Loading capacity automatic identification method and system for intelligent factory

Abstract

The invention discloses a loading capacity automatic identification method and system for an intelligent factory, and relates to the related technical field of goods identification, and the method comprises the steps: connecting a task platform to obtain loading task information, and obtaining order goods composition; searching a goods management feature library by using goods numbers formed by order goods as indexes, obtaining goods identification features, performing in-order interference feature identification on the goods identification features, and determining interference identification features; obtaining interference adjacent goods and adjacent interference characteristics thereof in the intelligent factory; performing feature selection processing on the goods identification features to obtain robust features; and carrying out loading task identification processing according to the number of goods by using the robust features as identification features. The technical problem of low loading accuracy and efficiency caused by the fact that goods combinations cannot be efficiently and accurately recognized for diversified and customized production requirements in existing loading quantity recognition is solved, and the technical effect of improving the efficiency and accuracy of factory loading is achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to goods identification, and specifically to a method and system for automatic identification of loading quantity for smart factories. Background Art

[0002] In today's wave of intelligent manufacturing, smart factories have gradually become the core driving force of modern manufacturing, and are gradually realizing the automation, intelligence and efficiency of the production process. With the vigorous development of the e-commerce industry and the increasing requirements of consumers for logistics and distribution efficiency, loading operations in smart factories are facing unprecedented challenges. Loading in smart factories usually requires the assembly and transportation of specific goods according to production orders and customer needs. With the expansion of factory production scale and the increase in the number of orders, loading tasks have become more complex and diversified. The traditional loading process is difficult to cope with a variety of different types of goods combinations, especially customized products. In the logistics warehousing of the factory, the goods are stacked tightly, and adjacent goods are prone to interfere with each other. There are often various interference factors at the loading site, such as confusion of similar goods, damage to packaging labels, etc. The traditional manual verification of the loading list and the quantity of goods is not only inefficient, but also difficult to accurately identify the target goods during the actual loading process, which is prone to errors, thus affecting the accuracy and efficiency of loading, and seriously restricting the overall efficiency of the logistics supply chain.

[0003] Therefore, at the current stage, the technologies related to loading quantity identification used in smart factories have technical problems such as being unable to efficiently and accurately identify product combinations in the face of diversified and customized production needs, resulting in low loading accuracy and efficiency. Summary of the invention

[0004] This application provides a method and system for automatic identification of loading quantities for smart factories, thereby solving the technical problem that existing loading quantity identification cannot efficiently and accurately identify the combination of goods in the face of diversified and customized production needs, resulting in low loading accuracy and efficiency, and achieves the technical effect of improving the efficiency and accuracy of factory loading.

[0005] The present application provides a method for automatically identifying loading quantities for smart factories, the method comprising: connecting to a task platform to obtain loading task information, performing task decomposition on the loading task information, and obtaining order goods composition, wherein the order goods composition is information on all goods that need to be assembled in the loading task, including a goods number and a quantity of goods; using the goods number of the order goods composition as an index, searching a goods management feature library to obtain a goods identification feature, performing order interference feature identification on the goods identification feature, and determining an interference identification feature; using the goods identification feature as an index, searching the goods management feature library to obtain interfering neighboring goods and their neighboring interference features in the smart factory; using the interference identification feature and the neighboring interference feature as constraints, performing feature selection processing on the goods identification feature to obtain a robust feature, wherein the robust feature is a core identification feature or a core identification feature combination that is not affected by the interference feature; using the robust feature as an identification feature, performing loading task identification processing according to the quantity of goods.

[0006] In a possible implementation, security triage is performed through the abnormal risk markers, and the following processing is performed: based on a machine learning algorithm, a vulnerability scanning detection model is established, and the vulnerability scanning detection model includes a quantitative feature identification scanning layer of event types and a trigger frequency feature identification scanning layer; based on the vulnerability scanning detection model, security triage is performed through the abnormal risk markers to determine high-confidence attack behaviors.

[0007] In a possible implementation, the search of the goods management feature library is preceded by the following processing: collecting the physical characteristics, image characteristics, and kit composition characteristics of all incoming goods in the smart factory, wherein the physical characteristics include weight, density, and material; the image characteristics include size, shape, color, texture, and light reflection characteristics; and the kit composition characteristics include kit component quantity and installation and assembly position; encoding the physical characteristics, image characteristics, and kit composition characteristics, converting them into a storable format, establishing a mapping relationship between the item number and the feature code of the incoming goods, and constructing the goods management feature library, wherein a multi-level mapping relationship from component to kit and from kit to goods number is established for the kit composition characteristics.

[0008] In a possible implementation, the product identification features are subjected to intra-order interference feature identification to determine the interference identification features, and the following processing is performed: partitioning is performed according to the product number, physical features, image features, and kit composition features; the product number is used as a unique index and integrated into the product number partition of the product management feature library; combined indexes are established respectively using the physical features, image features, and kit composition features as primary indexes and the specific features under the features as auxiliary indexes; the combined indexes corresponding to each feature category are respectively integrated into the physical features, image features, and kit composition feature partitions of the product management feature library.

[0009] In a possible implementation, the comprehensive similarity value is obtained, and the following processing is further performed: a set of loading abnormal cases is obtained, and the loading abnormal case set is clustered according to the factory operation process and the product number to construct an abnormal case cluster; through the abnormal case cluster, the abnormal characteristics and abnormal probabilities of the factory operation process and the product number are fitted respectively to obtain the abnormal characteristics and abnormal probabilities of the factory operation process, and the abnormal characteristics and abnormal probabilities of the product number; the factory operation process of the loading task is extracted according to the loading task information, and the abnormal characteristics and abnormal probabilities of the process are matched; according to the order cargo The product numbers of the products are matched with the abnormal features and abnormal probabilities of the product numbers; according to the abnormal features and abnormal probabilities of the matching process and the abnormal features and abnormal probabilities of the product numbers, the abnormal probability matching is performed on the features of each pair of product combinations and their feature category similarities, and the average abnormal probability is calculated, where the average abnormal probability is the average of the abnormal probabilities of each product in the product combination in the feature category and the abnormal probability of the matching process; the comprehensive similarity is weighted by using the average abnormal probability, and the similarity weighted calculation is performed on each pair of product combinations by using the weighting to obtain the comprehensive similarity value.

[0010] In a possible implementation, the interference identification feature and the adjacent interference feature are used as constraints to perform feature selection processing on the product identification feature to obtain a robust feature, and the following processing is also performed: the interference identification feature and the adjacent interference feature are removed from the product identification feature as removal features to construct a pure identification feature set;

[0011] Random sampling is performed from the pure recognition feature set to construct several feature subsets, each of which includes different numbers of features or feature combinations; the recognition probability of each random feature or feature combination is evaluated by the recognition evaluation model to obtain the recognition evaluation probability, and the randomly sampled feature with the largest recognition evaluation probability is screened as the candidate robust feature; it is determined whether the recognition evaluation probability reaches the preset recognition target, and if so, the candidate robust feature is used as the robust feature; if not, the interference recognition feature and the adjacent interference feature are used as additional features, randomly selected from the additional features, and added to the feature subset using the randomly added features to obtain a newly constructed feature combination; the recognition probability of the newly constructed feature combination is evaluated using the recognition evaluation model until the recognition evaluation probability reaches the preset recognition target, and the robust feature is determined.

[0012] In a possible implementation, the recognition evaluation model is used to evaluate the recognition probability of the newly constructed feature combination until the recognition evaluation probability reaches the preset recognition target, and the following processing is also performed: the recognition weight of the newly constructed feature combination is adjusted, wherein the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination; the recognition probability of the newly constructed feature combination is evaluated by the recognition evaluation model, and if the recognition evaluation probability does not reach the preset recognition target, the interference features are continued to be added to update the recognition feature combination; wherein, the recognition weight adjustment is performed before each iteration to ensure that the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination, and the iteration is repeated until the recognition evaluation probability reaches the preset recognition target.

[0013] The present application also provides a loading quantity automatic identification system for a smart factory, including: an order goods composition acquisition module, used to connect to a task platform to obtain loading task information, perform task decomposition on the loading task information, and obtain order goods composition, wherein the order goods composition is information on all goods that need to be assembled in the loading task, including goods number and goods quantity; an interference identification feature determination module, used to use the goods number of the order goods composition as an index, search the goods management feature library, obtain goods identification features, perform order interference feature identification on the goods identification features, and determine the interference identification features; goods A management feature library search module is used to search the product management feature library with the product identification feature as an index to obtain the interfering adjacent products and their adjacent interference features in the smart factory; a product identification feature processing module is used to use the interference identification feature and the adjacent interference feature as constraints to perform feature selection processing on the product identification feature to obtain a robust feature, wherein the robust feature is a core identification feature or a combination of core identification features that is not affected by the interference feature; a loading task identification processing module is used to use the robust feature as an identification feature to perform loading task identification processing according to the quantity of the products.

[0014] The automatic identification method and system for loading quantity for smart factories proposed in this application is intended to connect to the task platform to obtain loading task information and order goods composition; use the goods number of the order goods composition as an index to search the goods management feature library to obtain the goods identification features, perform interference feature identification within the order on the goods identification features, and determine the interference identification features; obtain the interference neighboring goods and their neighboring interference features within the smart factory; perform feature selection processing on the goods identification features to obtain robust features; use the robust features as identification features to perform loading task identification processing according to the quantity of goods. This solves the technical problem that the existing loading quantity identification cannot efficiently and accurately identify the combination of goods in the face of diversified and customized production needs, resulting in low loading accuracy and efficiency, and achieves the technical effect of improving the efficiency and accuracy of factory loading. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0016] Figure 1 A schematic diagram of a method for automatically identifying the amount of cargo for a smart factory provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of the structure of an automatic cargo quantity identification system for a smart factory provided in an embodiment of the present application.

[0018] Explanation of the reference numerals: order product composition acquisition module 10, interference identification feature determination module 20, product management feature library search module 30, product identification feature processing module 40, loading task identification processing module 50. DETAILED DESCRIPTION

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0022] The present application embodiment provides a method for automatically identifying the loading quantity for a smart factory, such as Figure 1 As shown, the method includes:

[0023] Step S100, connecting to the task platform to obtain loading task information, decomposing the loading task information, and obtaining order product composition, wherein the order product composition is information on all products that need to be assembled in the loading task, including product numbers and product quantities.

[0024] Preferably, the task platform is connected to the network or other communication means to obtain detailed information of the loading task to be executed, which may include customer orders, shipping instructions, loading time, etc., wherein the task platform is used to manage and distribute various production and logistics tasks in the factory, and is usually integrated with a production management system (such as an ERP system) or an order management system (such as an OMS system) to ensure that all tasks can be updated and distributed in real time. Task decomposition refers to decomposing the loading task information obtained from the task platform to obtain the relevant order product composition, wherein the loading task includes the classification and quantity of the goods specified for assembly in the order, and the order product composition refers to all the product information that needs to be assembled in the loading task. Each order usually contains different types and quantities of goods, including the product number and quantity of each product. The product number is a unique identifier of the product, which is used to identify and manage specific products in the system, such as barcodes, QR codes, etc. The quantity of goods refers to the quantity of each product that needs to be loaded in the loading order. By reading the quantity information in the order, it is ensured that the correct quantity can be accurately loaded during loading.

[0025] Step S200, using the product number of the order product composition as an index, searching the product management feature library to obtain product identification features, performing order interference feature identification on the product identification features, and determining interference identification features.

[0026] Preferably, the number of the goods in each order is used as a query keyword or index to search in the goods management feature library to find the feature information related to the goods. These features may include the physical characteristics of the goods (such as size, shape, color), packaging method, identifier, visual features (such as image or model), etc., wherein the goods management feature library is a database or information library that stores the feature information related to each product. These identification features are the key information used by the system to distinguish different goods in actual operation. The feature library is searched by the product number to extract the identification features related to the product. For example, for a specific part, the identification feature The characteristics of each item may be its unique shape and color information. The identification features of all items in the order are analyzed to identify the features that may cause interference. Because in the actual loading process, there may be multiple items in the order, and they may have similarities or overlaps in features. These similar features may cause interference or misidentification in the recognition process. For example, if two items have the same color but different shapes, it is necessary to identify and distinguish the interfering identification features to avoid confusing them during loading. Finally, the interfering identification features are determined so that corresponding measures can be taken during the loading process to ensure the accuracy and robustness of the final recognition results, that is, to maintain accurate product identification in the presence of interference.

[0027] In a possible implementation, step S200 further includes step S210, collecting the physical characteristics, image characteristics, and kit composition characteristics of all incoming goods in the smart factory, wherein the physical characteristics include weight, density, and material, the image characteristics include size, shape, color, texture, and light reflection characteristics, and the kit composition characteristics include kit component quantity and installation and assembly position; step S220, encoding the physical characteristics, image characteristics, and kit composition characteristics, converting them into a storable format, establishing a mapping relationship between the item number and the feature code of the incoming goods, and constructing the item management feature library, wherein a multi-level mapping relationship from component to kit and from kit to item number is established for the kit composition characteristics.

[0028] Preferably, the incoming goods refer to all goods that are about to be stored in the warehouse in the smart factory, which can be raw materials, semi-finished products or finished products. Each product needs to collect its key feature information, including physical features, image features, and kit composition features. Specifically, physical features refer to information related to the physical properties of the goods, including the weight, density, and material of the goods. Image features are appearance features of the goods collected through visual information, including the size, shape, color, texture, and light reflection features of the goods. Kit composition features refer to certain goods composed of multiple components or parts (such as mechanical equipment or electronic equipment, etc.), including the number of kit components (the number of components contained in a kit), installation, and assembly position (the installation position and assembly order of the components in the kit); then the collected physical features, image features, and kit composition features are converted into digital coding forms. The coding can be a combination of numbers, letters, or other specific identifiers to facilitate storage, retrieval, and processing by the system. For example, color can be encoded by RGB or HSV values, size can be represented by specific numerical values, and material can be represented by standardized codes (such as metal can be encoded as "MTL" and plastic can be encoded as "PLS"). , convert these encoded data into a format that the system can store and process, usually a structured data format, such as a table format in a database, XML or JSON file, etc., establish a mapping relationship between the product number and the feature code of the incoming goods, that is, each product number is associated with the feature code of the product to form a mapping relationship, and form a product management feature library with feature mapping relationships of multiple products, which records the mapping relationship between the physical characteristics, image characteristics, kit composition characteristics and their numbers of each product. Through this feature library, when identifying goods, managing inventory or performing loading tasks, relevant product characteristics can be quickly found and processed; among them, for the kit composition characteristics, a multi-level mapping relationship from component to kit and from kit to product number is established, that is, a mapping relationship is established for the components contained in each product, and the specific information of each component and its role in the kit are clarified. A product may be composed of multiple kits, and each kit is mapped to the final product number. For example, a product number may correspond to multiple kits, and each kit corresponds to multiple components. By establishing a multi-level mapping relationship, the system can quickly trace the detailed composition information of each kit and the component information in each kit from the product number.

[0029] In a possible implementation, step S220 further includes step S221, partitioning according to the product number, physical characteristics, image characteristics, and kit composition characteristics; step S222, using the product number as a unique index and integrating it into the product number partition of the product management feature library; step S223, using the physical characteristics, image characteristics, and kit composition characteristics as primary indexes, and using specific characteristics under the characteristics as auxiliary indexes, to establish combined indexes respectively; step S224, integrating the combined indexes corresponding to each feature category into the physical characteristics, image characteristics, and kit composition feature partitions of the product management feature library.

[0030] Preferably, the product is partitioned according to the product number, physical feature, image feature, and kit composition feature, an index is established for each category in the product management feature library, and multi-level indexes are used in the feature library to effectively store and retrieve this information, so as to facilitate fast search and matching. Specifically, different types of information are logically divided into independent areas for more efficient management and retrieval, that is, they are divided into product number partitions, physical feature partitions, image feature partitions, and kit composition feature partitions; the product number and its related information are stored in the product number partition using the product number as the unique index to ensure fast and unique access to all information about a product; the physical features, image features, and kit composition features are used as the primary indexes of their respective categories, so as to ensure fast and unique access to all information about a product. The specific features under the feature are used as auxiliary indexes to establish combined indexes respectively. Among them, the main index is the main index used to quickly find data of a specific category, and the auxiliary index is used to further refine the query based on the main index. For example, physical features are the main index of this category, and weight, density, material, etc. are auxiliary indexes. The main index and auxiliary index are combined to form a combined index, that is, you can first perform preliminary screening based on the main index (such as physical features), and then perform refined retrieval through the auxiliary index; the combined indexes established previously are integrated into different feature partitions, that is, the combined index of physical features is integrated into the physical feature partition, the combined index of image features is integrated into the image feature partition, and the combined index of kit composition features is integrated into the kit composition feature partition.

[0031] In a possible implementation, step S200 further includes step S230, performing feature similarity calculations according to the physical features, image features, and kit composition features to obtain the similarity of each feature category; step S240, establishing an N×N matrix, where N is the number of product categories, and each element represents the similarity of two products under a certain feature category; step S250, filling the matrix according to the feature category similarity, performing weighted averaging of the similarities for each pair of product combinations, and obtaining a comprehensive similarity value; step S260, assigning values ​​to the matrix using the comprehensive similarity values, and when the matrix elements reach a preset interference threshold, determining the corresponding features as the interference identification features.

[0032] Preferably, feature similarity calculations are performed according to physical features such as weight, density, and material, image features such as size, shape, color, and texture of goods, and kit composition features such as the number of components and assembly position of goods, respectively, to obtain similarity values ​​for each pair of goods in different feature categories; then an N×N matrix is ​​established with these calculated feature similarities, where N represents the number of types of goods. For example, if there are 5 types of goods, the size of the matrix is ​​5×5, and each element of the matrix represents the similarity of two goods in a certain feature category; the similarity calculation results of physical features, image features, and kit composition features are filled into different parts of the matrix, respectively, and each pair of goods combination will have a corresponding similarity value in the matrix. The similarity weighted average is performed on each pair of goods combination to obtain a comprehensive The combined similarity value is calculated by assigning different weights to the physical features, image features and kit composition features, and then filling these values ​​into the matrix. For example, if the comprehensive similarity of goods A and B is 0.85, then the elements at the A and B positions in the matrix will be assigned a value of 0.85. The preset interference threshold is a critical value used to determine whether two goods are similar to the extent that they may cause confusion or interference in recognition. If the similarity of the two goods exceeds the preset interference threshold, it indicates that some features of the two goods may be too similar and may easily cause interference in the recognition process. In this case, the corresponding features are determined as interference recognition features. For example, if the two goods are too similar in image features such as color and size, then the color and size features are marked as interference features.

[0033] In a possible implementation, step S250 further includes step S251, obtaining a set of loading abnormal cases, clustering the loading abnormal case set according to the factory operation process and the product number, and constructing an abnormal case cluster; step S252, fitting the abnormal characteristics and abnormal probabilities of the factory operation process and the product number through the abnormal case cluster, and obtaining the abnormal characteristics and abnormal probabilities of the factory operation process and the abnormal characteristics and abnormal probabilities of the product number; step S253, extracting the factory operation process of the loading task according to the loading task information, and matching the abnormal characteristics and abnormal probabilities of the process; step S254, according to The product numbers of the order product components are matched with the abnormal features and abnormal probabilities of the product numbers; step S255, according to the abnormal features and abnormal probabilities of the matching process and the abnormal features and abnormal probabilities of the product numbers, the abnormal probability matching is performed on the features of each pair of product combinations and their feature category similarities, and the average abnormal probability is calculated, and the average abnormal probability is the average of the abnormal probabilities of each product in the product combination in the feature category and the abnormal probability of the matching process; step S256, weighting the comprehensive similarity using the average abnormal probability, and performing weighted similarity calculation on each pair of product combinations using the weighting to obtain the comprehensive similarity value.

[0034] Preferably, the loading abnormal case set is a collection of all abnormal situations recorded by the factory during the loading process. For example, the wrong loading, missing loading, and incorrect identification of goods during the loading process can all be regarded as abnormal cases. According to the different operating procedures (such as loading, testing, and packaging) and product numbers of the factory, these abnormal cases are clustered and analyzed, and similar abnormal cases are put together to identify common features. For example, if the same type of goods frequently have abnormalities in the same process, they are clustered into a cluster, forming several abnormal case clusters, each cluster representing common abnormalities under similar operating procedures or product numbers; abnormal features refer to the abnormalities in a certain factory operation process. The features that often appear in the order or a certain product number are analyzed. By analyzing the abnormal case clusters, the abnormal occurrence probability of each process and product number is fitted to determine the probability of abnormal events. When performing the loading task, the specific processes involved in the task are extracted, such as loading, sorting, packaging, etc., and then the extracted processes are matched with the abnormal features and probabilities obtained from the previous analysis to determine what the abnormal features of each process are and what the probability of abnormality is in the current loading task. According to the product number in the order, the abnormal features and abnormal probabilities in the abnormal case cluster are matched to find out the common abnormal situations and probabilities of the product in historical operations.

[0035] Preferably, the similarity of each pair of product combinations is then matched with the corresponding process abnormality features and product number abnormality features, and the abnormality probabilities of these feature categories are calculated, that is, the possibility of abnormality under a specific process and product feature combination is calculated. For example, if the two products have a high similarity in color (0.9), and color recognition errors are a common abnormal feature, then the abnormality probability may be high; the abnormality probability of the feature category of each pair of products and the abnormality probability of the corresponding process are weighted averaged to calculate the mean abnormality probability, which represents the possibility of abnormality in this specific operation. For example, the comprehensive similarity calculation of product a and product b for color features, assuming that the confusion probability of product a in this process flow is 0.4 and the color probability is 0.1, then the abnormality The normal probability mean is 0.25, the historical process confusion anomaly of product b is 0.5, the color confusion probability is 0.05, and the probability mean is 0.275. When assigning weights, it should be noted that the sum of the weights of the same product feature pair is 1, so the weights are determined by converting according to 0.25 and 0.275. In this calculation, the weight of product a = 0.25 / (0.25+0.275), and the weight of product b = 0.275 / (0.25+0.275); according to the average probability of abnormality of each pair of product combinations, adjust its comprehensive similarity value. The higher the abnormal probability, the more likely the characteristics of this pair of products are to cause problems in operation. Therefore, the weight of the similarity may be reduced, and finally a weighted comprehensive similarity value is obtained, which not only considers the similarity of product characteristics, but also combines the possibility of abnormality.

[0036] Step S300, using the product identification feature as an index, searching the product management feature library to obtain interfering neighboring products and their neighboring interference features in the smart factory.

[0037] Preferably, the product identification feature is used as an index keyword to conduct a further search in the product management feature library to find other neighboring products and their features related to the current product identification feature, so as to ensure that those products that have similar features to the target product and may cause identification confusion can be identified, that is, to obtain interfering neighboring products and their neighboring interference features in the smart factory, wherein interfering neighboring products refer to products that are adjacent to the target product in the smart factory and have similar or interfering features. For example, neighboring products may be physically close to the target product (such as the position in the warehouse or on the production line), or they may have similar features, such as color, shape, etc., which are similar to the target product during the product identification process, resulting in identification confusion or misjudgment. After finding the interfering neighboring products, the interference features of these products will be further extracted and considered in the subsequent identification process. Neighboring interference features refer to features of interfering neighboring products, which may be similar or overlapping with the features of the target product, thereby causing interference in product identification. For example, if two products are similar in color or close in shape, these features will be marked as interference features.

[0038] Step S400, taking the interference identification feature and the adjacent interference feature as constraints, performing feature selection processing on the product identification feature to obtain a robust feature, wherein the robust feature is a core identification feature or a combination of core identification features that is not affected by the interference feature.

[0039] Preferably, an optimal search is used to find the feature or feature combination with the least interference. Specifically, using interference identification features and adjacent interference features as constraints means that when performing feature selection, interference identification features and adjacent interference features are used as a constraint or filtering rule to limit those features that are easily interfered with, and give priority to those features that are not affected by interference identification features and adjacent interference features, so as to avoid misjudgment due to interference and ensure the robustness of the system. Feature selection processing refers to selecting the most effective feature from multiple product identification features as the basis for identifying the product. The goal of feature selection is to reduce data dimensions while retaining the most discriminative and least interfered features. For example, the principal component analysis method is used to perform feature selection processing on the product identification features to obtain robust features. The selection of robust features ensures that the target product can be correctly identified even in the presence of interference factors. For example, if the color is easily interfered with, but the shape is unique and less interfered with, then the shape can be used as a robust feature, where the robust feature The core recognition feature or core recognition feature combination that is not affected by interference features means that the robust feature can be a single feature (such as shape) or a combination of multiple features (such as shape + label). By combining different robust features, it is possible to maintain a high accuracy in product recognition in a variety of interference environments. The core recognition feature refers to the most discriminative and important feature in the product recognition process. These features can accurately and quickly identify products and are not easily affected by interference factors in the environment. For example, in the case of similar colors, size may become a core recognition feature because size is relatively stable and not easily misidentified. Sometimes a single feature cannot provide sufficient distinguishing ability, and multiple features are required to form a core recognition feature combination. For example, a combination of color and shape may be more reliable than a single shape or color. By combining multiple uninterfered features, it is possible to more effectively identify target products in complex loading environments (such as stacked products, similar adjacent products, etc.), reduce the error rate of product recognition, and maintain high accuracy and reliability of product recognition.

[0040] In a possible implementation, step S400 further includes step S410, removing the interference identification feature and the adjacent interference feature as elimination features from the product identification feature to construct a pure identification feature set; step S420, randomly sampling from the pure identification feature set to construct a number of feature subsets, the subsets including different numbers of features or feature combinations, performing recognition probability evaluation on each random feature or feature combination through the recognition evaluation model to obtain the recognition evaluation probability, and screening the randomly sampled feature with the largest recognition evaluation probability as the candidate robust feature; step S430, judging whether the recognition evaluation probability reaches the preset recognition target, and if so, taking the candidate robust feature as the robust feature; step S440, when not reached, taking the interference identification feature and the adjacent interference feature as added features, randomly selecting them from the added features, adding them to the feature subset using the randomly added features to obtain a newly constructed feature combination, and performing recognition probability evaluation on the newly constructed feature combination using the recognition evaluation model until the recognition evaluation probability reaches the preset recognition target, and determining the robust feature.

[0041] Preferably, the determined interference identification features and adjacent interference features are removed from the product identification feature set as elimination features, and the remaining product identification features are constructed into a pure identification feature set, which no longer contains features that may cause confusion or interference; features are randomly selected from the pure identification feature set to generate several different feature subsets, which may include different numbers of features or feature combinations. For example, one feature subset may only include size and color, and another subset may include shape, color and weight. The purpose is to find the best robust feature combination among multiple feature combinations; the recognition probability of each random feature or feature combination is evaluated by the recognition evaluation model, which means that the recognition probability of each feature subset is calculated by the evaluation model, that is, the success probability of the subset in product identification. The higher the recognition probability, the more accurately and effectively the feature subset can identify the product. The recognition evaluation model is constructed based on machine learning (such as a classifier or regression model), and the performance of each feature subset is evaluated by simulating the product identification process. The subset with the highest recognition evaluation probability is selected from all feature subsets as the candidate robust feature.

[0042] Preferably, it is determined whether the recognition evaluation probability reaches a preset recognition target. The preset recognition target is a recognition accuracy target preset by the system. For example, the system requires the recognition accuracy rate to reach more than 95%. If the recognition probability of the candidate robust feature reaches or exceeds this preset target, the candidate robust feature will be formally selected as a robust feature for accurate identification of the goods. If the recognition probability of the candidate feature fails to reach the preset target, a new feature is introduced, that is, the previously eliminated interference recognition features and adjacent interference features are used as potentially useful additional features, some features are randomly selected from the additional features, and these features are added to the feature subset to construct a new feature combination to enhance the recognition ability and make up for the shortcomings of the current feature set. Then, through the recognition evaluation model, these newly constructed feature combinations are evaluated, their recognition probabilities are calculated, and it is determined whether the new feature combination is more effective than the previous one. If the new feature combination still fails to reach the preset recognition target, random sampling, construction and evaluation of feature combinations from the additional features are continued until a feature combination that meets the target requirements is found. When the recognition evaluation probability of a certain feature combination reaches or exceeds the preset recognition target, the feature combination is determined as the final robust feature.

[0043] In one possible implementation, step S440 further includes step S441, adjusting the recognition weight of the newly constructed feature combination, wherein the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination; step S442, evaluating the recognition probability of the newly constructed feature combination through the recognition evaluation model, and if the recognition evaluation probability does not reach the preset recognition target, continuing to add interference features to update the recognition feature combination; step S443, wherein the recognition weight is adjusted before each iteration to ensure that the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination, and repeating the iteration until the recognition evaluation probability reaches the preset recognition target.

[0044] Preferably, in order to avoid interference or weaken the influence of interfering features, the recognition weight of the newly constructed feature combination is adjusted, that is, the weight of each feature in the newly constructed feature combination is adjusted to optimize the recognition effect, and ensure that the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination, thereby reducing the influence of the interfering features in the recognition process and avoiding them from excessively interfering with the accurate recognition of the goods; the recognition probability of the adjusted new feature combination is calculated through the recognition evaluation model, that is, the accuracy and effect of these feature combinations in identifying goods are judged. The higher the recognition probability, the better the effect of the feature combination in identifying goods; the lower the recognition probability, the better the feature combination is in identifying goods; The performance is poor; if the recognition probability of the newly constructed feature combination does not reach the preset goal of the system (such as the recognition accuracy needs to reach a certain threshold), continue to randomly select more features from the interference features, add them to the current feature combination, and try to improve the recognition effect by expanding the feature set. In each iteration, adjust the weights before adding new interference features to ensure that the weights of the newly added interference features must always be smaller than the weights of the original recognition features to avoid too many interference factors affecting the final recognition effect. Each time the feature combination is adjusted and the recognition probability is calculated until the preset recognition goal is reached, stop the iteration, and regard the current feature combination as the final robust feature combination.

[0045] Step S500: using the robust features as identification features, performing loading task identification processing according to the quantity of goods.

[0046] Preferably, the obtained robust features are used as identification features, that is, each product is matched according to these robust features to ensure that the products can be accurately identified. For example, the shape and barcode of the product are used as core identification features. When the system scans the product, these features are used to distinguish different products and identify the loading task for the quantity of the products. The quantity of the products means that each order usually contains a variety of different products and their corresponding quantity information. The loading task requires the completion of the loading task according to the specific quantity of each product in the order. For example, the order may require the loading of 10 parts A and 5 parts B. The loading process not only needs to identify each product, but also needs to determine whether the loading task is completed according to the quantity information in the order. Specifically, the identified product features are compared with the quantity required in the order. For example, if the order requires loading 10 parts A, the system will automatically record the number of goods identified each time during loading, until a total of 10 parts A are identified, the loading task of the goods will be considered completed. If the number of identified goods is less than or more than the requirements in the order, the system will issue a prompt or warning to ensure that the goods are not over or under loaded. For example, if the order requires 5 parts B, and the system identifies that 6 parts B are loaded, the loader will be prompted to make adjustments. After each product is identified and its quantity is confirmed, the system will update the execution status of the task to ensure real-time tracking of the progress of the entire loading task; when the identification and quantity matching of all goods are completed, it is confirmed that the loading task has been completed, ensuring the accuracy and completeness of each loading task identification, thereby improving the automation and efficiency of the smart factory.

[0047] In the above, refer to Figure 1 A method for automatically identifying the amount of cargo for a smart factory according to an embodiment of the present invention is described in detail. Figure 2 A system for automatically identifying loading quantities for a smart factory according to an embodiment of the present invention is described.

[0048] According to an embodiment of the present invention, an automatic loading quantity identification system for a smart factory is used to solve the technical problem that the existing loading quantity identification cannot efficiently and accurately identify the combination of goods in the face of diversified and customized production needs, resulting in low loading accuracy and efficiency, and achieves the technical effect of improving the efficiency and accuracy of factory loading. An automatic loading quantity identification system for a smart factory includes: an order goods composition acquisition module 10, an interference identification feature determination module 20, a goods management feature library search module 30, a goods identification feature processing module 40, and a loading task identification processing module 50.

[0049] The order goods composition acquisition module 10 is used to connect to the task platform to obtain the loading task information, perform task decomposition on the loading task information, and obtain the order goods composition, which is the information of all goods that need to be assembled in the loading task, including the goods number and the goods quantity;

[0050] The interference identification feature determination module 20 is used to use the product number of the order product composition as an index to search the product management feature library to obtain the product identification feature, perform order interference feature identification on the product identification feature, and determine the interference identification feature;

[0051] A product management feature library search module 30 is used to search the product management feature library with the product identification feature as an index to obtain interfering neighboring products and their neighboring interference features in the smart factory;

[0052] The product identification feature processing module 40 is used to use the interference identification feature and the adjacent interference feature as constraint conditions to perform feature selection processing on the product identification feature to obtain a robust feature, wherein the robust feature is a core identification feature or a combination of core identification features that is not affected by the interference feature;

[0053] The loading task identification processing module 50 is used to use the robust features as identification features and perform loading task identification processing according to the quantity of goods.

[0054] The specific configuration of the interference identification feature determination module 20 will be described in detail below. The interference identification feature determination module 20 further includes: collecting the physical characteristics, image characteristics, and kit composition characteristics of all incoming goods in the smart factory, wherein the physical characteristics include weight, density, and material, the image characteristics include size, shape, color, texture, and light reflection characteristics, and the kit composition characteristics include the number of kit components and the installation and assembly position; encoding the physical characteristics, image characteristics, and kit composition characteristics, converting them into a storable format, establishing a mapping relationship between the product number and the feature code of the incoming goods, and constructing the product management feature library, wherein a multi-level mapping relationship between components and kits and kits and product numbers is established for the kit composition characteristics.

[0055] The specific configuration of the interference identification feature determination module 20 will be described in detail below. The interference identification feature determination module 20 may further include: partitioning according to the product number, physical feature, image feature, and kit composition feature; using the product number as a unique index, integrating it into the product number partition of the product management feature library; using the physical feature, image feature, and kit composition feature as the main index, and using the specific features under the feature as the auxiliary index, respectively, to establish a combined index; integrating the combined index corresponding to each feature category into the physical feature, image feature, and kit composition feature partition of the product management feature library.

[0056] The specific configuration of the interference identification feature determination module 20 will be described in detail below. The interference identification feature determination module 20 may further include: performing feature similarity calculations according to the physical features, image features, and kit composition features, respectively, to obtain the similarity of each feature category; establishing an N×N matrix, where N is the number of commodity categories, and each element represents the similarity of two commodities under a certain feature category; filling the matrix according to the feature category similarity, performing weighted average of the similarity of each pair of commodity combinations, and obtaining a comprehensive similarity value; assigning values ​​to the matrix using the comprehensive similarity value, and when the matrix element reaches a preset interference threshold, determining the corresponding feature as the interference identification feature.

[0057] Next, the specific configuration of the interference identification feature determination module 20 will be described in detail. The interference identification feature determination module 20 further includes: obtaining a set of loading abnormal cases, clustering the loading abnormal case set according to the factory operation process and the product number, and constructing an abnormal case cluster; fitting the abnormal characteristics and abnormal probabilities of the factory operation process and the product number through the abnormal case cluster, and obtaining the abnormal characteristics and abnormal probabilities of the factory operation process and the abnormal characteristics and abnormal probabilities of the product number; extracting the factory operation process of the loading task according to the loading task information, and matching the abnormal characteristics and abnormal probabilities of the process; matching the abnormal characteristics and abnormal probabilities of the product number according to the product number of the order product composition; matching the abnormal characteristics and abnormal probabilities of each pair of product combinations and their feature category similarity features according to the abnormal characteristics and abnormal probabilities of the matching process and the abnormal characteristics and abnormal probabilities of the product number, and calculating the average abnormal probability, which is the average of the abnormal probabilities of each product in the product combination in the feature category and the abnormal probability of the matching process; using the average abnormal probability to weight the comprehensive similarity, and using the weighting to perform similarity weighted calculation on each pair of product combinations to obtain the comprehensive similarity value.

[0058] The specific configuration of the product identification feature processing module 40 will be described in detail below. The product identification feature processing module 40 further includes: removing the interference identification feature and the adjacent interference feature as the elimination feature from the product identification feature to construct a pure identification feature set; randomly sampling from the pure identification feature set to construct a number of feature subsets, the subsets include different numbers of features or feature combinations, performing recognition probability evaluation on each random feature or feature combination through the recognition evaluation model to obtain the recognition evaluation probability, and screening the random sampling feature with the largest recognition evaluation probability as the candidate robust feature; judging whether the recognition evaluation probability reaches the preset recognition target, if so, taking the candidate robust feature as the robust feature; if not, taking the interference identification feature and the adjacent interference feature as the added feature, randomly selecting from the added features, adding the randomly added feature to the feature subset to obtain a newly constructed feature combination, and using the recognition evaluation model to evaluate the recognition probability of the newly constructed feature combination until the recognition evaluation probability reaches the preset recognition target, and determining the robust feature.

[0059] The specific configuration of the product identification feature processing module 40 will be described in detail below. The product identification feature processing module 40 may further include: adjusting the identification weight of the newly constructed feature combination, wherein the interference feature weight of the randomly added feature is less than the identification feature weight of the original feature combination; evaluating the identification probability of the newly constructed feature combination through the identification evaluation model, and if the identification evaluation probability does not reach the preset identification target, continuing to add interference features to update the identification feature combination; wherein the identification weight is adjusted before each iteration to ensure that the interference feature weight of the randomly added feature is less than the identification feature weight of the original feature combination, and repeating the iteration until the identification evaluation probability reaches the preset identification target.

[0060] An automatic loading quantity identification system for a smart factory provided by an embodiment of the present invention can execute an automatic loading quantity identification method for a smart factory provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0061] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0062] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for automatically identifying loading quantity for a smart factory, characterized in that: The method comprises: Connecting to the task platform to obtain loading task information, decomposing the loading task information, and obtaining the order goods composition, wherein the order goods composition is the information of all goods that need to be assembled in the loading task, including the goods number and the goods quantity; Using the product number of the order product composition as an index, searching the product management feature library to obtain product identification features, performing order interference feature identification on the product identification features, and determining the interference identification features; Using the commodity identification feature as an index, searching the commodity management feature library to obtain interfering neighboring commodities and their neighboring interference features in the smart factory; Taking the interference identification feature and the adjacent interference feature as constraint conditions, performing feature selection processing on the product identification feature to obtain a robust feature, wherein the robust feature is a core identification feature or a combination of core identification features that is not affected by the interference feature; The robust features are used as identification features to perform loading task identification processing according to the quantity of goods.

2. The method for automatically identifying the loading quantity for a smart factory according to claim 1, characterized in that: The searching of the product management feature library also includes: Collect the physical characteristics, image characteristics, and kit composition characteristics of all incoming goods in the smart factory. The physical characteristics include weight, density, and material. The image characteristics include size, shape, color, texture, and light reflection characteristics. The kit composition characteristics include the number of kit components and installation and assembly positions. The physical features, image features, and kit composition features are encoded and converted into a storable format, a mapping relationship between the product number and the feature code of the incoming goods is established, and the product management feature library is constructed, wherein a multi-level mapping relationship from component to kit and from kit to product number is established for the kit composition features.

3. The method for automatically identifying the loading quantity for a smart factory according to claim 2, characterized in that: Constructing the product management feature library also includes: Partitioning is performed according to the product number, physical features, image features, and kit composition features; The product number is used as a unique index and integrated into the product number partition of the product management feature library; The physical feature, the image feature, and the kit component feature are respectively used as the primary index, and the specific features under the feature are used as the auxiliary index, and a combined index is respectively established; The combined index corresponding to each feature category is respectively integrated into the physical feature, image feature, and kit composition feature partitions of the product management feature library.

4. The method for automatically identifying the loading quantity for a smart factory according to claim 2, characterized in that: Performing order interference feature identification on the product identification feature to determine the interference identification feature includes: According to the physical features, image features, and kit composition features, feature similarity calculations are performed respectively to obtain similarities of each feature category; Create an N×N matrix, where N is the number of product categories and each element represents the similarity between two products under a certain feature category; Filling the matrix according to the similarity of the feature categories, performing weighted average of the similarity of each pair of product combinations, and obtaining a comprehensive similarity value; The matrix is ​​assigned values ​​using the comprehensive similarity value, and when a matrix element reaches a preset interference threshold, a corresponding feature is determined as the interference identification feature.

5. The method for automatically identifying the loading quantity for a smart factory according to claim 4, characterized in that: The obtaining of the comprehensive similarity value further includes: Obtaining a set of loading abnormality cases, clustering the set of loading abnormality cases according to factory operation procedures and product numbers, and constructing abnormal case clusters; Through the abnormal case cluster, the abnormal characteristics and abnormal probability of the factory operation process and the product number are fitted respectively to obtain the abnormal characteristics and abnormal probability of the factory operation process and the abnormal characteristics and abnormal probability of the product number; Extracting the factory operation process of the loading task according to the loading task information, and matching the abnormal characteristics and abnormal probability of the process; According to the product number of the order product composition, match the abnormal characteristics and abnormal probability of the product number; According to the abnormal features and abnormal probabilities of the matching process and the abnormal features and abnormal probabilities of the product numbers, abnormal probability matching is performed on each pair of product combinations and the similarity of their feature categories, and the average abnormal probability is calculated. The average abnormal probability is the average of the abnormal probabilities of each product in the product combination in the feature category and the abnormal probability of the matching process; The comprehensive similarity is weighted using the abnormal probability mean, and the weighted similarity calculation is performed on each pair of product combinations to obtain the comprehensive similarity value.

6. The method for automatically identifying the loading quantity for a smart factory according to claim 1, characterized in that: Taking the interference identification feature and the adjacent interference feature as constraint conditions, performing feature selection processing on the product identification feature to obtain a robust feature includes: Eliminate the interference identification feature and the adjacent interference feature as elimination features from the product identification feature to construct a pure identification feature set; Randomly sampling from the pure recognition feature set to construct several feature subsets, the subsets include different numbers of features or feature combinations, performing recognition probability evaluation on each random feature or feature combination through the recognition evaluation model to obtain the recognition evaluation probability, and selecting the randomly sampled feature with the largest recognition evaluation probability as a candidate robust feature; Determine whether the recognition evaluation probability reaches a preset recognition target, and if so, use the candidate robust feature as the robust feature; When it is not reached, the interference identification feature and the adjacent interference feature are used as additional features, randomly selected from the additional features, and added to the feature subset using the randomly added features to obtain a new feature combination, and the recognition probability of the new feature combination is evaluated using the recognition evaluation model until the recognition evaluation probability reaches the preset recognition target, thereby determining the robust feature.

7. The method for automatically identifying the loading quantity for a smart factory according to claim 6, characterized in that: Using the recognition evaluation model to evaluate the recognition probability of the newly constructed feature combination until the recognition evaluation probability reaches the preset recognition target, further comprising: Adjusting the recognition weight of the newly constructed feature combination, wherein the interference feature weight of the randomly added feature is less than the recognition feature weight of the original feature combination; The recognition evaluation model is used to evaluate the recognition probability of the newly constructed feature combination. If the recognition evaluation probability does not reach the preset recognition target, interference features are continuously added to update the recognition feature combination. Among them, the recognition weight adjustment is performed before each iteration to ensure that the interference feature weight of the randomly added feature is smaller than the recognition feature weight of the original feature combination, and the iteration is repeated until the recognition evaluation probability reaches the preset recognition target.

8. An automatic loading quantity identification system for smart factories, characterized in that: The system is used to implement the method for automatically identifying the loading quantity for a smart factory according to any one of claims 1 to 7, and the system comprises: An order goods composition acquisition module is used to connect to the task platform to obtain loading task information, perform task decomposition on the loading task information, and obtain order goods composition, wherein the order goods composition is information on all goods that need to be assembled in the loading task, including the goods number and the goods quantity; An interference identification feature determination module is used to use the product number of the order product composition as an index to search the product management feature library, obtain the product identification feature, perform order interference feature identification on the product identification feature, and determine the interference identification feature; A product management feature library search module, used to search the product management feature library with the product identification feature as an index to obtain interfering neighboring products and their neighboring interference features in the smart factory; A product identification feature processing module, used to use the interference identification feature and the adjacent interference feature as constraints, perform feature selection processing on the product identification feature, and obtain a robust feature, wherein the robust feature is a core identification feature or a combination of core identification features that is not affected by the interference feature; The loading task identification processing module is used to use the robust features as identification features and perform loading task identification processing according to the quantity of goods.

Images