A material barcode intelligent analysis and material association method, device and equipment

Abstract

The application provides a material barcode intelligent analysis method, comprising: acquiring a barcode of a material and identifying basic attribute information of the barcode; matching the basic attribute information with various types of analysis rules in a preset analysis library to determine whether there is a matched analysis rule, and using the matched analysis rule to analyze the barcode to obtain a material identification mark of the material; wherein the analysis library comprises at least two types of analysis rules; when it is determined that there is no matched analysis rule, triggering an unknown barcode intelligent analysis process to analyze the barcode to obtain the material identification mark of the material. Through the cooperation of various analysis rules in the preset analysis library and the intelligent analysis of unknown barcodes, different barcodes can be analyzed, which can adapt to current barcode analysis scenes of different manufacturers and different specifications, improve the universality and adaptability of barcode analysis, and be suitable for business scenarios such as warehouse material warehousing, delivery, inventory, and whole-process tracing.

Description

Technical Field

[0001] This application relates to the field of warehouse management technology, and in particular to a method, apparatus and equipment for intelligent parsing and associating material barcodes. Background Technology

[0002] In modern warehouse management, barcodes serve as unique identifiers for materials and are widely used in core warehousing operations such as material receiving, issuing, inventory counting, and end-to-end traceability. Currently, there is no unified barcode coding standard across different industries, manufacturers, and material categories, resulting in significant differences in barcode length, character type, encoding segments, and storage format among various material types.

[0003] Existing traditional barcode parsing systems all use a single fixed parsing rule to decode, which can only adapt to pre-configured barcodes of a single specification. This has two prominent technical defects: First, the system has not built a unified parsing library containing multiple types of parsing rules. When faced with heterogeneous barcodes from various industries such as manufacturing, retail, and chemicals, it cannot automatically match the corresponding parsing logic, and a large number of common non-standard barcodes fail to be parsed directly. Second, existing solutions lack an intelligent fallback parsing process for unknown barcodes. If the barcode to be parsed is not within the range of preset rules, the system directly determines that the parsing has failed. It cannot automatically cluster and deduce approximate rules to complete the parsing. It must rely on manual rule input or manual material identification, which significantly reduces the efficiency of automated warehouse circulation.

[0004] In summary, existing technologies are neither compatible with multiple types of known standard barcodes, nor do they have the capability for automated parsing of newly added unknown barcodes, resulting in extremely poor barcode parsing universality and adaptability. Therefore, how to achieve automatic matching and parsing of known barcodes of multiple specifications, while providing intelligent fallback parsing for unknown barcodes, and improving the adaptability of barcode parsing across all warehousing scenarios, is a pressing technical challenge that needs to be addressed in the field of warehousing barcode parsing. Summary of the Invention

[0005] In view of this, existing traditional barcode parsing systems all use a single fixed parsing rule to achieve decoding, which has two prominent technical defects: First, the system has not built a unified parsing library containing multiple types of parsing rules. When faced with heterogeneous barcodes from various industries such as manufacturing, retail, and chemicals, it cannot automatically match the corresponding parsing logic, and a large number of common non-standard barcodes fail to be parsed directly. Second, existing solutions lack an intelligent fallback parsing process for unknown barcodes. If the barcode to be parsed is not within the preset rule range, the system directly determines that the parsing has failed, and cannot automatically cluster and deduce approximate rules to complete the parsing. It must rely on manual rule input or manual material identification, which significantly reduces the efficiency of automated warehouse circulation. Based on the above defects, this application proposes a method, device, and equipment for intelligent parsing of material barcodes, which can simultaneously be compatible with multiple known heterogeneous barcodes and provide intelligent fallback parsing for unknown barcodes, effectively improving the barcode parsing adaptability and automation level of the entire warehouse scenario.

[0006] According to a first aspect of this application, a method for intelligent parsing of material barcodes is provided, comprising: Obtain the barcode of the material and identify the basic attribute information of the barcode; Based on the basic attribute information, the barcode is matched with various parsing rules in the preset parsing library to determine whether a matching parsing rule exists. If a matching parsing rule exists, it is used to parse the barcode to obtain the material identification identifier of the material. The parsing library includes at least two types of parsing rules. When it is determined that no matching parsing rule exists, the unknown barcode intelligent parsing process is triggered to parse the barcode and obtain the material identification identifier of the material.

[0007] In one possible implementation, the basic attribute information of the barcode includes at least one of the following: barcode length, character type in the barcode, and encoding format.

[0008] In one possible implementation, the basic attribute information is matched with various parsing rules in a preset parsing library. When determining whether a matching parsing rule exists, the parsing rule is found by traversing all the parsing rules in the parsing library to find a parsing rule that matches all the basic attribute information of the barcode.

[0009] In one possible implementation, parsing the barcode using the matching parsing rules to obtain the material identification identifier for the material includes: The parsing sequence of the current barcode is constructed based on the preset parsing step template in the matching parsing rules; The material identification identifier of the material is obtained by parsing the barcode according to the parsing sequence.

[0010] In one possible implementation, when parsing the barcode using an unknown barcode intelligent parsing process, the process includes: Extract the feature vectors corresponding to the basic attributes of the barcode. Cluster analysis is performed based on the feature vector and the parsing rules in the parsing library to determine whether there is a rule cluster with the highest similarity; if a rule cluster with the highest similarity exists, the parsing rules in that rule cluster are used to complete the parsing. If no rule cluster with the highest similarity is found, the current barcode is determined to be a completely new and unknown type, and a manual association and parsing process is triggered.

[0011] According to another aspect of this application, a material association method is also provided for material association based on the aforementioned intelligent material barcode parsing method, including: The material is acquired and it is determined whether the material has a barcode. When it is determined that the material has a barcode, the material barcode intelligent parsing method is used to parse the barcode to obtain the material identification mark. The barcode is then compared with the stored material information to locate the corresponding material and associate it with the material.

[0012] In one possible implementation, determining that the material does not have a barcode includes: Obtain the attribute information of the material; wherein the attribute information includes at least one of material code and batch number; Based on the attribute information, complete material information is retrieved from the stored material information, and a corresponding barcode is generated based on the retrieved material information; Associate the barcode with the currently searched material information.

[0013] According to a second aspect of this application, a smart barcode parsing device for materials is provided, comprising: The data acquisition module is used to acquire the barcode of the material and identify the basic attribute information of the barcode; The parsing rule parsing module is used to match the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists. If a matching parsing rule exists, the matching parsing rule is used to parse the barcode to obtain the material identification identifier of the material. The parsing library includes at least two types of parsing rules. The unknown barcode parsing module is used to trigger the unknown barcode intelligent parsing process to parse the barcode and obtain the material identification mark of the material when it is determined that no matching parsing rule exists.

[0014] According to another aspect of this application, a material association apparatus is also provided for performing material association based on the aforementioned intelligent material barcode parsing method, comprising: The barcode presence determination module is used to acquire materials and determine whether the materials have barcodes. When it is determined that the materials have barcodes, the material barcode intelligent parsing method is used to parse the barcodes to obtain the material identification identifier. The association module is used to compare the material identification mark with the stored material information, find the corresponding material, and associate the barcode with the material.

[0015] According to a third aspect of this application, a material association device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to perform the method described in the first aspect of this application.

[0016] This application provides a method for intelligent parsing of material barcodes, comprising: acquiring a material barcode and identifying the basic attribute information of the barcode; matching the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists; if a matching parsing rule exists, using the matching parsing rule to parse the barcode to obtain a material identification identifier for the material; wherein the parsing library includes at least two types of parsing rules; and when it is determined that no matching parsing rule exists, triggering an intelligent parsing process for unknown barcodes to parse the barcode to obtain a material identification identifier for the material. This application completes matching and parsing through a parsing library containing at least multiple types of parsing rules; when no matching rules are available, it automatically triggers an intelligent parsing process for unknown barcodes to complete decoding, forming a dual parsing path of "known rule matching + intelligent fallback for unknown barcodes," bringing the following beneficial effects: 1. By constructing a unified parsing library of multiple types of parsing rules, it can automatically match the corresponding parsing logic according to barcode length, character type, and encoding format, and is compatible with heterogeneous material barcodes from different industries, manufacturers, and specifications, solving the problem that traditional single rules can only adapt to a single format of barcodes; 2. An intelligent clustering parsing fallback process for unknown barcodes is added, extracting feature vectors from barcodes without matching rules in the library for cluster analysis, reusing the rule cluster with the highest similarity to attempt parsing, and only transferring completely new barcodes with no similarity to the manual process, greatly reducing the frequency of manual intervention and ensuring the continuous automated flow of warehouse inbound and outbound and inventory operations; 3. The system includes a material association workflow. For materials with barcodes, it automatically parses and matches material information to complete the binding. For materials without barcodes, it automatically generates a unique barcode based on material attributes and associates it with the material. It also incorporates a rule self-learning optimization mechanism, continuously accumulating new rules based on added barcode features, dynamically adjusting rule priorities, and iteratively improving the parsing library to consistently enhance barcode parsing accuracy and compatibility coverage. This application relies on a four-layer logical collaborative mechanism of "multi-rule library matching, intelligent backup for unknown barcodes, automatic material association, and rule self-learning optimization" to achieve intelligent parsing of all types of warehouse labels and rapid material binding, significantly improving the overall operational efficiency of warehouse management.

[0017] Other features and aspects of this application will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0018] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this application together with the specification and serve to explain the principles of this application.

[0019] Figure 1 A flowchart illustrating a material barcode intelligent parsing method according to an embodiment of this application is shown; Figure 2 A schematic flowchart of a material association method according to an embodiment of this application is shown; Figure 3 A schematic block diagram of a material barcode intelligent parsing device according to an embodiment of this application is shown; Figure 4 A schematic block diagram of a material barcode intelligent parsing device according to an embodiment of this application is shown; Figure 5 A schematic block diagram of a material association device according to an embodiment of this application is shown. Detailed Implementation

[0020] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0021] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0022] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.

[0023] <Method Implementation> Figure 1 This illustrates a method for intelligent parsing of material barcodes according to an embodiment of this application. For example... Figure 1 As shown, the method includes steps S1100-S1300: S1100, acquiring the barcode of the material and identifying the basic attribute information of the barcode; S1200, matching the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists, and if so, using the matching parsing rule to parse the barcode to obtain the material identification identifier of the material; wherein, the parsing library includes at least two types of parsing rules; S1300, when it is determined that no matching parsing rule exists, triggering an unknown barcode intelligent parsing process to parse the barcode to obtain the material identification identifier of the material.

[0024] This application parses the barcode to obtain its material identification identifier by searching for a matching parsing rule in a preset parsing rule library. Simultaneously, when no matching parsing rule is found in the library, an intelligent unknown barcode parsing method is used. This approach allows for the parsing of all barcodes during intelligent material barcode parsing, improving its adaptability. Furthermore, this parsing method is adaptable to barcode parsing scenarios from different manufacturers and with different specifications, simultaneously supporting multiple known heterogeneous barcodes and providing intelligent fallback parsing for unknown barcodes, thus enhancing the overall versatility of warehouse barcode parsing.

[0025] In one possible implementation, the material barcode is obtained by scanning it using a barcode scanning device. The scanned barcode data is then input into a preset rule engine, which stores feature models for various barcode formats. The basic attribute information of the barcode can be dynamically identified using the dual-channel hierarchical pattern matching and multi-dimensional joint feature extraction algorithm developed in this application. The identification process for the basic attribute information includes at least one of the following: hierarchical flow segmentation, coarse image matching, string feature extraction, feature fusion verification, and fallback error tolerance.

[0026] In one possible implementation, the specific process of using a dual-channel hierarchical pattern matching and multi-dimensional joint feature extraction algorithm to perform the identification of basic attribute information is shown below: First, a layered and split-flow operation is performed, that is, the data output from the scan is split into barcode images and raw strings to perform dual-channel parallel processing. Next, a coarse image matching operation is performed on the barcode image, and a multi-dimensional string feature extraction operation is performed on the original string. Specifically, during image cluster matching of the barcode image, a template coarse screening is completed by adaptively adjusting the matching threshold, outputting candidate barcode types. Template coarse screening refers to using a pre-set barcode visual template as a benchmark to perform multi-layer filtering and matching on the cropped barcode grayscale image based on contour, bar-space ratio, and code shape. It should be noted that barcode visual templates come from two sources: first, during the system initialization phase, standard one-dimensional and two-dimensional barcode visual sample templates are pre-generated according to various standard barcode specifications in the manufacturing, retail, and chemical industries and stored in the rule engine template library; second, during the system's self-learning process, newly parsed barcode images that have been manually confirmed are automatically added as new visual templates and stored in the template library after size normalization. The stored content of a single visual template includes at least one of the following: barcode code shape contour, standard bar-space width ratio, image reference size, template-bound exclusive barcode type identifier, and template basic matching similarity threshold.

[0027] In one possible implementation, the complete screening steps for candidate barcode types, achieved by adaptively adjusting the matching threshold during image cluster matching of the barcode image, are as follows: First, the barcode image is processed by grayscale conversion, noise reduction, and binarization to extract the effective rectangular region of the barcode's outer contour. Second, all barcode visual templates in the template library are loaded, and a sliding window traversal method is used to perform similarity matching with the current barcode image. Third, the matching threshold is adaptively and dynamically adjusted based on image imaging quality. Specifically, this includes: increasing the similarity threshold for blurred, reflective, or incomplete images; and decreasing the threshold for high-definition, complete barcodes. Fourth, all templates with similarity greater than the current dynamic threshold are selected, the barcode types bound to each template are aggregated, and a set of candidate barcode types is output as the template coarse screening result. It should be noted here that the candidate barcode type set is mainly used as a constraint in two subsequent steps: First, in the two-layer feature fusion verification stage, only the string feature judgment logic that matches the candidate barcode type is retained, and illegal character combinations not in the candidate set are eliminated to narrow the feature verification scope; Second, in the parsing library rule matching stage, the parsing library is pruned according to the candidate barcode type, and only a subset of rules bound to that type of barcode are taken out to participate in the matching, reducing invalid rule traversal and reducing computational overhead.

[0028] In one possible implementation, the multi-dimensional string feature extraction of the original string refers to parsing three main attributes—barcode length, character distribution, and encoding prefix—from the original string. The specific parsing process is as follows: First, the barcode length is parsed by removing leading and trailing spaces and invalid separator padding characters, and counting the total number of remaining valid characters to obtain the effective barcode length. Second, the character distribution is parsed by traversing the effective string character by character, counting the occurrence and interval positions of numbers, uppercase letters, lowercase letters, and special symbols to generate a character distribution feature vector. Third, the encoding prefix is ​​parsed by truncating the string to a fixed length according to industry-preset standard prefixes, extracting the first character segment of the string, and performing a hash match with various barcode standard prefix sets in a preset encoding prefix template library. If a match is successful, the corresponding encoding prefix is ​​marked; otherwise, it is marked as an unknown prefix. Fourth, the barcode length, character distribution feature vector, and encoding prefix identifier are integrated into a multi-dimensional string feature group for subsequent feature fusion calculations.

[0029] Based on the aforementioned template coarse matching results and string feature extraction results, a two-layer feature fusion verification operation is performed. This involves using a weighted fusion of image matching results and string multidimensional features, automatically adjusting the dimensional confidence weights in case of conflicts. The image matching results used in the weighted fusion calculation include the candidate barcode type set output from the image coarse screening and the image matching similarity score corresponding to each candidate template. The string multidimensional features include the barcode length, character distribution, and encoding prefix feature group obtained from the aforementioned analysis. Both modality outputs serve as the weighted fusion input. Based on the aforementioned confirmed data used for calculation, this application sets a dedicated weighted fusion calculation formula. When performing weighted fusion, the formula for calculating the comprehensive confidence score of a single barcode type is as follows: Score = Sim_img×Wimg + Sim_str×Wstr In the formula, Sim_img represents the highest matching similarity of the image template corresponding to the current barcode type, with a value range of 0 to 1; Sim_str represents the matching similarity between the current barcode type and the multidimensional features of the string, which is calculated by weighted summation of three features: length, character distribution, and encoding prefix, with a value range of 0 to 1; Score is the comprehensive confidence score of the barcode type. After fusion, the barcode type corresponding to the maximum Score among all candidate types is selected as the initial judgment result of the barcode's basic attributes; Wimg represents the pre-configured basic weight of the image modality, and Wimg is initially set to 0.4; Wstr represents the basic weight of the string text modality, and Wstr is initially set to 0.6.

[0030] It should also be noted that the dimension confidence weights are divided into image dimension confidence weights (Wimg) and string dimension confidence weights (Wstr), used to characterize the confidence level of the corresponding modal output results, with values ​​ranging from 0 to 1. A higher image dimension confidence weight indicates a clearer barcode image and higher image matching reliability, while a higher string dimension confidence weight indicates that the scanned string is free of incomplete or garbled characters and that the character features are more reliable. The initial acquisition method for dimension confidence weights is as follows: the system presets the basic weights Wimg=0.4 and Wstr=0.6, and simultaneously collects the scanned image quality and string integrity indicators in real time to dynamically adjust the initial weights. When the image is blurry, reflective, or incomplete, the image confidence weight is reduced; when the string is truncated, missing, or garbled, the string confidence weight is reduced.

[0031] In one possible implementation, the weighted fusion process includes an adaptive adjustment of weights in case of feature conflicts. The complete logic for automatically adjusting dimensional confidence weights in case of feature conflicts is as follows: First, a conflict condition is determined. If the candidate barcode type output from the image coarse screening is inconsistent with the barcode type derived from the multidimensional features of the string, a bimodal feature conflict is determined. Second, the original confidence scores of the image and string modalities are compared. The weight of the modality with higher confidence is kept unchanged, while the weight of the modality with lower confidence is attenuated and reduced. The attenuation coefficient ranges from 0.2 to 0.5, and the larger the difference in confidence between the two modalities, the greater the attenuation. Then, the adjusted new weights are substituted into the above comprehensive confidence score calculation formula to calculate the comprehensive confidence score of each candidate barcode type a second time. The barcode type with the highest confidence score after the second calculation is selected as the basic attribute information of the barcode after the bimodal feature fusion verification is completed. If a unified high-confidence barcode type cannot be obtained after the weight adjustment, the image matching constraint is discarded, and the basic attributes of the barcode are output only based on the full string features.

[0032] It should be noted that before performing the dual-layer feature fusion verification, it is necessary to determine whether the image coarse matching matches the template. If a match is found, the basic attribute information of the barcode is obtained by using the weighted fusion method described above. If there is no image matching template, the basic attribute information of the barcode is obtained by relying solely on the full string features to complete attribute recognition.

[0033] In one possible implementation, the basic attribute information includes at least one of the following: barcode length, character type, and encoding format. Compared to traditional fixed-rule recognition methods, this method uses a rule engine combined with contextual information for dynamic inference, enabling it to adapt to barcodes from different manufacturers and with different specifications, thus improving recognition compatibility and accuracy.

[0034] It should be noted that when identifying the length of a barcode, it is based on the length of the string being identified; when identifying the character type, it is based on the character types contained in the string; and when identifying the encoding format, it is identified through encoding features.

[0035] After recognizing the basic attribute information of the barcode, the system performs an operation that matches the basic attribute information with various parsing rules in the preset parsing library to select the optimal parsing rule.

[0036] In one possible implementation, the various parsing rules are preset according to the characteristics of each industry and the material type. Specifically, when presetting the various parsing rules according to industry and material type, the parsing rules can be preset based on the barcode setting rules corresponding to each type of material in each industry. Each type of parsing rule includes at least one of the following: barcode length rules, encoding segmentation rules, character type rules, and keyword matching rules.

[0037] In one possible implementation, when setting the barcode parsing rules for manufacturing parts and materials, the corresponding barcode setting rules include length setting rules and segmentation setting rules. The length setting rule is: the barcode length is 14 digits; the segmentation setting rule is: the first 8 digits of the barcode represent the material code, the middle 4 digits represent the batch number, and the last 2 digits represent the specification code. Therefore, one type of parsing rule can be preset as follows: the length rule satisfies the 14-digit barcode; the preset segmentation rule is "to divide the 14-digit barcode into segments of 8 digits, 4 digits, and 2 digits respectively, and each segment string represents the following rule: the first 8 digits are the material code, the middle 4 digits are the batch number, and the last 2 digits are the specification code".

[0038] In one possible implementation, when pre-setting barcode parsing rules for retail goods and materials, the corresponding barcode setting rules include: barcode length setting rules, barcode character type setting rules, and keyword setting rules. Specifically, the barcode length setting rule is a fixed length of 13 characters; the barcode character type setting rule is a combination of numbers and letters; and the keyword setting rule is that the barcode contains a keyword (such as a product category abbreviation). Thus, one set of parsing rules can be pre-set as follows: Length rule: barcode length is fixed at 13 characters; Character type rule: combination of numbers and letters; Keyword matching rule: contains a keyword, and the keyword is a product category abbreviation.

[0039] In one possible implementation, when pre-setting the parsing rules for hazardous material barcodes in the chemical industry, the corresponding barcode setting rules include: barcode length setting rules, segmentation setting rules, and keyword setting rules. Specifically, the barcode length setting rule is 16 digits; the segmentation setting rule is that the last 3 digits are the safety level code; and the keyword setting rule is that the barcode contains hazardous material identification keywords. Thus, one set of parsing rules can be pre-set as follows: Length rule: barcode length is 16 digits; Segmentation rule: 13 digits and 3 digits, with the last 3 digits identifying the material safety level code; Keyword matching rule: contains hazardous material identification keywords.

[0040] Following the above example, parsing rules corresponding to different material types in various industries can be preset, initial priorities can be configured for each parsing rule, and the preset parsing rules and their corresponding priorities can be stored in a database to obtain a pre-built parsing library. The initial priority is calculated based on three dimensions: business priority, matching efficiency, and recognition accuracy. The business priority weight is 0.4, the matching efficiency weight is 0.3, and the recognition accuracy weight is 0.3. The initial priority score is obtained by weighted summation. It should be noted that in this application, the business priority score is preset manually; the higher the preset rule is for high-frequency materials or core industries, the higher the business priority score. The matching efficiency score and recognition accuracy score are obtained by system statistics; that is, the faster the rule matching and the fewer the steps, the higher the matching efficiency score; the higher the historical parsing accuracy, the higher the recognition accuracy score. The initial priority can be expressed by the formula: S= 0.4 × S1 + 0.3 × S2 + 0.3 × S3 Where: S is the initial priority score of the rule; S1 is the business priority score; S2 is the matching efficiency score; and S3 is the recognition accuracy score.

[0041] Based on the parsing library constructed above, the operation of matching basic attribute information with various parsing rules in the parsing library is performed: the rules in the parsing library are sorted from high to low priority scores, and matched with the basic attribute information in turn, and the first rule that is a complete match is selected as the optimal parsing rule.

[0042] Based on the parsing library constructed above, the operation is performed to determine whether there is a matching parsing rule by matching the basic attribute information with various parsing rules in the parsing library.

[0043] In one possible implementation, when matching the basic attribute information with various parsing rules in a preset parsing library to determine if a matching parsing rule exists, this is achieved by traversing all parsing rules in the parsing library to find a parsing rule that matches all the basic attribute information of the barcode. That is, when determining whether a parsing rule matching the barcode exists in the parsing library, the basic attribute information of the barcode is matched with each sub-rule in the current parsing rule. In other words, if the basic attribute information of the barcode does not match any sub-rule in the current parsing rule, it indicates that the current parsing rule is not a matching parsing rule for the barcode, and the current parsing rule is skipped to proceed to the next parsing rule. Specifically, the basic attribute information of the barcode, such as length, character type, and encoding format, is sequentially matched with each sub-rule (e.g., length rule, character type rule) in the current parsing rule. If any item does not match, it is determined that the current parsing rule is not a matching parsing rule for the barcode, and the current parsing rule is skipped. After traversing all the parsing rules in the parsing library, if no parsing rule matching the barcode is found, it is determined that there is no parsing rule matching the barcode in the current parsing library, indicating that the current barcode is an unknown barcode. At this time, the intelligent parsing process for unknown barcodes is triggered.

[0044] In one possible implementation, when matching based on basic attribute information with preset parsing rules, it can be done using the sequential matching method described above, or a quantitative matching method can be used, as shown below: In one possible implementation, when matching the basic attribute information with a preset parsing library, irrelevant rules can first be eliminated by rule pruning. That is, based on the current barcode's encoding format, a subset of rules with the same encoding format is selected from the parsing library, and irrelevant rules are eliminated. Then, the matching score of each selected parsing rule is calculated using an incremental matching method, and the parsing rule matching the current barcode is confirmed based on the matching score. Specifically, according to the basic attribute weights from high to low (encoding format weight 0.5 > barcode length weight 0.3 > character type weight 0.2), each basic attribute information of the current barcode is matched item by item with the parsing rules in the rule subset, allowing for tolerable deviations in some attributes. A comprehensive matching score is calculated, and only when the score exceeds a preset threshold is the current rule determined as a candidate matching rule, which can be used as the parsing rule for selection. Thus, the parsing rule matching the current barcode can be obtained.

[0045] In one possible implementation, when filtering matching parsing rules from the parsing library, the method further includes filtering multiple candidate matching rules, and then selecting the best matching parsing rule from the multiple candidate matching rules as the parsing rule that matches the current barcode.

[0046] In one possible implementation, when multiple candidate matching rules exist, the optimal parsing rule is selected by following these steps: a. Initial screening and tie-breaking based on total priority score: First, based on the initial total priority score of the rules calculated by weighted summation of business value, matching efficiency, and accuracy as described above, all candidate rules are sorted in descending order; if multiple candidate rules have the same initial total priority score, then the sub-item scores are compared in order of priority of sub-item weight: business value > matching efficiency > accuracy, with the rule with the higher sub-item score ranked first; b. Matching degree calculation: If the order still cannot be distinguished after comparing the total priority score and sub-indicators, the comprehensive matching degree of the multi-dimensional basic attributes of the current barcode and each candidate rule is recalculated, and the rule with the highest matching degree is selected. c. Context adaptation: If the matching degree is still the same, select the rule that is bound to the corresponding scenario based on the current business scenario, or select the rule with the highest historical parsing success rate; Output: The optimal parsing rule selected at the end will be used as the parsing rule for the current barcode, and subsequent parsing operations will be performed.

[0047] After selecting the optimal parsing rule, the barcode is parsed using the optimal parsing rule to obtain the material identification identifier. The material identification identifier includes at least one of a material code or a product ID.

[0048] In one possible implementation, when parsing the barcode using the matching parsing rules to obtain a material identification identifier, the process includes: constructing a parsing sequence for the current barcode based on a preset parsing step template in the matching parsing rules; and parsing the barcode according to the parsing sequence to obtain a material identification identifier for the material.

[0049] It should be noted that this application, when setting up the parsing library, also pre-sets a corresponding parsing step template for each parsing rule. A parsing step template refers to a set of parsing operation procedures set for a specific type of barcode parsing rule, including at least one combination of steps such as segmentation, keyword extraction, and information decoding. Specifically, it defines "how to extract meaningful material information from a barcode string." When pre-setting the parsing step template for each parsing rule, the corresponding parsing step template is manually configured according to the encoding specifications of the material barcode corresponding to the current parsing rule (such as length, segmentation method, keyword position, etc.). For example, for manufacturing component barcodes: the parsing step template is defined as "segmented by 8-4-3," extracting 8 digits as the material code, the middle 4 digits as the batch number, and the last 3 digits as the specification code. Based on the parsing step template, the parsing sequence of the current barcode can be obtained. The parsing sequence is a dynamically constructed list of executable instructions based on the template and the current specific barcode content. If the above segmentation template is defined, and the barcode content is "123456780512123", the parsing sequence generated by parsing according to the template is "12345678", which is the material code, "0512" is the batch number, and "123" is the specification code.

[0050] In one possible implementation, when parsing the current barcode using the optimal parsing rules, the specific steps include: Dynamic parsing pipeline generation: Based on the preset parsing step template in the optimal parsing rules, the parsing sequence of the current barcode is automatically constructed. The parsing steps include at least one of segmentation, keyword extraction, and information decoding, and the order and logic of the parsing steps are obtained from the parsing step template. Contextual Association Validation: After parsing each field, cross-field relationships are validated based on preset business rules. If the validation fails, fault tolerance correction is triggered or the barcode is marked as abnormal. Specifically, validating cross-field relationships based on business rules means checking the consistency of business logic between different fields after parsing each field. One possible implementation includes, but is not limited to, verifying whether the material code exists in the ERP system and verifying whether the production date is later than the warehousing date—at least one of these. Failure to validate triggers fault tolerance correction or marks the barcode as abnormal.

[0051] In one possible implementation, fault-tolerant modification refers to adaptive fault-tolerant parsing. That is, when verification fails, the parsing result is first checked against a preset tolerable deviation range. If within the preset tolerance range, fault-tolerant modification is triggered; otherwise, it is marked as an abnormal barcode. Tolerable deviation refers to a small, non-destructive error between the actual barcode data and the preset standard parameters of the parsing rules. The error range is limited by a system preset threshold and does not change the core encoding semantics of the barcode. This mainly includes two categories: length deviation and character deviation. Length deviation refers to the difference between the actual barcode length and the standard length being within a preset threshold range (e.g., ±2 digits), often caused by printing defects or redundant characters during scanning. Character deviation refers to individual characters being printed blurry, causing recognition confusion, but without disrupting the barcode segmentation and keyword structure. Deviations exceeding the preset tolerance threshold are considered intolerable deviations and are directly marked as abnormal barcodes.

[0052] When the difference between the actual barcode string length and the preset standard length is within a tolerable deviation range, the barcode string is automatically repaired by padding or truncation before re-parsing. Padding involves adding check characters or fixed padding characters to the end of the barcode if the actual barcode is shorter than the standard length, according to the rules; if a check digit is missing, it is generated based on the preceding fields. Truncation involves identifying the coded keywords at the top of the barcode and removing redundant prefix characters or invalid separators and whitespace characters at the end of the barcode, retaining only the valid main content of the standard length. When keyword extraction fails, fuzzy matching is performed based on character set constraints to improve the parsing success rate of low-quality or incomplete barcodes; finally, a verified material identification label is output.

[0053] In one possible implementation, when it is determined that no matching parsing rule exists, triggering intelligent parsing of the unknown barcode to parse the barcode and obtain the material identification identifier of the material includes: clustering the basic attribute information with each parsing rule in the parsing library, determining whether there is a most similar rule cluster, and if so, using the most similar parsing rule for parsing; if not, determining that the current barcode is a completely new unknown type, and triggering a manual association parsing process. Specifically, when performing intelligent parsing of the unknown barcode, the operation of unknown barcode feature extraction and clustering is first performed, that is, extracting the basic feature vectors of the current barcode such as encoding format, length, and character distribution, and classifying it into the most similar known rule cluster through a clustering model, or marking it as a completely new unknown type; When the clustering result assigns the current barcode to the most similar known rule cluster, a hierarchical intelligent parsing approach can be used to generate multiple candidate parsing results. This hierarchical intelligent parsing includes at least one of the following strategies: similarity rule migration, general template parsing, or context-assisted parsing.

[0054] In one possible implementation, similarity rule migration refers to first attempting to parse the code using a known parsing rule within the cluster that is closest to the current barcode feature, and then checking if it succeeds.

[0055] If the migration fails, a generic template is used for parsing, which means using a "generic template" corresponding to the cluster. This generic template is the "greatest common divisor" of all rule features within the cluster. For example, for a "14-bit pure numeric rule cluster", the generic template may simply "output the 14-bit numeric string as a whole".

[0056] If all the above fail, context-assisted parsing can be used, which means that the parsing method can be inferred by combining the current business scenario (such as supplier, material type). For example, if the system knows that the goods currently being received come from supplier X, and supplier X's barcode is usually in a certain format, then it will try to parse it using that format.

[0057] Based on the above parsing methods, the parsing can be performed sequentially according to the above order, or each method can be executed to obtain multiple candidate parsing results.

[0058] Result verification and confidence assessment: The candidate parsing results are formatted and business-related, and the confidence level is calculated. Only the results with a confidence level ≥ the preset threshold are retained as intelligent parsing results. Rule self-learning update: If automatic parsing is successful, the basic attributes and parsing steps of the current barcode are saved as candidate new rules, which are updated to the parsing library after conflict detection, and the initial priority is configured to provide data support for subsequent parsing; If the above process still fails to yield a valid parsing result, a fallback manual association operation is triggered. After the user manually enters the material information, the parsing method of the barcode is simultaneously recorded in the learning library for subsequent rule optimization.

[0059] When clustering results cannot classify a material into any known rule cluster, it is considered a completely new unknown type, triggering a "fallback manual association operation." This means it cannot be automatically parsed and the user must manually input the material information. After the user manually associates the material, the system records the barcode and the user-inputted material information as data in the "learning library," used to generate new parsing rules later. In other words, completely new unknown types will not be automatically parsed; only manual processing and rule self-learning raw material collection are performed.

[0060] Based on the parsed material identifier, see [link / reference]. Figure 2As shown, this application also proposes a material association method for material association based on the aforementioned material barcode parsing method. The method includes steps S2100-S2200: S2100, acquiring materials and determining whether the materials have barcodes; when it is determined that the materials have barcodes, using a material barcode intelligent parsing method to parse the barcodes to obtain a material identification identifier; S2200, comparing the material identification identifier with the stored material information to find the corresponding materials and associating the barcodes with the materials.

[0061] This application utilizes the aforementioned barcode parsing method to associate real-time materials with material information in the ERP (Enterprise Resource Planning) system, thereby achieving material management. Furthermore, based on this barcode parsing method, it is applicable to barcode parsing across different industries and materials. Therefore, it can associate materials with large quantities of different industries and types, effectively improving the efficiency and adaptability of warehouse management. It should be noted that the ERP system stores complete information on various materials, including at least one of the following: material name, specifications, material, inventory quantity, category, and manufacturer.

[0062] When a material is found to have a barcode, the barcode is parsed using the aforementioned barcode parsing method. The parsed material identification is then compared with the material information stored in the ERP system. If the comparison matches, the current material is associated with the material information in the ERP system.

[0063] In one possible implementation, when it is determined that the material does not have a barcode, the attribute information of the material is obtained; wherein, the attribute information includes at least one of material code and batch number, and the acquisition method includes at least one of user manual input, IoT device collection, business context association extraction or OCR recognition; based on the attribute information, the complete material information is searched from the stored material information, and a corresponding barcode is generated based on the searched complete information; the barcode is associated with the currently searched material information.

[0064] Specifically, when it is determined that a material does not have a barcode, a barcode printing instruction is triggered to obtain the material's attribute information. The obtained attribute information is compared with the complete information of various materials stored in the ERP system. If the attribute information matches, the complete information of the current material is obtained. The material's identification information is extracted from the complete information, and the identification information is embedded into the barcode to generate a unique barcode for the current material. See the following for details: 1. The identification information is broken down into fields such as material code, batch number, factory code, and business type label according to business semantics; 2. Combine the data into a structured string according to the preset hierarchical format, and configure the corresponding error correction level and checksum for key fields; 3. Adaptively select either one-dimensional or two-dimensional barcode format based on material type and information capacity to encode structured strings; 4. Associate the generated unique barcode with the complete information of the material being searched.

[0065] Based on the material association method of this application, it can be ensured that unlabeled materials can also be quickly identified and information retrieved through the generated exclusive barcode, thus solving the problem that unlabeled materials cannot be managed by barcode.

[0066] In one possible implementation, based on the aforementioned barcode intelligent parsing and material association method, this application further includes an operation of optimizing the parsing rules to generate new parsing rules, enabling the parsing rules of this application to self-optimize. Specifically, optimizing the parsing rules includes: acquiring each parsing data; performing statistical analysis based on each parsing data to generate analysis results; and optimizing the parsing rules based on the analysis results to generate new parsing rules. Specifically, after the aforementioned parsing and association of the barcodes of each material is completed, a verification of the correctness of the association results is first performed, specifically including: 1. Verify whether the parsed material identification tags conform to the ERP system's material coding format specifications; 2. Verify whether the identifier corresponds to material information in the ERP system; 3. Verify whether the material information conforms to the constraints of the current scenario in conjunction with the business context; 4. Manual review is triggered for high-value materials; Record the verification result (correct / incorrect) and the reason for the error to the parsed data.

[0067] Record the parsing data for each barcode; the parsing data includes, but is not limited to, at least one of the following: barcode content, parsing rules used, parsing result (success / failure), association result (correct / incorrect), parsing time, failure reason (such as rule mismatch, barcode blurry, encoding error, etc.), and association error reason (such as format error, material not found, business mismatch, etc.).

[0068] After recording the parsed data, statistical analysis is performed on the recorded data periodically to calculate various core indicators. These core indicators include at least one of the following: barcode parsing success rate, material association error rate, adaptation rate of each parsing rule, distribution rate of failed barcode types, percentage of high-frequency causes of association errors, and rule execution index. The specific calculation methods for each core indicator are shown below: Barcode parsing success rate = (Number of successful parsings - Number of abnormal parsings) / (Total number of parsings - Number of invalid parsings) × 100% Material association error rate = (Number of incorrect associations + Number of fuzzy associations × Fuzzy coefficient) / Total number of associations × 100% The fit rate of each parsing rule = (Number of rule hits × Hit weight + Number of rule correct parsings × Correct weight) / Total number of barcode parsings × 100% Type distribution rate of failed barcode parsing = (Number of failures for a specific failure type / Total number of failures) × 100% Percentage of high-frequency causes of related errors = (Number of occurrences of a specific error cause / Total number of errors) × 100% Rule confidence index = (Number of correct rule parsings + Number of successful rule hits × Confidence coefficient) / (Total number of rule usages + 1) × 100% When optimizing parsing rules based on computational core metrics, at least one of the following is involved: generating new parsing rules, adjusting and optimizing existing parsing rules, and dynamically managing the rule base based on the adaptation rate.

[0069] When generating new parsing rules for optimization, the first step is to determine whether the barcodes for each type of rule are new specification barcodes based on the parsing success rate and the reasons for parsing failure. Specifically, it is determined whether the parsing success rate of a certain type of barcode is lower than a threshold. If it is confirmed that the parsing success rate of a certain type of barcode is lower than the threshold, the current barcode is confirmed to be a new specification barcode. At this time, the common features of the current specification barcode (including length, segmentation method, and character combination) are extracted, and new parsing rules are generated based on the common features and added to the parsing library.

[0070] Furthermore, after determining that the success rate of parsing a certain barcode specification is below a threshold, the reason for parsing failure can be used to further confirm whether the current barcode specification is a new specification barcode. For example, if the reason for parsing failure is "the preset rule does not cover the segmentation method of this barcode", it can be determined that the current barcode specification is a new specification barcode and new parsing rules need to be generated.

[0071] When dynamically managing the rule base based on the fit rate of each parsing rule, a rule optimization alarm is triggered for parsing rules whose fit rate has been below the threshold for a long time; for parsing rules whose fit rate has been rising, their matching priority is automatically increased.

[0072] When adjusting and optimizing existing parsing rules, at least one of rule parameter adjustment and priority adjustment is included.

[0073] When adjusting rule parameters, specific parameter adjustments are made based on the calculated material association error rate and the high-frequency causes of association errors. For example, if statistics show that a certain parsing rule has a high association error rate, and the error reason is always "fuzzy keyword matching," then it is determined that the keyword matching precision of this rule needs to be adjusted.

[0074] Priority adjustment includes two methods: configuring priorities for new rules and adjusting the priorities of old rules. When configuring the priority of a new parsing rule, an initial priority is first configured. This initial priority is calculated based on a weighted average of three dimensions: business value, matching efficiency, and recognition accuracy: Initial Priority = 0.4 × Business Value Score + 0.3 × Matching Efficiency Score + 0.3 × Accuracy Score When performing parsing rule matching, matching can be performed sequentially based on priority order.

[0075] After generating new parsing rules and configuring their initial priorities, historical performance scores are statistically analyzed through small-scale gray-scale testing. The new rule priority is then calculated using a mixed approach, and can be expressed by the formula: New rule priority = 0.6 × initial priority + 0.4 × (rule fit rate × 10 - average parsing time × 0.1) Priority adjustments for old parsing rules mainly refer to the increase in priority of old parsing rules. Specifically, when the core metrics of an old parsing rule meet the conditions of adaptation rate ≥ 90%, usage frequency ≥ 20%, or association error rate ≤ 1%, its priority is increased according to the following formula: New priority = min(current priority + (rule fit rate - 0.8) × 20, 100) It should be noted that the priority adjustment cannot be increased by more than 20 points at a time, and can be adjusted a maximum of once per day, so as to shorten the parsing time while ensuring system stability.

[0076] In one possible implementation, the optimized parsing rules of this application can be directly added to the parsing library for parsing, or they can be added to the parsing library after user approval; the specific implementation is not limited. User approval ensures the rationality and accuracy of the optimized parsing rules, effectively improving the efficiency and accuracy of barcode parsing.

[0077] Based on this, this application provides a method for intelligent parsing of material barcodes, including: acquiring the barcode of the material and identifying the basic attribute information of the barcode; matching the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists; if a matching parsing rule exists, using the matching parsing rule to parse the barcode to obtain the material identification identifier of the material; wherein, the parsing library includes at least two types of parsing rules; when it is determined that no matching parsing rule exists, triggering an unknown barcode intelligent parsing process to parse the barcode to obtain the material identification identifier of the material. This application obtains the material identification identifier of the barcode by searching for a parsing rule that matches the current barcode in each parsing rule in the preset parsing rule library, and simultaneously, when no matching parsing rule is found in the parsing library, using an unknown barcode intelligent parsing method to parse the barcode. In this way, it is possible to parse all barcodes during intelligent parsing of material barcodes, thus improving the adaptability of barcode parsing. Meanwhile, this application also incorporates parsing rule optimization technology. Through self-optimization of the parsing rules, it continuously adjusts various parsing rules in the parsing library, thereby simultaneously ensuring compatibility with known heterogeneous barcodes of multiple specifications and providing intelligent fallback parsing for unknown barcodes, thus improving the overall universality of warehouse barcode parsing. This application achieves intelligent parsing of warehouse labels and material association through the collaborative operation of three layers of logic: "full-specification adaptation, accurate association, and self-learning optimization." By pre-setting multiple types of parsing rules, self-optimization of parsing rules, and automatic association of unlabeled materials, the adaptability of label parsing and the efficiency of warehouse management can be effectively improved.

[0078] <Device Embodiment> Figure 3 A schematic block diagram of a material barcode intelligent parsing device according to an embodiment of this application is shown. Figure 3 As shown, the device 100 includes: a data acquisition module 110, a parsing rule parsing module 120, and an unknown barcode parsing module 130. The data acquisition module 110 is used to acquire the barcode of the material and identify the basic attribute information of the barcode. The parsing rule parsing module 120 is used to match the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists. If a matching parsing rule exists, the matching parsing rule is used to parse the barcode to obtain the material identification identifier of the material. The parsing library includes at least two types of parsing rules. The unknown barcode parsing module 130 is used to trigger an intelligent unknown barcode parsing process to parse the barcode to obtain the material identification identifier of the material when it is determined that no matching parsing rule exists.

[0079] Figure 4 A schematic block diagram of a material association device according to an embodiment of this application is shown, such as Figure 4As shown, the device 300 is used for material association based on the aforementioned intelligent material barcode parsing method, and the device 300 includes: a barcode presence determination module 310 and an association module 320. The barcode presence determination module 310 is used to acquire materials and determine whether the materials have barcodes. When it is determined that the materials have barcodes, the intelligent material barcode parsing method is used to parse the barcodes to obtain the material identification identifier. The association module 320 is used to compare the material identification identifier with stored material information, find the corresponding materials, and associate the barcode with the materials.

[0080] <Equipment Example> Figure 5 A schematic block diagram of a material association device according to an embodiment of this application is shown. Figure 5 As shown, the material association device 200 includes a processor 210 and a memory 220 for storing executable instructions of the processor 210. The processor 210 is configured to implement any of the aforementioned material association methods when executing the executable instructions.

[0081] It should be noted here that the number of processors 210 can be one or more. Furthermore, the material association device 200 in this embodiment may also include an input device 230 and an output device 240. The processors 210, memory 220, input device 230, and output device 240 can be connected via a bus or other means, which are not specifically limited here.

[0082] The memory 220, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the material association method in the embodiments of this application. The processor 210 executes various functional applications and data processing of the material association device 200 by running the software program or module stored in the memory 220.

[0083] Input device 230 can be used to receive input digital numbers or signals. These signals may include key signals related to user settings and function control of the device / terminal / server. Output device 240 may include a display device such as a screen.

[0084] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technological improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for intelligent parsing of material barcodes, characterized in that, include: Obtain the barcode of the material and identify the basic attribute information of the barcode; Based on the basic attribute information, the barcode is matched with various parsing rules in the preset parsing library to determine whether a matching parsing rule exists. If a matching parsing rule exists, it is used to parse the barcode to obtain the material identification identifier of the material. The parsing library includes at least two types of parsing rules. When it is determined that no matching parsing rule exists, the unknown barcode intelligent parsing process is triggered to parse the barcode and obtain the material identification identifier of the material.

2. The intelligent barcode parsing method for materials according to claim 1, characterized in that, The basic attribute information of the barcode includes at least one of the following: barcode length, character type in the barcode, and encoding format.

3. The intelligent barcode parsing method for materials according to claim 1, characterized in that, Based on the basic attribute information, the parsing rules in the preset parsing library are matched. When determining whether there is a matching parsing rule, the parsing rules in the parsing library are traversed to find the parsing rule that matches all the basic attribute information of the barcode.

4. The intelligent barcode parsing method for materials according to claim 1, characterized in that, When parsing the barcode using the matching parsing rules to obtain the material identification identifier of the material, the process includes: The parsing sequence of the current barcode is constructed based on the preset parsing step template in the matching parsing rules; The material identification identifier of the material is obtained by parsing the barcode according to the parsing sequence.

5. The intelligent barcode parsing method for materials according to claim 1, characterized in that, When parsing the barcode using the unknown barcode intelligent parsing process, the following steps are included: Extract the feature vectors corresponding to the basic attributes of the barcode. Cluster analysis is performed based on the feature vector and the parsing rules in the parsing library to determine whether there is a rule cluster with the highest similarity; if a rule cluster with the highest similarity exists, the parsing rules in that rule cluster are used to complete the parsing. If no rule cluster with the highest similarity is found, the current barcode is determined to be a completely new and unknown type, and a manual association and parsing process is triggered.

6. A material association method, characterized in that, For material association based on the intelligent material barcode parsing method according to any one of claims 1 to 5, including: The material is acquired and it is determined whether the material has a barcode. When it is determined that the material has a barcode, the material barcode intelligent parsing method is used to parse the barcode to obtain the material identification mark. The barcode is then compared with the stored material information to locate the corresponding material and associate it with the material.

7. A material association method according to claim 6, characterized in that, When determining that the material does not have a barcode, the following applies: Obtain the attribute information of the material; wherein the attribute information includes at least one of material code and batch number; Based on the attribute information, complete material information is retrieved from the stored material information, and a corresponding barcode is generated based on the retrieved material information; Associate the barcode with the currently searched material information.

8. A material barcode intelligent parsing device, characterized in that, include: The data acquisition module is used to acquire the barcode of the material and identify the basic attribute information of the barcode; The parsing rule parsing module is used to match the basic attribute information with various parsing rules in a preset parsing library to determine whether a matching parsing rule exists. If a matching parsing rule exists, the matching parsing rule is used to parse the barcode to obtain the material identification identifier of the material. The parsing library includes at least two types of parsing rules. The unknown barcode parsing module is used to trigger the unknown barcode intelligent parsing process to parse the barcode and obtain the material identification mark of the material when it is determined that no matching parsing rule exists.

9. A material association device, characterized in that, For material association based on the intelligent material barcode parsing method according to any one of claims 1 to 5, including: The barcode presence determination module is used to acquire materials and determine whether the materials have barcodes. When it is determined that the materials have barcodes, the material barcode intelligent parsing method is used to parse the barcodes to obtain the material identification identifier. The association module is used to compare the material identification mark with the stored material information, find the corresponding material, and associate the barcode with the material.

10. A material association device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 7 when executing the executable instructions.

Images