex environments.

[0031] In addition, self-supervised auxiliary training and boundary refinement are introduced. The introduction of a self-supervised mechanism uses data enhancement without relying on additional annotations, and guides the model to learn transformation-invariant semantic features through consistency constraints, thereby enhancing the adaptability of the model in unseen environments and complex scenarios. By processing the gradient information of the semantic probability map and enhancing the boundaries, the clarity and accuracy of the segmentation boundaries are significantly improved, and the blurring and confusion of category boundaries are reduced. This refinement process ensures the refinement of the segmentation results, which facilitates the subsequent accurate extraction and matching of environmental features.

[0032] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of analyzing the safety risk index of workers based on the score record ledger specifically includes: S501, calculate the worker's safety risk index RI, RI= ; S502, when the safety risk index RI is greater than the risk threshold, calling the corresponding point reduction type; S503: Analyze the points reduction type and generate safety training information and a work adjustment plan.

[0033] In the embodiment of the present invention, the safety risk index can reflect whether a worker has a certain risk. The calculation formula of the safety risk index RI is: RI= , represents the deduction of points for the i-th reduction in points, It represents the bonus points of the jth integral increase, TD is the time decay factor, TD= , λ is the attenuation coefficient, t is the time from the deduction or addition of points to the present, n and m represent the total number of points reduction and the total number of points increase in the recent set time period. Then the RI will be judged. When the safety risk index RI is greater than the risk threshold, it means that there is a certain risk. The corresponding point reduction type will be retrieved, and the point reduction type will be automatically analyzed to generate safety training information and work adjustment plan, so as to provide safety training to the worker in time and transfer him to a job with a low safety index.

[0034] As a preferred embodiment of the present invention, the step of analyzing the points reduction type and generating safety training information and a work adjustment plan specifically includes: According to the amount of deduction points, all points reduction types are sorted, and the points reduction types ranked in the top N positions are selected to obtain a number of points reduction type sets; Given a job safety database, based on the job information, retrieve a set of safety accident types related to the corresponding job information from the job safety database to obtain a list of safety accident types corresponding to the job; Matching and judging the set of points reduction types with the list of safety accident types corresponding to the types of work, and obtaining a matching result between the points reduction type and the safety accident type; Retrieving the safety training plan and key training content for the matching result from the safety training content library to obtain a safety training information set for the worker; Extracting unqualified safety items related to the list of safety accident types corresponding to the types of work from the work safety database to obtain a list of unqualified safety items; A set of work types that meet the standards on the list of unqualified safety items is selected from the work type safety database to obtain a set of work types that meet the standards; Based on the list of qualified jobs, combined with the current workers' skills, length of service and the company's employment needs, a specific work adjustment plan is generated.

[0035] In the embodiment of the present invention, by selecting the top-ranked points reduction types, focusing on the main safety hazards, avoiding data redundancy and noise interference, and ensuring the pertinence and efficiency of the analysis. Matching is performed in combination with the types of safety accidents related to the types of work, so that safety training and work adjustments are more targeted and effective, and irrelevant content interference is reduced. And according to the type of work information and deduction type of different workers, the safety training content and work adjustment plan are tailored to avoid "one-size-fits-all" management. Personalized plans are more in line with the actual risk status of workers, which helps to improve training effectiveness and job suitability, thereby effectively reducing the risk of safety accidents.

[0036] like Figure 4 As shown, as a preferred embodiment of the present invention, the steps of analyzing the points reduction type and generating safety training information and work adjustment plan specifically include: S5031, determine the points reduction type ranked in the top N according to the deduction points, and retrieve the safety accident type related to the corresponding worker's job type; S5032, matching the point reduction type with the safety accident type to determine safety training information and unqualified safety items; S5033, input unqualified safety items into the job safety database, determine qualified jobs, and generate a work adjustment plan.

[0037] In the embodiment of the present invention, the points reduction type ranked in the top N (for example, the top three) is determined based on the deduction points, and the safety accident type related to the corresponding worker's job type is retrieved. Each job type has a corresponding strongly related safety accident type. The points reduction type and the safety accident type are matched. When a certain type is successfully matched, targeted training is required for that type, and unqualified safety items are determined based on the corresponding safety accident type. Finally, the unqualified safety items are input into the job type safety library, the qualified jobs are determined, and a work adjustment plan is generated based on the qualified jobs.

[0038] As a preferred embodiment of the present invention, the method further includes: determining the total score of each worker based on the score record account, determining the incentive level based on the total score and constructing a red and black list of workers, displaying the red and black list of workers, with the red list displaying workers with higher scores and the black list displaying workers with lower scores or potential safety hazards, so as to encourage workers to actively participate in safety management. A safety trend chart for each worker will also be drawn based on the score record account, and the analysis results will be displayed to managers in an intuitive manner to help them better understand the safety status of the workers and take corresponding measures, and generate warning information based on the safety trend chart. For example, when the safety risk index of a worker continues to rise, a warning information will be generated.

[0039] As a preferred embodiment of the present invention, the method further includes: receiving the points exchange demand information input by the worker, the points exchange demand information includes the item name and the exchange points; summarizing and counting all the points exchange demand information to generate worker demand information, the worker demand information includes several item names, each item name corresponds to a demand point. In this way, the worker can exchange for more desired items, which is more humane.

[0040] like Figure 5 As shown, the embodiment of the present invention also provides a traceable industrial worker points whole process management system, the system comprising: A worker points updating module 100 is used to receive a points updating instruction, wherein the points updating instruction includes a worker number, a points increase or decrease type, and a verification picture; The relevant weight determination module 200 is used to determine the basic points according to the point increase or decrease type, identify the verification image to determine the environmental information, retrieve the environmental weight, and determine the job type weight and historical performance weight based on the worker number; The score update information module 300 is used to calculate the score increase or decrease value based on the basic score, the environment weight, the job type weight and the historical performance weight to obtain the score update information; The points redemption information module 400 is used to receive a points redemption instruction, wherein the points redemption instruction includes redeeming goods, and obtain points redemption information; The worker safety risk module 500 is used to generate a points record ledger based on the points update information and the points redemption information, and analyze the worker's safety risk index based on the points record ledger.

[0041] As a preferred embodiment of the present invention, the relevant weight determination module 200 includes: An environmental feature matching unit is used to extract the background environment of the verification image and input the background environment into the environment type library for feature matching; An environment weight determination unit, used to output matching environment information and determine the environment weight according to the environment information; The job performance weight unit is used to determine the job information and the performance information of the most recent set time period based on the worker number, determine the job weight based on the job information, and determine the historical performance weight based on the performance information.

[0042] As a preferred embodiment of the present invention, the worker safety risk module 500 includes: Safety risk index unit, used to calculate the worker's safety risk index RI, RI= , represents the deduction of points for the i-th reduction in points, It represents the bonus points of the jth integral increase, TD is the time decay factor, TD= , λ is the attenuation coefficient, t is the time from the deduction or addition of points to the present; An integral type retrieving unit, used for retrieving a corresponding integral reduction type when the safety risk index RI is greater than a risk threshold; The point type analysis unit is used to analyze the point reduction type and generate safety training information and work adjustment plan.

[0043] As a preferred embodiment of the present invention, the integral type analysis unit includes: The safety accident type subunit is used to determine the points reduction type ranked in the top N according to the deduction points, and retrieve the safety accident type related to the corresponding worker's job type; A safety training information subunit, used to match the point reduction type with the safety accident type to determine safety training information and unqualified safety items; The work adjustment plan subunit is used to input unqualified safety items into the work safety database, determine the qualified work types, and generate a work adjustment plan.

[0044] The above only describes in detail the preferred embodiments of the present invention, which is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0045] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0046] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0047] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A traceable whole-process management method for industrial workers' points, characterized by: The method comprises the following steps: receiving a points update instruction, wherein the points update instruction includes a worker number, a points increase or decrease type, and a verification picture; Determine the basic points based on the points increase or decrease type, identify the verification image to determine the environmental information, retrieve the environmental weight, and determine the job type weight and historical performance weight based on the worker number; Calculate the points increase or decrease based on the basic points, environmental weight, job type weight and historical performance weight to obtain the points update information; Receiving a points redemption instruction, the points redemption instruction including redeeming goods, and obtaining points redemption information; Generate a points record ledger based on the points update information and points redemption information, and analyze the workers' safety risk index based on the points record ledger; The steps of identifying the verification image to determine the environmental information, retrieving the environmental weight, and determining the job type weight and historical performance weight based on the worker number specifically include: Extract the background environment of the verification image, and input the background environment into the environment type library for feature matching; Outputting matching environmental information, and determining environmental weights according to the environmental information; The type of work information and the performance information of the most recent set time period are determined based on the worker number, the type of work weight is determined based on the type of work information, and the historical performance weight is determined based on the performance information.

2. The traceable industrial worker integral whole process management method according to claim 1 is characterized in that: The step of extracting the background environment of the verification image and inputting the background environment into the environment type library for feature matching specifically includes: The verification images are processed using a multi-scale and multi-modal visual feature fusion method to obtain basic visual characteristics containing multi-scale information and multi-modal visual information; Perform semantic segmentation on the verification image based on deep learning semantic analysis to obtain a semantic segmentation map containing high-level semantic information; The semantic segmentation map is encoded into a semantic feature vector, and the corresponding first fusion weights are assigned to the basic visual characteristics and the semantic feature vector according to the complexity of the current environment and the accuracy of historical matching, and weighted fusion is performed to obtain a multi-dimensional environmental feature set; Establish an environment type library, which contains information of multiple environment types. Each environment type corresponds to a set of feature descriptions and an environment weight value. The environment type library is used as a standard template. The feature set of each environment type in the environment type library is matched with the multi-dimensional environment feature set. During the matching process, a matching weight is assigned to each type of feature according to its historical performance accuracy and weighted calculation is performed to obtain a weighted similarity score corresponding to each environment type. Compare the weighted similarity scores of all environment types and select the environment type with the highest score as the best environment match result corresponding to the verification image.

3. The traceable industrial worker integral whole process management method according to claim 2 is characterized in that: The verification image is processed using a multi-scale and multi-modal visual feature fusion method to obtain basic visual characteristics containing multi-scale information and multi-modal visual information, specifically including: Convert the input verification image into multiple color space channels and use each color channel as an independent input image; Using color histogram statistics, the number of color histogram bins is set according to the complexity of color information of the input image and normalized to obtain the normalized color histogram vector corresponding to each color channel; Set the number of scale layers, perform Gaussian blur and smooth sampling on the verification images in sequence according to the number of scale layers, gradually reduce the image resolution to form a multi-layer image pyramid, and obtain a set of images at multiple scale layers; Extract edge features and texture features of each scale in a set of images at multiple scale layers to obtain texture feature vectors and multi-scale edge features; On the verification image, a sliding window of fixed size is used as a unit to calculate the variance of the pixel grayscale value in each window to form a local variance matrix; Statistically verify the probability distribution of the gray level of the image and obtain the overall entropy value; The local variance matrix is ​​normalized and fused with the overall entropy value to obtain the local importance weight matrix; The normalized color histogram vector, texture feature vector and multi-scale edge feature corresponding to each color channel are assigned an initial second fusion weight and weighted fused to obtain the current fused multimodal visual feature vector.

4. The traceable industrial worker integral whole process management method according to claim 3 is characterized in that: Extracting edge features and texture features of each scale in an image set of multiple scale layers to obtain texture feature vectors and multi-scale edge features specifically includes the following steps: Apply a multi-directional texture filter to each scale layer image to obtain a multi-directional filter response map of multiple scale layers; Apply a nonlinear activation function to the multi-directional filter response map of each scale layer to obtain an activated filter response map; The activated filter response map is divided into a number of local grid units, and the local statistical features are calculated in each local grid unit to obtain the local statistical feature vector corresponding to each scale and direction; The local statistical feature vectors corresponding to all scales and directions are concatenated in sequence to form a complete set of texture feature vectors; Apply the gradient operator to each scale layer image to calculate the gradient components in the horizontal and vertical directions, and obtain the gradient magnitude map and gradient direction map corresponding to each scale layer; The gradient direction is divided into a predetermined number of angle intervals, and the gradient amplitude map of each scale layer image and the gradient direction map in each angle interval region are counted to obtain the directional gradient histogram feature vector of each scale layer; The sum of the overall gradient strength of each scale layer is calculated according to the gradient amplitude map corresponding to each scale layer, and the weight value of each scale layer is obtained after normalization; The directional gradient histogram feature vector of each scale layer is weighted with the weight value of each scale layer, and the weighted results are spliced ​​to obtain a comprehensive multi-scale edge feature; The gradient magnitude map and gradient direction map of the current scale layer are divided into several local grid units. In each local grid unit, the sum of the gradient magnitudes of different gradient directions is counted to obtain the directional gradient histogram of all local units of the current scale layer. The directional gradient histograms of all local units of the current scale layer are weighted by the total edge strength index weight of the current scale layer to obtain the weighted edge feature vector of the current scale layer; The weighted edge feature vectors of different scale layers are concatenated to form an overall edge feature vector containing multi-scale information, and a complete multi-scale weighted edge feature vector is output.

5. The traceable industrial worker integral whole process management method according to claim 4 is characterized in that: The steps of performing semantic segmentation on the verification image based on semantic analysis of deep learning and obtaining a semantic segmentation map containing high-level semantic information include: The image of each scale layer is extracted with a deep convolutional neural network to obtain a multi-dimensional feature representation corresponding to each scale layer; Calculate the amount of information represented by the multidimensional features corresponding to each scale layer, and assign corresponding information weights according to the amount of information; The multi-dimensional features corresponding to all scale layers are represented and weighted summed according to the corresponding information weights to obtain the fused multi-scale features; The fused multi-scale features are processed by the spatial attention mechanism and the channel attention mechanism respectively to obtain the spatial attention matrix and the channel attention matrix respectively; The spatial attention matrix and the channel attention matrix are fused by element-by-element multiplication, and the fused multi-scale features are weighted by the fusion result to obtain the final weighted feature map; The final weighted feature map is fed into a deep convolutional decoder network to convert the high-dimensional feature map into a semantic prediction map that matches the size of the verification image, and a preliminary semantic segmentation probability map is obtained. Based on the preliminary semantic segmentation probability map, self-supervision is used to assist in optimizing the learnable parameters. After the optimization is completed, the optimized semantic segmentation probability map is output; Calculate the gradient information of the optimized semantic segmentation probability map to obtain a gradient map; Input the gradient map into the pre-trained boundary refinement network to increase the boundary and obtain the boundary enhancement map; The boundary enhancement map is weighted according to the proportional coefficient and superimposed on the optimized semantic segmentation probability map to enhance the category confidence of the boundary area, and the semantic segmentation probability map after boundary refinement and fusion is obtained.

6. The traceable industrial worker integral whole process management method according to claim 5 is characterized in that: The steps of analyzing the worker's safety risk index based on the score record ledger specifically include: Calculate the worker's safety risk index RI, RI= , represents the deduction of points for the i-th reduction in points, It represents the bonus points of the jth integral increase, TD is the time decay factor, TD= , λ is the attenuation coefficient, t is the time from the deduction or addition of points to the present; When the safety risk index RI is greater than the risk threshold, the corresponding point reduction type is retrieved; The points reduction types are analyzed to generate safety training information and work adjustment plans.

7. The traceable whole-process management method for industrial workers' points according to claim 6 is characterized in that: The step of analyzing the worker's safety risk index based on the score record ledger specifically includes: analyzing the score reduction type, generating safety training information and a work adjustment plan, specifically including: According to the amount of deduction points, all points reduction types are sorted, and the points reduction types ranked in the top N positions are selected to obtain a number of points reduction type sets; Given a job safety database, based on the job information, retrieve a set of safety accident types related to the corresponding job information from the job safety database to obtain a list of safety accident types corresponding to the job; Matching and judging the set of points reduction types with the list of safety accident types corresponding to the types of work, and obtaining a matching result between the points reduction type and the safety accident type; Retrieving the safety training plan and key training content for the matching result from the safety training content library to obtain a safety training information set for the worker; Extract unqualified safety items related to the list of safety accident types corresponding to the types of work from the work safety database to obtain a list of unqualified safety items; A set of work types that meet the standards on the list of unqualified safety items is selected from the work type safety database to obtain a set of work types that meet the standards; Based on the list of qualified jobs, combined with the current workers' skills, length of service and the company's employment needs, a specific work adjustment plan is generated.

8. The traceable whole-process management method for industrial workers' points according to claim 7 is characterized in that: The steps of analyzing the points reduction type and generating safety training information and work adjustment plans specifically include: Determine the points reduction type for the top N according to the deduction points, and retrieve the safety accident type related to the corresponding worker's job type; Matching the points reduction type with the safety accident type to determine safety training information and unqualified safety items; Input unqualified safety items into the job safety database, determine qualified jobs, and generate a work adjustment plan.

9. The traceable industrial worker integral whole process management method according to claim 8 is characterized in that: The method further comprises: Determine the total points of each worker based on the points record, determine the incentive level based on the total points, build a red and black list of workers, and display the red and black list of workers; Draw a safety trend chart for each worker based on the points record ledger, and generate early warning information based on the safety trend chart; Receiving point redemption demand information input by a worker, wherein the point redemption demand information includes an item name and redemption points; All points redemption demand information is summarized and counted to generate worker demand information, which includes several item names, each of which corresponds to a demand point.

10. A traceable industrial worker integral whole process management system, the system being applied to the traceable industrial worker integral whole process management method as claimed in any one of claims 1 to 9, characterized in that: The system comprises: A worker points update module, used to receive a points update instruction, wherein the points update instruction includes a worker number, a points increase or decrease type, and a verification picture; The relevant weight determination module is used to determine the basic points according to the point increase or decrease type, identify the verification image to determine the environmental information, retrieve the environmental weight, and determine the job type weight and historical performance weight based on the worker number; The points update information module is used to calculate the points increase or decrease value based on the basic points, environmental weight, job type weight and historical performance weight to obtain the points update information; The points redemption information module is used to receive a points redemption instruction, wherein the points redemption instruction includes redeeming goods and obtaining points redemption information; The worker safety risk module is used to generate a points record ledger based on the points update information and the points redemption information, and analyze the worker's safety risk index based on the points record ledger.

Images