Animation surrounding inventory management method based on image recognition

Abstract

The invention belongs to the technical field of image recognition, and particularly relates to a cartoon surrounding inventory management method based on image recognition, which comprises the following steps: inputting a collected real object image into an image recognition model, extracting image features by the image recognition model based on a convolutional neural network, and comparing the image features with an image feature database; identifying the types and the number of commodities; in combination with an identification result and pre-stored inventory data, inventory information is updated in real time, and an inventory early warning signal is automatically generated if it is identified that the number of inventory commodities changes abnormally; parameters of the image recognition model are dynamically adjusted according to the recognition accuracy, and if the recognition accuracy is lower than a preset threshold value, a model retraining process is started; the updated inventory data is synchronized to the sales platform, the storage scheduling system and the financial management system through the inventory management system, multi-terminal sharing and synchronous updating of the inventory data are realized, and the effects of realizing automatic management of the whole process from warehousing, storage to ex-warehouse of commodities and greatly improving the efficiency of inventory management are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to an animation peripheral inventory management method based on image recognition. Background Art

[0002] Current inventory management for anime-related merchandise relies primarily on manual counting or barcoding, which is inefficient, error-prone, and costly. This is especially true for large quantities of goods and frequently changing inventory, where manual counting is a massive workload and difficult to update in real time. With the advancement of image recognition technology, automated inventory management using cameras or scanners to identify product images has become a new and efficient method, significantly reducing manual workload and improving the accuracy and timeliness of inventory counts.

[0003] In recent years, image recognition technology has made significant progress in various application scenarios, particularly in retail, e-commerce, and warehousing and logistics. Using image recognition technology, products can be automatically classified and identified, enabling rapid inventory updates. By combining it with existing barcode or RFID technologies, image recognition can further improve the accuracy of inventory management. However, existing inventory management systems often lack efficient product identification and real-time inventory updates, particularly in large-scale inventory management and the automated management of a wide variety of products.

[0004] To address these issues, this paper introduces advanced image recognition technology to provide an image-based inventory management method for anime-related merchandise, aiming to improve the efficiency and accuracy of inventory management. This method not only automatically identifies and categorizes anime-related merchandise but also effectively updates inventory information, automating the entire process from entry, storage, and shipment. This significantly improves the efficiency and flexibility of inventory management, meeting the complex and diverse needs of anime-related merchandise.

[0005] In response to the above technical defects, a solution for animation peripheral inventory management method based on image recognition is proposed. Summary of the Invention

[0006] In order to solve the above problems, the present invention provides the following technical solutions:

[0007] A method for managing animation peripheral inventory based on image recognition, comprising:

[0008] The collected physical image is input into the image recognition model, which extracts image features based on a convolutional neural network and compares them with the image feature database to identify the type and quantity of the product;

[0009] Combining the recognition results with pre-stored inventory data, the inventory information is updated in real time. If an abnormal change in the quantity of inventory items is identified, an inventory warning signal is automatically generated.

[0010] Dynamically adjust the parameters of the image recognition model based on recognition accuracy. If the recognition accuracy falls below a preset threshold, the model retraining process is initiated.

[0011] The updated inventory data is synchronized to the sales platform, warehouse scheduling system and financial management system through the inventory management system, realizing multi-terminal sharing and synchronous updating of inventory data.

[0012] Furthermore, the step of inputting the captured physical image into an image recognition model, wherein the image recognition model extracts image features based on a convolutional neural network, includes using a high-resolution camera or other image acquisition device to capture the physical image of the product. Ensure sufficient lighting and an appropriate angle to obtain a clear image;

[0013] Adjust the collected images, including denoising, adjusting brightness and contrast, and cropping, to optimize image quality and improve recognition accuracy. The pre-processed images are input into an image recognition model based on a convolutional neural network. The convolutional neural network extracts image features layer by layer through multi-layer convolution and pooling operations, and compares the extracted image features with a pre-established image feature database.

[0014] By comparison, find the product that best matches the current image features. Based on the comparison results, determine the type of product. By analyzing the number of products in the image, determine the quantity of products. The identified product type and quantity are output for subsequent inventory management.

[0015] Furthermore, the step of combining the recognition result with pre-stored inventory data to update inventory information in real time and automatically generating an inventory warning signal if an abnormal change in the quantity of inventory goods is identified includes obtaining pre-stored inventory data from the inventory management system, including the type, quantity and storage location of the goods, and comparing the image recognition result with the pre-stored inventory data;

[0016] Based on the comparison results, the inventory information in the inventory management system is updated to ensure the real-time and accuracy of inventory data. A threshold for inventory quantity changes is set. When the difference between the identified product quantity and the pre-stored data exceeds the threshold, an abnormal change alert is triggered. Abnormal changes may include sudden increases or decreases in inventory quantity that exceed the normal range.

[0017] When abnormal changes are detected, the system automatically generates an inventory warning signal. The warning signal can be sent in a variety of ways. The management personnel who receive the warning signal will take corresponding measures according to the specific situation, continuously collect images and update inventory data, maintain real-time synchronization of inventory information, regularly evaluate the performance of the inventory management system, optimize the recognition model and warning mechanism, and improve the accuracy and response speed of the system.

[0018] Furthermore, the automatic generation of inventory warning signals includes building a warning analysis model, which is calculated as follows:

[0019]

[0020] Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days;

[0021] Calculate the inventory standard deviation and build the model as follows:

[0022]

[0023] Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days;

[0024] Based on the above model, the upper and lower limits of warning are set, the upper limit of warning is μ+kσ, and the lower limit of warning is μ-k.

[0025] Furthermore, the parameters of the image recognition model are dynamically adjusted according to the recognition accuracy. If the recognition accuracy is lower than a preset threshold, the step of initiating the model retraining process includes continuously monitoring the recognition accuracy of the model in actual applications, regularly testing the performance of the model on the verification data set, and comparing the current recognition accuracy with the preset threshold. If the recognition accuracy is lower than the threshold, the model adjustment or retraining process is triggered. By analyzing the performance of the model and data changes, the specific reasons for the decline in recognition accuracy, such as overfitting and changes in data distribution, are determined. According to the analysis results, the parameters of the model, such as the learning rate, regularization term or network structure, are adjusted to improve the model performance. If the performance is not significantly improved after adjusting the parameters, the model retraining process is initiated, which may include adding training data and adjusting the training strategy. After the retraining is completed, the model is updated and deployed to actual applications, and its performance continues to be monitored to ensure stable operation of the system.

[0026] Furthermore, the step of finding the product that best matches the current image features through comparison and determining the type of the product based on the comparison result includes collecting inventory data from various channels, including sales data, warehouse inbound and outbound data, and procurement data;

[0027] Clean the collected data, remove invalid or erroneous data, and ensure data accuracy and consistency. At the same time, convert the data into a unified format for subsequent processing;

[0028] Based on business needs, choose to update inventory data in real time or in batches periodically. Update inventory quantities based on collected data. Based on the system architecture and business needs, select appropriate data synchronization technologies, including API interfaces, Web Services, message queues, and database replication. Define the data transmission paths and methods from the inventory management system to the sales platform, warehouse scheduling system, and financial management system.

[0029] Updated inventory data is pushed to sales platforms, warehouse scheduling systems, and financial management systems through standard APIs or WebService interfaces. Message queues are used for asynchronous data transmission, improving system responsiveness and stability. The inventory management system sends data change messages to message queues, and each target system subscribes to the relevant topics to receive data updates in real time. For scenarios with large data volumes, file transfer can be used to package inventory data into files and transfer them to the target system via FTP or SFTP.

[0030] According to one aspect of the present invention, there is provided an animation peripheral inventory management system based on image recognition, comprising:

[0031] Image acquisition module: used to collect physical images of inventory goods;

[0032] Image recognition module: used to identify the type and quantity of goods in the physical image;

[0033] Inventory management module: used to compare the recognition results with the pre-stored inventory data and update the inventory information in the inventory management system in real time;

[0034] Anomaly Detection Module: This module detects abnormal changes in the quantity of inventory items and automatically generates an inventory warning signal when the difference between the identified quantity and the pre-stored data exceeds a preset threshold.

[0035] Data synchronization module: used to synchronize updated inventory data to the sales platform, warehouse scheduling system and financial management system, realizing multi-terminal sharing and synchronous updating of inventory data.

[0036] Furthermore, the data synchronization module includes cleaning the collected inventory data, removing invalid or erroneous data, and converting it into a unified format; selecting real-time updates or periodic batch updates of inventory data based on business needs; performing consistency checks after data updates to ensure that inventory data in various systems is consistent; if there is any inconsistency, initiating a data repair process to ensure data accuracy and consistency;

[0037] After receiving the data update, the target system sends feedback to the inventory management system to confirm that the data has been successfully received and processed. Based on monitoring data and user feedback, the data synchronization process is optimized to improve synchronization efficiency and stability. For example, this includes optimizing API interface performance, adjusting message queue parameters, and optimizing data transmission protocols.

[0038] The anomaly detection model is trained using historical inventory data to learn the distribution of normal inventory quantities; new inventory quantities are predicted and their anomaly scores are calculated; and a threshold is determined through cross-validation to ensure that the anomaly score of normal data is lower than the threshold, while the anomaly score of anomaly data is higher than the threshold.

[0039] Store inventory data in a database, supporting fast query and update; perform statistical analysis on inventory data and generate inventory reports to support business decisions.

[0040] During data transmission, encryption technology is used to protect the security of inventory data; during data synchronization, inventory data is backed up regularly to ensure data security and recoverability, collect user feedback on the inventory management system, understand system usage and improvement needs; and based on user feedback and monitoring data, make system optimization suggestions.

[0041] According to one aspect of the present invention, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program and the processor implements the steps of the above-mentioned method for managing animation peripheral inventory based on image recognition when executing the computer program.

[0042] According to one aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for managing animation peripheral inventory based on image recognition are implemented.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention provides an animation peripheral inventory management method based on image recognition. The method comprises the following steps: inputting a collected image of an actual object into an image recognition model; the image recognition model extracts image features based on a convolutional neural network, compares the features with an image feature database, and identifies the type and quantity of goods; combining the recognition results with pre-stored inventory data, updating inventory information in real time; and automatically generating an inventory warning signal if an abnormal change in the quantity of inventory goods is identified; dynamically adjusting the parameters of the image recognition model according to the recognition accuracy; and starting a model retraining process if the recognition accuracy is lower than a preset threshold; and synchronizing the updated inventory data to a sales platform, a warehouse scheduling system, and a financial management system through an inventory management system, thereby realizing multi-terminal sharing and synchronous updating of inventory data. The method has the effect of realizing automated management of the entire process from warehousing, storage, to delivery of goods, and greatly improving the efficiency of inventory management. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0046] Figure 1 This is an overall schematic diagram of an animation peripheral inventory management method based on image recognition according to the present invention;

[0047] Figure 2 This is a schematic diagram of the framework of an animation peripheral inventory management system based on image recognition according to the present invention;

[0048] Figure 3 The figure is a schematic diagram of the computer structure in an animation peripheral inventory management system based on image recognition according to the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] like Figure 1-Figure 3 As shown, the present application provides an animation peripheral inventory management method based on image recognition, comprising:

[0051] S1: Input the collected physical image into the image recognition model, which extracts image features based on a convolutional neural network and compares them with the image feature database to identify the type and quantity of the product;

[0052] S2: Combine the recognition results with the pre-stored inventory data to update the inventory information in real time. If an abnormal change in the number of inventory items is identified, an inventory warning signal will be automatically generated;

[0053] S3: Dynamically adjust the parameters of the image recognition model according to the recognition accuracy. If the recognition accuracy is lower than the preset threshold, start the model retraining process;

[0054] S4: Synchronize the updated inventory data to the sales platform, warehouse scheduling system and financial management system through the inventory management system to achieve multi-terminal sharing and synchronous updating of inventory data.

[0055] In one embodiment, the system architecture and hardware configuration includes image acquisition devices, including high-resolution cameras and scanners installed at warehouse entrances, shelves, and sales areas to capture real-time product images. High-performance servers are configured for running image recognition models and storing large amounts of image data and feature databases. Network infrastructure includes stable local area network and internet connections to ensure data transmission and communication between devices.

[0056] The image recognition module, developed based on a convolutional neural network (CNN), is responsible for extracting features from product images and performing classification and identification. A feature database stores the image features of all known anime-related products for comparison and identification. The inventory management module integrates recognition results with existing inventory data, updates inventory information in real time, and triggers inventory alerts. The model training module dynamically adjusts model parameters based on recognition accuracy, automatically initiating retraining when accuracy falls below a threshold. The data synchronization module synchronizes updated inventory data to the sales platform, warehousing system, and financial system via an API.

[0057] Real-time capture: Cameras and scanners capture product images periodically or in real time to ensure clear images of the product are captured. Image upload: The captured images are uploaded to the server via the network and enter the image recognition process.

[0058] Feature extraction: The image recognition module uses CNN to extract features from product images, including key information such as color, shape, and pattern. Feature comparison compares the extracted features with data in the feature database to identify product type and quantity.

[0059] Result output: Outputs recognition results, including product type, quantity, and recognition accuracy. Inventory data update and integration: Integrates recognition results with pre-stored data in the inventory management system to update inventory information. Anomaly detection: The system monitors changes in inventory quantity and triggers an inventory warning signal if an anomaly is detected (such as an increase or decrease outside the preset range).

[0060] Accuracy monitoring continuously monitors recognition accuracy. If it falls below a preset threshold, the system automatically records the error and triggers model retraining. Data collection collects new product images and recognition results as a dataset for training a new model. Model retraining uses the new data to retrain the CNN model, optimize parameters, and improve recognition accuracy. Model update replaces the old model with the optimized one to enhance the system's recognition capabilities.

[0061] Data synchronization and sharing, including data push, allows the inventory management system to push updated inventory data to the sales platform, warehouse scheduling system, and financial management system via APIs. Real-time updates ensure consistency by updating local inventory information upon receiving data. Multi-terminal sharing allows managers to view the latest inventory data on various system terminals and make appropriate decisions and operations.

[0062] Cameras capture real-time images of merchandise on shelves, and a recognition system rapidly categorizes and counts the items. The system automatically updates inventory data. If the number of anime figures falls below safety stock, an alert is triggered, notifying the purchasing department to restock. Cameras in the sales area monitor merchandise displays in real time, and the recognition system compiles sales data. If sales of a particular product surge, the system automatically adjusts inventory allocation to ensure timely restocking of shelves.

[0063] Warehouse Scheduling Optimization: The warehouse system optimizes picking and delivery routes based on real-time inventory data, improving logistics efficiency. Inventory alerts help dispatchers prepare countermeasures in advance, reducing customer complaints caused by out-of-stock situations.

[0064] Financial management and cost control: The financial system monitors inventory dynamics in real time and accurately calculates costs and profits. By optimizing inventory management, we can reduce inventory overstock and losses, and improve capital turnover.

[0065] Efficient and accurate, image recognition technology significantly improves the efficiency and accuracy of inventory management and reduces human error. Real-time monitoring: The system updates inventory data in real time, enabling timely detection and resolution of inventory anomalies. Intelligent optimization dynamically adjusts model parameters to continuously improve recognition accuracy and adapt to changing product types and market demands. Environmental factors, such as product placement, lighting, and obstructions, can affect image recognition accuracy. Model maintenance costs: Regular updates and training require significant technical and human resources.

[0066] By implementing an image recognition-based inventory management approach for animation peripherals, companies can significantly improve the efficiency and accuracy of inventory management, reduce manual intervention, optimize resource allocation, and lower costs. Although there are some challenges in practical application, these challenges will gradually be resolved with continuous technological advancement and system optimization, bringing greater benefits and competitiveness to the animation peripherals industry.

[0067] Specifically, the step of inputting the captured physical image into an image recognition model, wherein the image recognition model extracts image features based on a convolutional neural network, includes using a high-resolution camera or other image acquisition device to capture the physical image of the product. Ensure sufficient lighting and an appropriate angle to obtain a clear image;

[0068] Adjust the collected images, including denoising, adjusting brightness and contrast, and cropping, to optimize image quality and improve recognition accuracy. The pre-processed images are input into an image recognition model based on a convolutional neural network. The convolutional neural network extracts image features layer by layer through multi-layer convolution and pooling operations, and compares the extracted image features with a pre-established image feature database.

[0069] By comparison, find the product that best matches the current image features. Based on the comparison results, determine the type of product. By analyzing the number of products in the image, determine the quantity of products. The identified product type and quantity are output for subsequent inventory management.

[0070] Specifically, the step of combining the recognition result with pre-stored inventory data to update inventory information in real time and automatically generating an inventory warning signal if an abnormal change in the quantity of inventory goods is identified includes obtaining pre-stored inventory data from an inventory management system, including the type, quantity, and storage location of goods, and comparing the image recognition result with the pre-stored inventory data;

[0071] Based on the comparison results, the inventory information in the inventory management system is updated to ensure the real-time and accuracy of inventory data. A threshold for inventory quantity changes is set. When the difference between the identified product quantity and the pre-stored data exceeds the threshold, an abnormal change alert is triggered. Abnormal changes may include sudden increases or decreases in inventory quantity that exceed the normal range.

[0072] When abnormal changes are detected, the system automatically generates an inventory warning signal. The warning signal can be sent in a variety of ways. The management personnel who receive the warning signal will take corresponding measures according to the specific situation, continuously collect images and update inventory data, maintain real-time synchronization of inventory information, regularly evaluate the performance of the inventory management system, optimize the recognition model and warning mechanism, and improve the accuracy and response speed of the system.

[0073] Specifically, the automatic generation of inventory warning signals includes building a warning analysis model, and the calculation is as follows:

[0074]

[0075] Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days;

[0076] Calculate the inventory standard deviation and build the model as follows:

[0077]

[0078] Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days;

[0079] Based on the above model, the upper and lower limits of warning are set, the upper limit of warning is μ+kσ, and the lower limit of warning is μ-k.

[0080] Specifically, the parameters of the image recognition model are dynamically adjusted according to the recognition accuracy. If the recognition accuracy is lower than a preset threshold, the steps of initiating the model retraining process include continuously monitoring the recognition accuracy of the model in actual applications, regularly testing the performance of the model on the verification data set, and comparing the current recognition accuracy with the preset threshold. If the recognition accuracy is lower than the threshold, the model adjustment or retraining process is triggered. By analyzing the performance of the model and data changes, the specific reasons for the decline in recognition accuracy, such as overfitting and changes in data distribution, are determined. According to the analysis results, the parameters of the model, such as the learning rate, regularization term or network structure, are adjusted to improve the model performance. If the performance is not significantly improved after adjusting the parameters, the model retraining process is initiated, which may include adding training data and adjusting the training strategy. After the retraining is completed, the model is updated and deployed to actual applications, and its performance continues to be monitored to ensure stable operation of the system.

[0081] Specifically, the step of finding the product that best matches the current image features through comparison and determining the type of the product based on the comparison result includes collecting inventory data from various channels, including sales data, warehouse inbound and outbound data, and procurement data;

[0082] Clean the collected data, remove invalid or erroneous data, and ensure data accuracy and consistency. At the same time, convert the data into a unified format for subsequent processing;

[0083] Based on business needs, choose to update inventory data in real time or in batches periodically. Update inventory quantities based on collected data. Based on the system architecture and business needs, select appropriate data synchronization technologies, including API interfaces, Web Services, message queues, and database replication. Define the data transmission paths and methods from the inventory management system to the sales platform, warehouse scheduling system, and financial management system.

[0084] Updated inventory data is pushed to sales platforms, warehouse scheduling systems, and financial management systems through standard APIs or WebService interfaces. Message queues are used for asynchronous data transmission, improving system responsiveness and stability. The inventory management system sends data change messages to message queues, and each target system subscribes to the relevant topics to receive data updates in real time. For scenarios with large data volumes, file transfer can be used to package inventory data into files and transfer them to the target system via FTP or SFTP.

[0085] According to one aspect of the present invention, there is provided an animation peripheral inventory management system based on image recognition, comprising:

[0086] Image acquisition module: used to collect physical images of inventory goods;

[0087] Image recognition module: used to identify the type and quantity of goods in the physical image;

[0088] Inventory management module: used to compare the recognition results with the pre-stored inventory data and update the inventory information in the inventory management system in real time;

[0089] Anomaly Detection Module: This module detects abnormal changes in the quantity of inventory items and automatically generates an inventory warning signal when the difference between the identified quantity and the pre-stored data exceeds a preset threshold.

[0090] Data synchronization module: used to synchronize updated inventory data to the sales platform, warehouse scheduling system and financial management system, realizing multi-terminal sharing and synchronous updating of inventory data.

[0091] Specifically, the data synchronization module includes cleaning the collected inventory data, removing invalid or erroneous data, and converting it into a unified format; based on business needs, it can choose to update inventory data in real time or in batches on a regular basis; after the data update is completed, it performs a consistency check to ensure that the inventory data in each system is consistent; if there is any inconsistency, it initiates the data repair process to ensure data accuracy and consistency;

[0092] After receiving the data update, the target system sends feedback to the inventory management system to confirm that the data has been successfully received and processed. Based on monitoring data and user feedback, the data synchronization process is optimized to improve synchronization efficiency and stability. For example, this includes optimizing API interface performance, adjusting message queue parameters, and optimizing data transmission protocols.

[0093] The anomaly detection model is trained using historical inventory data to learn the distribution of normal inventory quantities; new inventory quantities are predicted and their anomaly scores are calculated; and a threshold is determined through cross-validation to ensure that the anomaly score of normal data is lower than the threshold, while the anomaly score of anomaly data is higher than the threshold.

[0094] Store inventory data in a database, supporting fast query and update; perform statistical analysis on inventory data and generate inventory reports to support business decisions.

[0095] During data transmission, encryption technology is used to protect the security of inventory data; during data synchronization, inventory data is backed up regularly to ensure data security and recoverability, collect user feedback on the inventory management system, understand system usage and improvement needs; and based on user feedback and monitoring data, make system optimization suggestions.

[0096] In one embodiment,

[0097] The present invention provides an animation peripheral inventory management method based on image recognition. The method comprises the following steps: inputting a collected image of an actual object into an image recognition model; the image recognition model extracts image features based on a convolutional neural network, compares the features with an image feature database, and identifies the type and quantity of goods; combining the recognition results with pre-stored inventory data, updating inventory information in real time; and automatically generating an inventory warning signal if an abnormal change in the quantity of inventory goods is identified; dynamically adjusting the parameters of the image recognition model according to the recognition accuracy; and starting a model retraining process if the recognition accuracy is lower than a preset threshold; and synchronizing the updated inventory data to a sales platform, a warehouse scheduling system, and a financial management system through an inventory management system, thereby realizing multi-terminal sharing and synchronous updating of inventory data. The method has the effect of realizing automated management of the entire process from warehousing, storage, to delivery of goods, and greatly improving the efficiency of inventory management.

[0098] The present invention also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method for managing animation peripheral inventory based on image recognition when executing the computer program.

[0099] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned method for managing animation peripheral inventory based on image recognition are implemented.

[0100] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0101] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0102] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for managing animation peripheral inventory based on image recognition, characterized in that: include: The collected physical image is input into the image recognition model, which extracts image features based on a convolutional neural network and compares them with the image feature database to identify the type and quantity of the product; Combining the recognition results with pre-stored inventory data, the inventory information is updated in real time. If an abnormal change in the quantity of inventory items is identified, an inventory warning signal is automatically generated. Dynamically adjust the parameters of the image recognition model based on recognition accuracy. If the recognition accuracy falls below a preset threshold, the model retraining process is initiated. The updated inventory data is synchronized to the sales platform, warehouse scheduling system and financial management system through the inventory management system, realizing multi-terminal sharing and synchronous updating of inventory data.

2. The method for managing animation peripheral inventory based on image recognition according to claim 1, characterized in that: The step of inputting the captured physical image into an image recognition model, wherein the image recognition model extracts image features based on a convolutional neural network, includes using a high-resolution camera or other image acquisition device to capture the physical image of the product. Ensure sufficient lighting and an appropriate angle to obtain a clear image; Adjust the collected images, including denoising, adjusting brightness and contrast, and cropping, to optimize image quality and improve recognition accuracy. The pre-processed images are input into an image recognition model based on a convolutional neural network. The convolutional neural network extracts image features layer by layer through multi-layer convolution and pooling operations, and compares the extracted image features with a pre-established image feature database. By comparison, find the product that best matches the current image features. Based on the comparison results, determine the type of product. By analyzing the number of products in the image, determine the quantity of products. The identified product type and quantity are output for subsequent inventory management.

3. The method for managing animation peripheral inventory based on image recognition according to claim 1, characterized in that: The step of combining the recognition result with pre-stored inventory data to update inventory information in real time and automatically generating an inventory warning signal if an abnormal change in the number of inventory items is identified includes obtaining pre-stored inventory data from an inventory management system, including the type, quantity, and storage location of the items, and comparing the image recognition result with the pre-stored inventory data; Based on the comparison results, the inventory information in the inventory management system is updated to ensure the real-time and accuracy of inventory data. A threshold for inventory quantity changes is set. When the difference between the identified product quantity and the pre-stored data exceeds the threshold, an abnormal change alert is triggered. Abnormal changes may include sudden increases or decreases in inventory quantity that exceed the normal range. When abnormal changes are detected, the system automatically generates an inventory warning signal. The warning signal can be sent in a variety of ways. The management personnel who receive the warning signal will take corresponding measures according to the specific situation, continuously collect images and update inventory data, maintain real-time synchronization of inventory information, regularly evaluate the performance of the inventory management system, optimize the recognition model and warning mechanism, and improve the accuracy and response speed of the system.

4. The method for managing animation peripheral inventory based on image recognition according to claim 1, characterized in that: The automatic generation of inventory warning signals includes building a warning analysis model, which is calculated as follows: Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days; Calculate the inventory standard deviation and build the model as follows: Among them, μ is the average value, x i is the inventory quantity on day i, and N is the number of days; Based on the above model, the upper and lower limits of warning are set, the upper limit of warning is μ+kσ, and the lower limit of warning is μ-kσ.

5. The method for managing animation peripheral inventory based on image recognition according to claim 1, characterized in that: The parameters of the image recognition model are dynamically adjusted according to the recognition accuracy. If the recognition accuracy is lower than the preset threshold, the steps of initiating the model retraining process include continuously monitoring the recognition accuracy of the model in actual applications, regularly testing the performance of the model on the verification data set, and comparing the current recognition accuracy with the preset threshold. If the recognition accuracy is lower than the threshold, the model adjustment or retraining process is triggered. By analyzing the performance of the model and data changes, the specific reasons for the decline in recognition accuracy, such as overfitting and changes in data distribution, are determined. According to the analysis results, the parameters of the model, such as the learning rate, regularization term or network structure, are adjusted to improve the model performance. If the performance is not significantly improved after adjusting the parameters, the model retraining process is initiated, which may include adding training data and adjusting the training strategy. After the retraining is completed, the model is updated and deployed to actual applications, and its performance continues to be monitored to ensure stable operation of the system.

6. The method for managing animation peripheral inventory based on image recognition according to claim 5, characterized in that: The step of finding the product that best matches the current image features through comparison and determining the type of the product based on the comparison result includes collecting inventory data from various channels, including sales data, warehouse inbound and outbound data, and procurement data; Clean the collected data, remove invalid or erroneous data, and ensure data accuracy and consistency. At the same time, convert the data into a unified format for subsequent processing; Based on business needs, choose to update inventory data in real time or in batches periodically. Update inventory quantities based on collected data. Based on the system architecture and business needs, select appropriate data synchronization technologies, including API interfaces, Web Services, message queues, and database replication. Define the data transmission paths and methods from the inventory management system to the sales platform, warehouse scheduling system, and financial management system. Updated inventory data is pushed to sales platforms, warehouse scheduling systems, and financial management systems through standard APIs or WebService interfaces. Message queues are used for asynchronous data transmission, improving system responsiveness and stability. The inventory management system sends data change messages to message queues, and each target system subscribes to the relevant topics to receive data updates in real time. For scenarios with large data volumes, file transfer can be used to package inventory data into files and transfer them to the target system via FTP or SFTP.

7. An animation peripheral inventory management system based on image recognition, characterized in that: include: Image acquisition module: used to collect physical images of inventory goods; Image recognition module: used to identify the type and quantity of goods in the physical image; Inventory management module: used to compare the recognition results with the pre-stored inventory data and update the inventory information in the inventory management system in real time; Anomaly Detection Module: This module detects abnormal changes in the quantity of inventory items and automatically generates an inventory warning signal when the difference between the identified quantity and the pre-stored data exceeds a preset threshold. Data synchronization module: used to synchronize updated inventory data to the sales platform, warehouse scheduling system and financial management system, realizing multi-terminal sharing and synchronous updating of inventory data.

8. The animation peripheral inventory management system based on image recognition according to claim 7 is characterized in that: The data synchronization module cleans the collected inventory data, removes invalid or erroneous data, and converts it into a unified format. Based on business needs, inventory data can be updated in real time or in batches at regular intervals. After the data update is complete, a consistency check is performed to ensure that the inventory data in each system is consistent. If there is any inconsistency, start the data repair process to ensure the accuracy and consistency of the data; After receiving the data update, the target system sends feedback to the inventory management system to confirm that the data has been successfully received and processed; Based on monitoring data and user feedback, optimize the data synchronization process to improve synchronization efficiency and stability. For example, optimize API interface performance, adjust message queue parameters, and optimize data transmission protocols. Use historical inventory data to train anomaly detection models to learn the distribution of normal inventory quantities; predict new inventory quantities and calculate their anomaly scores; The threshold is determined through cross-validation to ensure that the anomaly score of normal data is lower than the threshold, while the anomaly score of anomaly data is higher than the threshold. Store inventory data in a database, supporting fast query and update; Conduct statistical analysis on inventory data and generate inventory reports to support business decisions. During data transmission, encryption technology is used to protect the security of inventory data; During data synchronization, inventory data is backed up regularly to ensure data security and recoverability, and user feedback on the inventory management system is collected to understand system usage and improvement needs; Based on user feedback and monitoring data, make system optimization suggestions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the processor implements the steps of the method for managing animation peripheral inventory based on image recognition according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for managing animation peripheral inventory based on image recognition according to any one of claims 1 to 6 are implemented.

Images