Intelligent commodity warehousing management system and method

Abstract

The invention relates to a commodity warehousing management technology, and discloses an intelligent commodity warehousing management system and method, which are used for efficiently and accurately verifying warehoused commodities and improving the accuracy, efficiency and fine management level of warehousing management. According to the scheme, firstly, commodity information is read from a commodity label, preliminary verification is conducted on the shape, size and color features of a commodity through a point cloud recognition algorithm, if verification is passed, a subsequent detailed comparison process is started, and if not, it is judged that the commodity is a suspected counterfeit commodity; in the detailed comparison process, a pre-trained classification model is utilized to identify commodity categories, detailed feature comparison is carried out, if the comparison is passed, it is judged that the commodities are authentic commodities, the commodities are put in storage according to the commodity categories, and commodity information is synchronized to a block chain database; and carrying out risk early warning and processing on the suspected counterfeit commodity, and recording related information to a block chain database. In addition, based on the point cloud data of commodity distribution in the warehouse, the actual distribution condition of the commodities in the warehouse is analyzed, and a warehouse layout optimization suggestion is provided.

Description

Technical Field

[0001] The present invention relates to the technology of goods warehousing management, and particularly to an intelligent goods warehousing management system and method. Background Art

[0002] In the current consumer market, the problem of fake and counterfeit goods is becoming increasingly prominent, bringing many challenges to enterprises. To address this issue, enterprises need to strictly verify and manage goods during the warehousing process to ensure that the inventory only contains genuine products. In the existing technology, enterprises mainly manage warehousing through the following methods:

[0003] (1) Automatic identification technology

[0004] Automatic identification technology is the core of intelligent warehousing management, mainly including barcode identification and RFID (Radio Frequency Identification) and other technologies. Barcode identification technology can quickly and accurately obtain product information by scanning the barcode on the product.

[0005] (2) Information technology

[0006] Information technology plays an important role in intelligent warehousing management. By introducing advanced information systems, such as the WMS (Warehouse Management System) warehouse management system, real-time monitoring and analysis of inventory data can be achieved. The WMS system can provide detailed operation records, inventory status reports, and data analysis tools to help managers better grasp the inventory situation. In addition, information technology can be integrated with other systems, such as the ERP (Enterprise Resource Planning) enterprise resource planning system, SCM (Supply Chain Management) supply chain management system, etc., to achieve information sharing and collaboration. This helps improve the efficiency and accuracy of inventory management and reduce labor costs.

[0007] (3) Internet of Things technology

[0008] The Internet of Things technology connects goods with information systems through devices such as sensors and RFID tags, realizing real-time monitoring and tracking of goods. This helps prevent the mixing of fake and counterfeit goods and improves the transparency and security of inventory management. The Internet of Things technology can also be combined with other technologies, such as big data analysis and artificial intelligence, to achieve a higher level of inventory management and prediction. For example, by analyzing historical sales data and market trends, future inventory requirements can be predicted, and more reasonable procurement plans and inventory strategies can be formulated.

[0009] Although the existing warehousing management technologies have improved the warehousing efficiency and the accuracy of inventory management to a certain extent, there are still some defects, specifically as follows:

[0010] (1) Automatic identification technology: Barcode identification technology is vulnerable to barcode tearing or damage, resulting in difficult identification and affecting the warehousing efficiency. Although RFID technology has a long identification distance and strong identification ability, it has a high cost and there are identification obstacles for certain special material goods, which limits its wide application.

[0011] (2) Information technology: During the intelligent warehousing process, if the information system fails, it may lead to incorrect or missing entry of commodity information, affecting the accuracy of inventory data and bringing inconvenience to subsequent sales and logistics links. In addition, the existing information system may have delays in data update, resulting in inventory data not being able to reflect the actual situation in real time and affecting the refinement level of inventory management.

[0012] (3) Internet of Things technology: With the continuous development of technology, new means and ways of counterfeiting and imitating goods emerge in an endless stream, and the update of existing Internet of Things technology may lag behind the emergence of these new means, resulting in a decline in technical prevention ability. In addition, the data generated by the Internet of Things technology is huge, and the complexity of data processing and analysis is relatively high, requiring more powerful computing power and more efficient algorithm support.

[0013] In addition, the existing warehousing management system has deficiencies in optimizing the inventory layout, unable to dynamically adjust the warehouse layout according to the real-time distribution of goods, resulting in low space utilization and low outbound efficiency.

[0014] In summary, although the existing warehousing management technology has improved the warehousing efficiency and the accuracy of inventory management to a certain extent, there are still deficiencies in anti-counterfeiting, traceability, data update, technology update, and inventory layout optimization. These problems not only affect the operation efficiency of enterprises, but also may lead to counterfeit and imitated goods mixing into the inventory, damaging the brand image of enterprises and the rights and interests of consumers. Therefore, enterprises need a more efficient and intelligent warehousing management system to cope with these challenges. Summary of the Invention

[0015] The technical problem to be solved by the present invention is: to propose a commodity intelligent warehousing management system and method to efficiently and accurately verify the incoming goods and improve the accuracy, efficiency, and refinement management level of warehousing management.

[0016] The technical solution adopted by the present invention to solve the above technical problems is:

[0017] On the one hand, the present invention provides a commodity intelligent warehousing management method, including the following steps:

[0018] S1. Read the commodity information from the commodity label and use the point cloud recognition algorithm to preliminarily verify the shape, size, and color characteristics of the commodity. If the verification is passed, go to step S2; otherwise, determine it as a suspected counterfeit commodity and go to step S4;

[0019] S2. Use the pre-trained classification model to identify the product category and conduct a detailed feature comparison. If the comparison passes, determine that the product is genuine and proceed to step S3; otherwise, determine it as a suspected counterfeit product and proceed to step S4;

[0020] S3. Store the products determined to be genuine according to the product category and synchronize the product information to the blockchain database;

[0021] S4. Conduct risk warning and handling for the products determined to be suspected counterfeit products, and record the relevant information in the blockchain database.

[0022] Further, in step S1, the use of the point cloud recognition algorithm to preliminarily verify the shape, size, and color features of the product includes:

[0023] Use a 3D scanner to scan the product to obtain the point cloud data of the product;

[0024] Generate a 3D model of the product based on the point cloud data of the product;

[0025] Extract the shape, size, and color features of the 3D model respectively, and compare them with the shape, size, and color features of the corresponding 3D models of the products in the pre-established genuine product database.

[0026] Further, in step S2, the detailed feature comparison includes: refined feature comparison between the 3D model of the product and the corresponding 3D models of the products in the genuine product database, texture comparison between the product and the genuine product, and brand logo comparison between the product and the genuine product.

[0027] Further, in step S4, the conduct of risk warning and handling for the products determined to be suspected counterfeit products includes:

[0028] When the product is determined to be a suspected counterfeit product, store the product in isolation and automatically trigger a warning to notify the management staff for handling.

[0029] Further, the method further includes the step:

[0030] S5. Optimize the storage layout of the products in the warehouse:

[0031] Obtain the point cloud data of the product distribution in the warehouse, analyze the actual distribution of the products in the warehouse, including the stacking method of the products and the relative positions of adjacent products;

[0032] According to the actual distribution of the products in the warehouse, identify the separate idle spaces and adjustable idle spaces in the warehouse. Among them, the separate idle space refers to the space in the warehouse that is completely unoccupied, and the adjustable idle space refers to the idle space above or around the stored products that can accommodate other products;

[0033] Generate recommended storage locations based on the individual idle space and adjustable idle space in the warehouse, combined with the size and category of the goods to be warehoused, and optimize the layout of goods storage.

[0034] On the other hand, the present invention also provides an intelligent goods warehousing management system, which includes:

[0035] An information reading module for reading goods information from goods tags;

[0036] A preliminary verification module for preliminarily verifying the shape, size, and color characteristics of goods using a point cloud recognition algorithm;

[0037] A classification module for identifying the category of goods using a pre-trained classification model;

[0038] A detailed verification module for performing a detailed feature comparison on the goods that have passed the preliminary verification, and judging whether the goods are genuine according to the comparison results;

[0039] A warehousing processing module for warehousing the goods determined to be genuine according to the category of the goods, and synchronizing the goods information to the blockchain database;

[0040] An exception handling module for performing risk early warning and handling on the goods determined to be suspected of being counterfeit, and recording relevant information in the blockchain database.

[0041] Furthermore, the preliminary verification module is specifically used for:

[0042] Scanning the goods using a 3D scanner to obtain the point cloud data of the goods;

[0043] Generating a 3D model of the goods based on the point cloud data of the goods;

[0044] Respectively extracting the shape, size, and color characteristics of the 3D model, and comparing them with the shape, size, and color characteristics of the corresponding 3D model of the goods in the pre-established genuine goods database.

[0045] Furthermore, the detailed verification module performs a detailed feature comparison on the goods that have passed the preliminary verification, including:

[0046] Performing a refined feature comparison between the 3D model of the goods and the corresponding 3D model of the goods in the genuine goods database, comparing the texture between the goods and the genuine goods, and comparing the brand logo between the goods and the genuine goods.

[0047] Furthermore, the exception handling module performs risk early warning and handling on the goods determined to be suspected of being counterfeit, including:

[0048] When a commodity is determined to be a suspected counterfeit commodity, the commodity is stored in isolation and a warning is automatically triggered to notify the management staff for handling.

[0049] Furthermore, the system further includes:

[0050] A layout optimization module for optimizing the layout of commodity storage in the warehouse:

[0051] Obtain the point cloud data of the commodity distribution in the warehouse, analyze the actual distribution of commodities in the warehouse, including the stacking method of commodities and the relative positions of adjacent commodities;

[0052] According to the actual distribution of commodities in the warehouse, identify the separate idle spaces and adjustable idle spaces in the warehouse. Among them, the separate idle space refers to the space in the warehouse that is completely unoccupied, and the adjustable idle space refers to the idle space above or around the stored commodities that can accommodate other commodities;

[0053] According to the separate idle spaces and adjustable idle spaces in the warehouse, combined with the size and category of the commodities to be warehoused, generate recommended storage locations to optimize the commodity storage layout.

[0054] The beneficial effects of the present invention are:

[0055] (1) By using the point cloud recognition algorithm, combined with deep learning, machine learning and computer vision technologies, the rapid and accurate recognition of commodity features such as shape, size, and color is realized, thus greatly improving the warehousing efficiency, enabling a large number of commodities to be preliminarily processed in a short time. On the basis of preliminary classification, the system further conducts detailed feature comparison to ensure the accuracy of the recognition results. This dual verification mechanism effectively reduces the risk of counterfeit commodities entering the warehouse.

[0056] (2) Combined with blockchain technology, the system records key data such as the production information and warehousing time of commodities, ensuring the accuracy and immutability of the data, and through the establishment of a risk warning mechanism, it can timely detect and handle suspected counterfeit commodities, improving the safety of commodities.

[0057] (3) It can analyze the actual distribution of commodities in the warehouse based on the point cloud data of the commodity distribution in the warehouse, and automatically identify the idle spaces according to the actual distribution, and put forward suggestions for optimizing the warehouse layout, thereby improving the space utilization rate and warehousing efficiency. It can also formulate more reasonable inventory management strategies by analyzing the distribution of various commodities, such as regular inventory checks, inventory warnings, and management of hot-selling commodities. Description of the Drawings

[0058] Figure 1 It is a flow chart of the intelligent warehousing management method for commodities in the present invention;

[0059] Figure 2 This is the structural block diagram of the intelligent goods warehousing management system in the present invention. Specific implementation manners

[0060] The present invention aims to provide an intelligent goods warehousing management system and method, which can efficiently and accurately verify the incoming goods, and improve the accuracy, efficiency and refined management level of warehousing management. Its core idea is: by combining advanced point cloud recognition technology, deep learning algorithms, blockchain technology and intelligent robot automation operations, to achieve efficient and accurate verification and management of incoming goods, ensuring that only verified genuine goods can enter the inventory system, and at the same time effectively identifying and isolating suspected counterfeit goods.

[0061] More specifically, the present invention preliminarily verifies the shape, size and color characteristics of goods through point cloud recognition technology, thus greatly improving the warehousing efficiency, enabling a large number of goods to be preliminarily processed in a short time. On the basis of preliminary classification, through the refined feature comparison between the point cloud three-dimensional model of the goods and the corresponding three-dimensional model of the goods in the genuine product database, the texture comparison between the goods and the genuine products, and the brand logo comparison between the goods and the genuine products, to meticulously verify whether the goods are genuine products, so as to effectively distinguish genuine products from suspected counterfeit goods, significantly improving the recognition efficiency and accuracy; for genuine products, the automatic handling and warehousing operations of the goods can be realized through intelligent robots and automation equipment, reducing manual intervention, improving the warehousing efficiency and reducing the labor cost; in addition, the present invention also utilizes the immutability and traceability of the blockchain to upload the information of the goods to the blockchain for storage, providing a reliable guarantee for the identity verification and data management of the goods. At the same time, based on the risk warning mechanism, when identifying suspected counterfeit goods, they are isolated and the management personnel are notified in a timely manner; finally, the present invention can also analyze the actual distribution of goods in the warehouse based on the point cloud data of the goods distribution in the warehouse, automatically identify the idle space according to the actual distribution situation, and put forward suggestions for optimizing the warehouse layout, thereby improving the space utilization rate and warehousing efficiency. It can also formulate more reasonable inventory management strategies, such as regular inventory taking, inventory warning, management of hot-selling goods, etc. by analyzing the distribution of various goods.

[0062] In summary, the solution of the present invention covers the entire process from goods information reading, preliminary verification, classification comparison, risk warning to intelligent warehousing and inventory management. Each module works together to form a complete intelligent warehousing management system, which can comprehensively solve the problems of counterfeit goods mixing in and inventory management optimization.

[0063] The intelligent goods warehousing management method provided by the present invention is shown in Figure 1 and includes the following implementation processes:

[0064] S1. Read the goods information and conduct preliminary verification:

[0065] In this step, RFID can be used to read the production information on the product label, such as the production date, batch number, manufacturer, etc. These information can be used as auxiliary verification for preliminary verification and can also be used as part of the product information uploaded to the blockchain after being determined to be genuine.

[0066] After reading the product information, this step also needs to conduct a preliminary verification of the product, aiming to quickly identify the product and preliminarily screen genuine products and suspected counterfeit products. In an exemplary implementation, a high-precision 3D scanner can be used to scan the product omnidirectionally to obtain the point cloud data on the product surface. The point cloud data is composed of a large number of points in three-dimensional space, and each point contains its coordinates (X, Y, Z) in space and color information (R, G, B). The scanning process includes scanning the product from multiple angles to ensure that the complete surface information of the product is obtained.

[0067] Then, the product is verified based on the point cloud recognition algorithm to ensure that its features such as shape, size, and color match those of genuine products. The point cloud recognition algorithm aims to automatically identify the product through features such as its shape, size, and color to improve the warehousing efficiency and ensure accuracy. This algorithm combines deep learning, machine learning, and computer vision technologies and can automatically adjust the recognition parameters according to different product features.

[0068] Among them, for the shape feature: a 3D model of the product is generated using the point cloud data, and the product is identified by comparing the geometric shapes of the 3D models; for the size feature: the size of the product is determined by measuring parameters such as the distance from each point in the point cloud to the center point; for the color feature: the point cloud data is converted into feature vectors such as color histograms or color moments for identifying the color of the product.

[0069] By separately extracting the shape, size, and color features of the 3D model and comparing them with the shape, size, and color features of the corresponding 3D models of products in the pre-established genuine product database, the product is preliminarily verified whether it is genuine. If the preliminary verification passes, it enters the detailed verification in step S2. If the preliminary verification fails, it is determined as a suspected counterfeit product and enters the risk warning mechanism process in step S4.

[0070] It should be noted that in this step, a laser rangefinder can also be used instead of a high-precision 3D scanner for product size measurement. Although the accuracy is slightly lower than that of the 3D scanner, in some cases, such as when the product has a regular shape and obvious size differences, the laser rangefinder can also provide sufficient information for the preliminary screening of counterfeit products.

[0071] In addition, in terms of the method for obtaining the 3D model of the commodity, multiple cameras can be combined to capture images of the commodity, and then the 3D model of the commodity can be generated through image stitching and 3D reconstruction technologies. This method has a lower cost, but requires more complex algorithms to process the image data.

[0072] S2. Commodity category identification and detailed feature comparison:

[0073] In this step, based on the preliminary verification, more detailed feature comparison of the commodity is carried out, such as the refined feature comparison between the 3D model of the commodity and the corresponding 3D model of the commodity in the genuine product database, the texture comparison between the commodity and the genuine product, the brand logo comparison between the commodity and the genuine product, etc. This stage focuses on accuracy to ensure the reliability of the identification result. After the detailed feature comparison, if the comparison passes, the commodity is determined to be a genuine product and enters step S3; otherwise, it is determined to be a suspected counterfeit commodity and enters step S4.

[0074] In specific implementation, deep learning algorithms (such as CNN or RNN) can be used to extract features and classify point cloud data to improve the generalization ability and robustness of the algorithm. And according to historical data and commodity features, continuously optimize the parameters of the algorithm, such as feature extraction methods, classifier parameters, etc., to further improve the identification performance and stability of the algorithm.

[0075] It should be noted that traditional image processing technologies (such as edge detection, contour extraction, etc.) combined with machine learning algorithms (such as SVM, decision tree, etc.) can also be used for commodity identification. Although this method is not as accurate as the deep learning model, it can also achieve good identification results in certain specific scenarios.

[0076] In addition, two-dimensional code or bar code recognition technology can also be introduced as an auxiliary means for commodity identity verification. Although two-dimensional codes or bar codes are easily copied, when combined with other anti-counterfeiting means (such as special materials, hidden information, etc.), the recognition rate of counterfeit commodities can still be improved.

[0077] S3. Storage of genuine products in the warehouse and uploading of information to the blockchain:

[0078] In this step, the commodities determined to be genuine are warehoused according to the commodity category, and the commodity information is synchronized to the blockchain database. In an exemplary implementation scheme, intelligent robots can be used to automatically send the genuine products to the designated storage locations corresponding to the commodity categories. At the same time, the information of the warehoused commodities is synchronized to the blockchain database in real time to ensure the accuracy and immutability of the data. When the commodities are out of the warehouse or there are quality problems, the source and warehousing process of the commodities can be quickly traced so as to take timely measures.

[0079] It is understandable that automated guided vehicles (AGVs) can also be used to replace intelligent robots for commodity handling. AGVs have lower costs and are more flexible in some warehouse environments. Or an intelligent warehousing method assisted by manual labor can be adopted, that is, the system provides warehousing suggestions, and the warehouse administrator operates manually according to the suggestions. Although this method reduces the degree of automation, it is more suitable for small warehouses or scenarios with limited budgets.

[0080] S4. Perform special processing on suspected counterfeit commodities and upload the information to the blockchain:

[0081] In this step, special processing such as isolated storage is performed on the determined suspected counterfeit commodities to avoid confusion with genuine products. And trigger a risk warning mechanism, such as notifying relevant personnel for further inspection, such as unpacking inspection, chemical analysis, etc. And record the information of suspected counterfeit commodities on the blockchain for subsequent traceability.

[0082] S5. Optimize the layout of commodity storage in the warehouse:

[0083] In this step, by regularly or as needed obtaining the point cloud data of the commodity distribution in the warehouse, analyze the actual distribution of commodities in the warehouse, including the stacking method of commodities and the relative positions of adjacent commodities; according to the actual distribution of commodities in the warehouse, identify the separate idle spaces and adjustable idle spaces in the warehouse; according to the separate idle spaces and adjustable idle spaces in the warehouse, combined with the size and category of the to-be-stored commodities, generate recommended storage locations, such as concentrating similar commodities in the same area for more efficient management and search; place the hot-selling commodities in easily accessible positions to improve the outbound efficiency.

[0084] In addition, an inventory warning mechanism can also be established. When the inventory level is lower than the safety inventory level, a warning is issued in a timely manner and relevant personnel are notified, which can help enterprises replenish goods in a timely manner and adjust the inventory strategy to avoid the risks of out-of-stock and overstocked inventory.

[0085] In summary, the present invention provides a commodity intelligent warehousing management method that can efficiently and accurately verify inbound commodities, improving the accuracy, efficiency, and refined management level of warehousing management.

[0086] Correspondingly, the present invention also provides a commodity intelligent warehousing management system, see Figure 2 and it includes the following components:

[0087] An information reading module for reading commodity information from commodity labels;

[0088] A preliminary verification module for preliminarily verifying the shape, size, and color characteristics of commodities using a point cloud recognition algorithm;

[0089] A classification module for identifying product categories using a pre-trained classification model;

[0090] A detailed verification module for performing a detailed feature comparison on the products that have passed the preliminary verification, and determining whether the product is genuine based on the comparison results;

[0091] An inbound processing module for warehousing the products determined to be genuine according to the product categories, and synchronizing the product information to the blockchain database;

[0092] An exception handling module for performing risk early warning and handling on the products determined to be suspected of being counterfeit, and recording the relevant information in the blockchain database;

[0093] A layout optimization module for optimizing the layout of product storage in the warehouse.

[0094] Since the functions of the various modules in the above system correspond to the steps described in the foregoing intelligent inbound management method for products, the specific implementation of the functions will not be elaborated herein.

[0095] Embodiment

[0096] In this embodiment, taking the product to be warehoused as a liquor product as an example, the implementation method of the intelligent inbound management solution is specifically described:

[0097] A certain liquor enterprise adopts an intelligent inbound management system to conduct strict identity verification on the inbound liquor products.

[0098] Before warehousing, the system first reads the production information on the product label, such as the production date, batch number, place of origin, etc. At the same time, the point cloud recognition algorithm is used to verify the shape, size, and color of the wine bottle (and the wine bottle's packaging box).

[0099] Product label information: The production date is May 1, 2023, the batch number is 001234, and the place of origin is Sichuan, China.

[0100] Point cloud recognition data:

[0101] Shape features: The wine bottle is a typical cylinder, slightly wider at the bottom, gradually tapering at the top, and has a special thread design at the bottleneck.

[0102] Size features: The height of the wine bottle is 30 cm, the diameter is 8 cm, and the diameter of the bottleneck is 3 cm.

[0103] Color features: The wine bottle is dark red, and the color histogram shows that its main color components are red and dark colors.

[0104] Based on the point cloud recognition algorithm, the system compares the read shape, size, and color features with the data in the genuine product database, confirms that the product is genuine, and allows it to enter the next stage of processing.

[0105] After passing the preliminary verification, the system enters the stage of commodity classification and feature comparison. First, a pre-trained classification model is used to quickly identify commodity categories, such as red wine, white wine, beer, etc. Then, based on the preliminary classification, a more detailed feature comparison of the commodity is carried out, such as accurately measuring the dimensions of the wine bottle and analyzing the details of the 3D model.

[0106] For example: Quick identification result: This commodity is identified as the red wine category.

[0107] Detailed feature comparison: Dimension comparison: After accurate measurement, the height and diameter of the wine bottle are exactly the same as the data in the genuine product database. Analysis of 3D model details: Details such as the thread design of the wine bottle and the position and size of the bottle label are consistent with the genuine product.

[0108] Comparison result: The system confirms that this commodity passes the verification in both the classification and feature comparison stages and meets the requirements of genuine products.

[0109] At the same time, upload blockchain records: Key data such as the production information (production date, batch number, place of origin) of this commodity and the warehousing time (14:00 on May 5, 2023) are recorded on the blockchain.

[0110] When warehousing and storing, the system is set to conduct a comprehensive point cloud data scan of all commodities in the warehouse every weekend (or after each storage). During the scan, the system captures the three-dimensional shape, dimensions, and position information of each commodity and generates a detailed point cloud data set. The point cloud data set reflects the overall volume of the commodities stored in the warehouse. The system can obtain the overall volume (including its shape) of all commodities in the warehouse and optimize the storage layout for the next incoming commodity.

[0111] By comparing the point cloud data, the system can accurately identify the individual idle spaces in the warehouse. For example, in area B, due to an increase in the sales volume of a certain commodity, there are a large number of empty spaces on the shelf originally used to store this commodity, and these empty spaces are individual idle spaces.

[0112] In addition to individual idle spaces, the system can also identify adjustable idle spaces based on the dimensions and stacking methods of the commodities. For example, in area C, although the shelf is full of commodity A, the space above commodity A is not enough to place another layer of commodity A. However, the system finds that if commodity B (with a smaller size than commodity A) is placed above commodity A, this part of the space can be fully utilized. Therefore, this part of the space is an adjustable idle space.

[0113] Based on the above analysis, the system will recommend the optimal storage location for newly incoming goods according to the distribution of goods in the current warehouse, the individual idle space, and the adjustable idle space. For example, for the newly incoming Product C (with moderate size and high expected sales volume), the system will recommend storing it in the individual idle space just identified in Area A to make full use of the space and improve the picking efficiency.

[0114] In addition to recommending storage locations for new goods, the system will also propose optimization suggestions for the warehouse layout based on the long-term data analysis results. For example, the system may recommend concentrating goods with high sales volume near the entrance of the warehouse to respond to order demands more quickly; or recommend storing goods with large weight and large volume on the lower floors of the warehouse or near the exit to reduce handling costs and improve the outbound efficiency.

[0115] Finally, it should be noted that the above embodiments are only preferred embodiments and are not intended to limit the present invention. It should be pointed out that for those of ordinary skill in the art in this technical field, without departing from the spirit and scope protected by the claims of the present invention, several modifications, equivalent replacements, improvements, etc. can be made, and all of them should be included in the protection scope of the present invention.

Claims

1. A method for intelligent warehousing management of commodities, characterized in that, Including the following steps: S1. Read product information from the product label, and use a point cloud recognition algorithm to preliminarily verify the shape, size, and color features of the product. If the verification passes, proceed to step S2; otherwise, determine it as a suspected counterfeit product and proceed to step S4; S2. Use a pre-trained classification model to identify the product category and conduct a detailed feature comparison. If the comparison passes, determine the product as a genuine product and proceed to step S3; otherwise, determine it as a suspected counterfeit product and proceed to step S4; S3. Warehousing the products determined to be genuine according to the product category, and synchronize the product information to the blockchain database; S4. Conduct risk warning and handling for the products determined to be suspected counterfeit products, and record the relevant information in the blockchain database.

2. A method for intelligent warehousing management of products according to claim 1, characterized in that In step S1, the use of the point cloud recognition algorithm to preliminarily verify the shape, size, and color features of the product includes: Scanning the product using a 3D scanner to obtain the point cloud data of the product; Generating a 3D model of the product based on the point cloud data of the product; Extracting the shape, size, and color features of the 3D model respectively, and comparing them with the shape, size, and color features of the corresponding 3D model of the product in the pre-established genuine product database.

3. A method for intelligent warehousing management of products according to claim 2, characterized in that In step S2, the detailed feature comparison includes: refined feature comparison between the 3D model of the product and the corresponding 3D model of the product in the genuine product database, texture comparison between the product and the genuine product, and brand logo comparison between the product and the genuine product.

4. The intelligent warehousing management method for commodities according to claim 1, wherein In step S4, the risk warning and handling for the products determined to be suspected counterfeit products includes: When the product is determined to be a suspected counterfeit product, isolate and store the product, and automatically trigger a warning to notify the management staff for handling.

5. A method for intelligent warehousing management of products according to any one of claims 1-4, characterized in that This method further includes the step: S5. Optimize the storage layout of products in the warehouse: Obtain the point cloud data of the product distribution in the warehouse, analyze the actual distribution of products in the warehouse, including the stacking method of products and the relative positions of adjacent products; According to the actual distribution of products in the warehouse, identify the separate idle spaces and adjustable idle spaces in the warehouse. Among them, the separate idle space refers to the space in the warehouse that is completely unoccupied, and the adjustable idle space refers to the idle space above or around the stored products that can accommodate other products; According to the separate idle spaces and adjustable idle spaces in the warehouse, combined with the size and category of the products to be warehoused, generate recommended storage locations to optimize the product storage layout.

6. A commodity intelligent warehousing management system, characterized in that, Including: An information reading module for reading product information from the product label; A preliminary verification module for preliminarily verifying the shape, size, and color features of the product using a point cloud recognition algorithm; A classification module for identifying the product category using a pre-trained classification model; A detailed verification module for conducting a detailed feature comparison on the products that have passed the preliminary verification, and judging whether the product is a genuine product according to the comparison result; The warehousing processing module is used to warehousing the products determined to be genuine according to the product categories, and synchronize the product information to the blockchain database; The exception handling module is used to conduct risk early warning and handling for the products determined to be suspected of being counterfeit, and record the relevant information to the blockchain database.

7. A smart warehousing management system for products as claimed in claim 6, wherein, The preliminary verification module is specifically used for: Scanning the product using a 3D scanner to obtain the point cloud data of the product; Generating a 3D model of the product based on the point cloud data of the product; Extracting the shape, size, and color features of the 3D model respectively, and comparing them with the shape, size, and color features of the corresponding 3D models of the products in the pre-established genuine product database.

8. A smart warehousing management system for products as claimed in claim 7, wherein, The detailed verification module conducts detailed feature comparison on the products that have passed the preliminary verification, including: Conducting refined feature comparison between the 3D model of the product and the corresponding 3D models of the products in the genuine product database, comparing the texture between the product and the genuine product, and comparing the brand logo between the product and the genuine product.

9. A smart warehousing management system for products as claimed in claim 6, wherein, The exception handling module conducts risk early warning and handling for the products determined to be suspected of being counterfeit, including: When the product is determined to be a suspected counterfeit product, isolating the product for storage, automatically triggering an early warning, and notifying the management personnel for handling.

10. A smart warehousing management system for products as claimed in any one of claims 6-9, wherein, This system further includes: The layout optimization module is used to optimize the storage layout of the products in the warehouse: Obtaining the point cloud data of the product distribution in the warehouse, analyzing the actual distribution of the products in the warehouse, including the stacking method of the products and the relative positions of the adjacent products; Identifying the separate idle spaces and adjustable idle spaces in the warehouse according to the actual distribution of the products in the warehouse, wherein, The separate idle space refers to the space in the warehouse that is completely unoccupied, and the adjustable idle space refers to the idle space above or around the stored products that can accommodate other products; Generating recommended storage locations according to the separate idle spaces and adjustable idle spaces in the warehouse, combined with the size and category of the products to be warehoused, and optimizing the storage layout of the products.

Images