Full-automatic mobile phone film pasting inventory management system based on big data

Abstract

The invention discloses a full-automatic mobile phone film pasting inventory management system based on big data, which comprises an image acquisition module, a database module, a detection and identification module, an inventory management module and a terminal module, and is characterized in that the detection and identification module comprises a model construction training module and a detection module. The goods shelf management module comprises a multi-goods shelf scheduling module and an inventory detection module, mutual inventory scheduling of a plurality of goods shelves can be achieved through distributed storage, overall inventory distribution is optimized, mobile phone films to be stored are concluded and stored, and the mobile phone models are obtained. And meanwhile, dynamic functions of inventory early warning, inventory distribution analysis and warehousing and ex-warehousing are provided. The method has the advantages that different types of mobile phone films and multi-shelf scheduling monitoring are accurately identified through the feature learning technology, the inventory distribution and storage efficiency is optimized, demand changes are flexibly coped with, and resource waste is reduced.

Description

Technical Field

[0001] The present invention relates to intelligent identification and management, and particularly to a fully automatic mobile phone film inventory management system based on big data. Background Art

[0002] A fully automatic mobile phone film inventory management system based on big data automatically identifies different models of mobile phones through image recognition technology, accurately matches corresponding mobile phone films according to the recognition results, obtains the optimal path for storing and retrieving mobile phone films through a path planning algorithm, so as to realize intelligent inventory management. It monitors the inventory situation of mobile phone films in real time and automatically records the inbound and outbound information of each film. Based on big data analysis, the system can accurately predict and dynamically adjust the inventory. The system also has the ability of distributed storage and multi-shelf scheduling, and will automatically schedule the resources of other shelves for replenishment to ensure the balance and timely supply of inventory. All data can be updated in real time and shared among different shelves, improving the overall operation efficiency and response speed. This intelligent and automatic management method greatly improves the operation efficiency, reduces the labor cost, optimizes the resource allocation, and finally enhances the collaboration and response ability of the entire supply chain.

[0003] The current mobile phone film inventory management systems on the market are mainly targeted at mobile phone accessory merchants, repair shops and film specialty stores, helping merchants efficiently manage inventory, purchase, sales and inventory, order and sales data. Most systems provide basic functions such as inventory tracking, purchase management, sales out, etc., and can predict demand through intelligent algorithms to ensure sufficient inventory and avoid out-of-stock or overstock phenomena. Some systems also support technologies such as barcode scanning and two-dimensional code management to improve the efficiency of inventory checking and product tracking. Some systems also integrate customer management functions, which can record customer purchase history and preferences and provide personalized promotion and service suggestions. Summary of the Invention

[0004] In order to improve the existing mobile phone film inventory management system, a fully automatic mobile phone film inventory management system based on big data is provided. This method accurately identifies the mobile phone model through intelligent recognition technology and automatically matches the corresponding mobile phone film to realize real-time update and accurate management of inventory data. The system combines distributed storage and multi-shelf scheduling functions to ensure the balance and timely allocation of inventory among shelves and improve the inventory management efficiency.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A fully automatic mobile phone film inventory management system based on big data, comprising:

[0007] Image acquisition module: The image acquisition module includes an automated fixture device and an image preprocessing device. The automated fixture device performs access operations on the mobile phone film based on the optimal path, and the image preprocessing device is used to process the image to make the mobile phone have integrity and continuity;

[0008] Database module: The database module is mainly used to store the identified mobile phone model information and the inventory of mobile phone films of each model;

[0009] Detection and recognition module: The detection and recognition module includes a model construction and training module and a detection module. The model construction and training module is mainly used to obtain the mobile phone model features through feature learning to construct a recognition model, and train the model with the data in the database. The detection module is used to identify and detect the mobile phone image obtained by the image acquisition module through the mobile phone model detection model to obtain the mobile phone model;

[0010] Inventory management module: The inventory management module is used to store mobile phone films of each model in a distributed manner, including a multi-shelf scheduling module and an inventory detection module. The multi-shelf scheduling module is mainly used to enable multiple shelves to schedule inventory with each other, optimize the overall inventory distribution, and classify and store the mobile phone films to be stored. The inventory detection module is mainly used to obtain the real-time storage situation in the shelf, including the storage quantity, available storage quantity, and quantity to be stored;

[0011] Terminal module: The terminal module includes a mobile phone application, a tablet computer, and a computer application, which can browse the inventory of mobile phone films of each model in the database and has the function of issuing an inventory shortage warning.

[0012] Preferably, the image acquisition module specifically includes:

[0013] Automated fixture device, which is used to clamp the mobile phone in front of the camera for acquisition and obtain the recognition result to be processed;

[0014] Image preprocessing device, which eliminates image noise through Gaussian filtering technology, uses the Canny algorithm for edge detection to highlight the outline of the mobile phone frame, and simplifies the image through histogram equalization and binarization operations to present the mobile phone image features.

[0015] Preferably, the automated fixture device specifically includes:

[0016] Obtain the storage positions P1(x1,y1), P2(x2,y2),..., P n (x n ,y n ) of the mobile phone film to be grabbed;

[0017] Create an n×n distance matrix based on the storage positions of the mobile phone film;

[0018] Obtain the shortest path of the automated fixture device's movement based on the dynamic programming algorithm.

[0019] Preferably, the obtaining of the shortest path of the automated fixture device's movement based on the dynamic programming algorithm specifically includes:

[0020] Define the state C(S, i) to represent the shortest path length from the starting point, after visiting all the mobile phone film storage points to be grasped, and finally reaching point i. The state transition equation is:

[0021] C(S, i) = min j∈S,j≠i (C(S - {i}, j) + d ji )

[0022] where d ji is the distance from point j to point i;

[0023] Obtain the shortest path from the starting point O, passing through all the target points and finally returning to point O. The formula is:

[0024] L min = min i∈{1,2,...,n} (C({1, 2,..., n}, i) + d io )

[0025] Preferably, the detection and recognition module specifically includes:

[0026] Model construction and training module: The model construction and training module includes a model construction unit, a convolutional layer feature extraction unit, an activation function unit, a pooling operation unit, a fully connected layer unit, a loss function unit, and a model training unit;

[0027] Detection module: The detection module includes a feature extraction unit, a model prediction unit, and an identification result acquisition unit.

[0028] Preferably, the model construction and training module specifically includes:

[0029] Construct a mobile phone model recognition model based on ResNet;

[0030] Filter the local area of the mobile phone image through a convolutional kernel to extract local features of the image, such as edges, textures, etc. Each convolutional kernel is a learned parameter. The formula is:

[0031] F(i, j) = (I * K)(i, j) + b

[0032] where I is the input image, K is the convolutional kernel, b is the bias term, * is the convolutional operation, F is the output feature map, and i, j are position indices;

[0033] Processing through the ReLU activation function can turn negative values into 0, increasing the non-linear expression ability of the image network;

[0034] Through pooling operations for downsampling, the size of the feature map is reduced, the computational amount is reduced, and important spatial information of mobile phone features is maintained;

[0035] After extracting features in the convolutional layer and pooling layer, the high-dimensional feature map is flattened through the fully connected layer to obtain a higher-level non-linear relationship;

[0036] Through the Softmax layer, the output of the fully connected layer is converted into a probability distribution, representing the possibility of each mobile phone model recognition result. The formula is:

[0037]

[0038] Among them, is the grid output, and P(y i ) is the probability of class i;

[0039] During the training process, the loss function is minimized through cross-entropy loss, thereby minimizing the classification error and updating the model parameters.

[0040] Preferably, the detection module specifically includes:

[0041] Input the mobile phone image obtained by the image acquisition module into the mobile phone model recognition model;

[0042] Obtain the classification probability distribution result given by the model, select the class with the highest probability, and obtain the mobile phone type.

[0043] Preferably, the inventory management module specifically includes:

[0044] Multi-shelf scheduling module: The multi-shelf scheduling module manages the inventory scheduling between multiple shelves, optimizes the overall inventory distribution, ensures that the inventory levels and resource utilization efficiencies of each shelf reach the optimal, and intelligently inductively stores the mobile phone films to be stored;

[0045] Inventory detection module: The inventory detection module is mainly used to obtain the storage situation in the shelf in real time and monitor the inventory status of the shelf in real time, including the current storage quantity, the available storage quantity, and the quantity to be stored.

[0046] Preferably, the multi-shelf scheduling module specifically includes:

[0047] Adopt a distributed storage architecture, so that each shelf acts as a distributed node to manage the inventory. Each shelf node stores its own inventory information and communicates with other shelf nodes to achieve real-time data synchronization;

[0048] For each storage shelf, obtain its current storage quantity C i and the maximum storage quantity M i , and calculate and obtain the remaining available storage space R i ;

[0049] Based on the remaining space of each shelf, initially allocate the quantity of mobile phone films to be stored. The formula is:

[0050]

[0051] Detect whether each shelf meets the maximum storage constraint. If the result exceeds the remaining available space of the shelf, adjust its quantity to be stored;

[0052] Randomly generate multiple initial solutions based on the genetic algorithm. Each solution represents the allocation quantity T to be stored on a shelf i ;

[0053] Calculate the fitness F of each allocation quantity T to be stored through the objective function i . The formula is:

[0054]

[0055] Among them, α and β are weight coefficients, C i , C j are the storage quantities of the shelves, and D i is the storage cost;

[0056] Obtain the optimal solution for shelf scheduling through crossover operation and mutation operation;

[0057] Establish an optimization objective function and optimize the overall inventory distribution by adjusting the quantity T to be stored i . If there are n shelves, the cost minimization formula is:

[0058]

[0059] Preferably, the inventory detection module specifically includes:

[0060] Obtain the current storage quantity and the quantity to be stored of each shelf in real time;

[0061] Process and analyze the collected data to form a comprehensive and real-time inventory status report, providing functions such as inventory warning, inventory distribution analysis, and dynamic inbound and outbound.

[0062] Compared with the prior art, the advantages of the present invention are:

[0063] Through the model construction and training module, the system uses feature learning technology to construct an accurate mobile phone model recognition model, which can quickly and accurately identify mobile phone films of different models. Based on the path planning algorithm, the system obtains the best path for storing and retrieving mobile phone films, greatly improving the automation level and reducing human intervention. The continuous training and optimization of the model enable the system to adapt to different model changes, enhancing the adaptability and accuracy of recognition. Meanwhile, the inventory management module realizes the inventory allocation between multiple shelves through the multi-shelf scheduling module, optimizing the inventory distribution. The coordinated scheduling between shelves not only reduces inventory backlogs but also can quickly adjust resource allocation according to demand, reducing operating costs. In addition, the inventory detection module can monitor the storage situation of the shelves in real time, timely obtain the storage quantity, available storage quantity, and pending storage quantity of each shelf, helping managers quickly understand the inventory status and make adjustments according to actual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 FIG. is a schematic diagram of the overall system flow of the system proposed by the present invention;

[0065] Figure 2 FIG. is a schematic diagram of the automated fixture operation of the system proposed by the present invention;

[0066] Figure 3 FIG. is a schematic diagram of the path dynamic planning of the system proposed by the present invention;

[0067] Figure 4 FIG. is a schematic diagram of the model construction and training of the system proposed by the present invention;

[0068] Figure 5 FIG. is a schematic diagram of the mobile phone model detection of the system proposed by the present invention;

[0069] Figure 6 FIG. is a schematic diagram of the multi-shelf scheduling of the system proposed by the present invention;

[0070] Figure 7 FIG. is a schematic diagram of the inventory detection of the system proposed by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0071] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0072] Refer to Figure 1 as shown, a fully automatic mobile phone film inventory management system based on big data includes:

[0073] Image acquisition module: The image acquisition module includes an automated fixture device and an image preprocessing device. The automated fixture device performs access operations on the mobile phone film based on the optimal path, and the image preprocessing device is used to process the image to make the mobile phone have integrity and continuity;

[0074] Database module: The database module is mainly used to store the recognized mobile phone model information and the inventory of mobile phone films of each model;

[0075] Detection and recognition module: The detection and recognition module includes a model construction and training module and a detection module. The model construction and training module is mainly used to obtain the mobile phone model features through feature learning to construct a recognition model, and train the model with the data in the database. The detection module is used to perform recognition and detection on the mobile phone images obtained by the image acquisition module through the mobile phone model detection model to obtain the mobile phone model;

[0076] Inventory management module: The inventory management module is used to store mobile phone films of each model in a distributed manner, including a multi-shelf scheduling module and an inventory detection module. The multi-shelf scheduling module is mainly used to enable multiple shelves to schedule inventory with each other, optimize the overall inventory distribution, and classify and store the mobile phone films to be stored. The inventory detection module is mainly used to obtain the real-time storage situation in the shelves, including the storage quantity, the available storage quantity, and the quantity to be stored;

[0077] Terminal module: The terminal module includes a mobile phone application, a tablet computer, and a computer application, which can browse the inventory of mobile phone films of each model in the database and has the function of issuing inventory shortage warnings.

[0078] Refer to Figure 1 As shown in the figure, the image acquisition module specifically includes:

[0079] Automated fixture device, which is used to clamp the mobile phone in front of the camera for acquisition and obtain the recognition result to be processed;

[0080] Image preprocessing device, which eliminates image noise through Gaussian filtering technology, uses the Canny algorithm for edge detection to highlight the outline of the mobile phone frame, and simplifies the image through histogram equalization and binarization operations to present the mobile phone image features.

[0081] Refer to Figure 2 As shown in the figure, the automated fixture device specifically includes:

[0082] Obtain the storage positions P1(x1,y1), P2(x2,y2),..., P n (x n ,y n ) of the mobile phone film to be grabbed;

[0083] Create an n×n distance matrix based on the storage locations of the mobile phone films;

[0084] Obtain the shortest path of the automated fixture device's movement based on the dynamic programming algorithm.

[0085] Refer to Figure 3 As shown, obtaining the shortest path of the automated fixture device's movement based on the dynamic programming algorithm specifically includes:

[0086] Define the state C(S,i) to represent the shortest path length from the starting point, after visiting all the mobile phone film storage points that need to be grasped, and finally reaching point i. The state transition equation is:

[0087] C(S,i) = min j∈S,j≠i (C(S - {i},j) + d ji )

[0088] where d ji is the distance from point j to point i;

[0089] Obtain the shortest path from the starting point O, passing through all the target points and finally returning to point O. The formula is:

[0090] L min = min i∈{1,2,...,n} (C({1,2,...,n},i) + d io )

[0091] Refer to Figure 1 As shown, the detection and recognition module specifically includes:

[0092] Model construction and training module: The model construction and training module includes a model construction unit, a convolutional layer feature extraction unit, an activation function unit, a pooling operation unit, a fully connected layer unit, a loss function unit, and a model training unit;

[0093] Detection module: The detection module includes a feature extraction unit, a model prediction unit, and an identification result acquisition unit.

[0094] Refer to Figure 4 As shown, the model construction and training module specifically includes:

[0095] Build a mobile phone model recognition model based on ResNet;

[0096] Filter the local area of the mobile phone image through a convolutional kernel to extract the local features of the image, such as edges, textures, etc. Each convolutional kernel is a learned parameter. The formula is:

[0097] F(i,j) = (I * K)(i,j) + b

[0098] Among them, I is the input image, K is the convolution kernel, b is the bias term, * is the convolution operation, F is the output feature map, and i, j are position indices;

[0099] By processing through the ReLU activation function, negative values can be turned into 0, increasing the non-linear expression ability of the image network;

[0100] Through the pooling operation for downsampling, the size of the feature map is reduced, the computational amount is reduced, and important spatial information of mobile phone features is retained;

[0101] After extracting features in the convolutional layer and pooling layer, the high-dimensional feature map is flattened through the fully connected layer to obtain a higher-level non-linear relationship;

[0102] Through the Softmax layer, the output of the fully connected layer is converted into a probability distribution, representing the possibility of the recognition result of each mobile phone model. The formula is:

[0103]

[0104] Among them, is the grid output, and P(y i ) is the probability of class i;

[0105] During the training process, the loss function is minimized through cross-entropy loss, thereby minimizing the classification error and updating the model parameters.

[0106] Specifically, after completing the recognition and analysis of mobile phone images, a process of repeated training is carried out. By integrating the analyzed mobile phone image data into the existing training set, the recognition and prediction capabilities of the model are enhanced. First, the mobile phone images are added to the training data set for the mobile phone model recognition model to learn. Subsequently, using the updated data set, a new round of training process is started. The model adjusts its internal parameters through backpropagation and gradient descent algorithms to better adapt to the new data features. After training is completed, verification and testing are carried out to evaluate the improvement of the model performance. If the performance meets the expectations, the updated model will be deployed for actual mobile phone image analysis tasks, completing the closed-loop process of the entire repeated training.

[0107] Refer to Figure 5 As shown, the detection module specifically includes:

[0108] Input the mobile phone images obtained by the image acquisition module into the mobile phone model recognition model;

[0109] Obtain the classification probability distribution result given by the model, select the category with the highest probability, and obtain the mobile phone type.

[0110] Refer to Figure 1 As shown, the inventory management module specifically includes:

[0111] Multi-shelf Scheduling Module: The multi-shelf scheduling module manages the inventory scheduling among multiple shelves, optimizes the overall inventory distribution, ensures that the inventory levels and resource utilization efficiencies of each shelf reach the optimal state, and intelligently classifies and stores the mobile phone films to be stored.

[0112] Inventory Detection Module: The inventory detection module is mainly used to obtain the storage situation in the shelf in real time and monitor the inventory status of the shelf in real time, including the current storage quantity, the available storage quantity, and the quantity to be stored.

[0113] See Figure 6 As shown, the multi-shelf scheduling module specifically includes:

[0114] Adopt a distributed storage architecture, so that each shelf acts as a distributed node to manage the inventory. Each shelf node stores its own inventory information and communicates with other shelf nodes to achieve real-time data synchronization.

[0115] For each storage shelf, obtain its current storage quantity C i and the maximum storage quantity M i , and calculate and obtain the remaining available storage space R i ;

[0116] According to the remaining space of each shelf, initially allocate the quantity of mobile phone films to be stored. The formula is:

[0117]

[0118] Detect whether each shelf meets the maximum storage constraint. If the result exceeds the remaining available space of the shelf, adjust its quantity to be stored.

[0119] Randomly generate multiple initial solutions based on the genetic algorithm. Each solution represents the quantity of mobile phone films to be stored T i ;

[0120] Calculate the fitness F of each quantity of mobile phone films to be stored T i through the objective function. The formula is:

[0121]

[0122] Among them, α and β are weight coefficients, C i , C j are the storage quantities of the shelves, and D i is the storage cost;

[0123] Obtain the optimal solution of shelf scheduling through crossover operation and mutation operation;

[0124] Establish an optimization objective function, and optimize the overall inventory distribution by adjusting the quantity of mobile phone films to be stored T i . If there are n shelves, the cost minimization formula is:

[0125]

[0126] Refer to Figure 7 As shown, the inventory detection module specifically includes:

[0127] Obtain the current storage quantity and the quantity to be stored of each shelf in real time;

[0128] Process and analyze the collected data to form a comprehensive and real-time inventory status report, providing functions such as inventory warning, inventory distribution analysis, and dynamic functions of inbound and outbound.

[0129] It should be noted that: the above order of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. And the above specific embodiments of this specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0130] Each embodiment in this specification is described in a progressive manner. The same or similar parts between each embodiment can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

[0131] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A fully automated mobile phone film inventory management system based on big data, characterized in that, Including: Image acquisition module: The image acquisition module includes an automated fixture device and an image preprocessing device. The automated fixture device performs access operations on the mobile phone film based on the optimal path, and the image preprocessing device is used to process the image to make the mobile phone have integrity and continuity; Database module: The database module is mainly used to store the identified mobile phone model information and the inventory of mobile phone films of each model; Detection and recognition module: The detection and recognition module includes a model construction and training module and a detection module. The model construction and training module is mainly used to obtain the mobile phone model features through feature learning to construct an identification model, and train the model with the data in the database. The detection module is used to identify and detect the mobile phone image obtained by the image acquisition module through the mobile phone model detection model to obtain the mobile phone model; Inventory management module: The inventory management module is used to store mobile phone films of each model in a distributed manner, including a multi-shelf scheduling module and an inventory detection module. The multi-shelf scheduling module is mainly used to enable multiple shelves to schedule inventory with each other, optimize the overall inventory distribution, and classify and store the mobile phone films to be stored. The inventory detection module is mainly used to obtain the real-time storage situation in the shelf, including the storage quantity, the available storage quantity, and the quantity to be stored; Terminal module: The terminal module includes a mobile phone application, a tablet computer, and a computer application, which can browse the inventory of mobile phone films of each model in the database and has the function of issuing an inventory shortage warning.

2. The fully automatic mobile phone film inventory management system based on big data according to claim 1, wherein, Specifically, the image acquisition module includes: Automated fixture device, which is used to clamp the mobile phone in front of the camera for acquisition and obtain the recognition result to be processed; Image preprocessing device, which eliminates image noise through Gaussian filtering technology, uses the Canny algorithm for edge detection to highlight the contour of the mobile phone frame, and simplifies the image through histogram equalization and binarization operations to present the mobile phone image features.

3. The fully automatic mobile phone film inventory management system based on big data according to claim 2, characterized in that, Specifically, the automated fixture device includes: Obtain the storage positions P1(x1, y1), P2(x2, y2),..., P of the mobile phone films to be grabbed n (x n , y n ); Create an n×n distance matrix based on the storage location of the mobile phone film; Obtain the shortest path of the movement of the automated fixture device based on the dynamic programming algorithm.

4. The fully automatic mobile phone film inventory management system based on big data according to claim 3, characterized in that, Specifically, the obtaining of the shortest path of the movement of the automated fixture device based on the dynamic programming algorithm includes: Define the state C(S, i) to represent the shortest path length from the starting point, visiting all the storage points of the mobile phone films to be grabbed, and finally reaching point i. The state transition equation is: C(S, i) = min j∈S，j≠i (C(S - {i}, j) + d ji ) where d ji is the distance from point j to point i; Obtain the shortest path from the starting point O passing through all the target points and finally returning to point O. The formula is: L min = min i∈{1，2，...，n} (C({1, 2, …, n}, i) + d io ) 5. The fully automatic mobile phone film inventory management system based on big data according to claim 1, characterized in that, Specifically, the detection and recognition module includes: Model construction and training module: The model construction and training module includes a model construction unit, a convolutional layer feature extraction unit, an activation function unit, a pooling operation unit, a fully connected layer unit, a loss function unit, and a model training unit; Detection module: The detection module includes a feature extraction unit, a model prediction unit, and an identification result acquisition unit.

6. The fully automatic mobile phone film inventory management system based on big data according to claim 3, characterized in that, Specifically, the model construction and training module includes: Construct a mobile phone model recognition model based on ResNet; Filter the local area of the mobile phone image through a convolutional kernel, extract the local features of the image, such as edges, textures, etc. Each convolutional kernel is a learned parameter. The formula is: F(i, j) = (I * K)(i, j) + b Where I is the input image, K is the convolutional kernel, b is the bias term, * is the convolution operation, F is the output feature map, and i, j are position indices; Processing through the ReLU activation function can turn negative values into 0, increasing the non-linear expression ability of the image network; Through pooling operations for downsampling, the size of the feature map is reduced, the amount of computation is reduced, and important spatial information of mobile phone features is retained; After extracting features in the convolutional layer and pooling layer, the high-dimensional feature map is flattened through a fully-connected layer to obtain a higher-level non-linear relationship; The output of the fully-connected layer is converted into a probability distribution through the Softmax layer, representing the possibility of each mobile phone model recognition result. The formula is: Among them, is the grid output, and P(yi) is the probability of class i; During the training process, the loss function is minimized through cross-entropy loss, thereby minimizing the classification error and updating the model parameters.

7. An automatic mobile phone film inventory management system based on big data according to claim 3, characterized in that, The detection module specifically includes: Input the mobile phone image obtained by the image acquisition module into the mobile phone model recognition model; Obtain the classification probability distribution result given by the model, select the category with the highest probability, and obtain the mobile phone type.

8. A fully automatic mobile phone film inventory management system based on big data according to claim 1, characterized in that, The inventory management module specifically includes: Multi-shelf scheduling module: The multi-shelf scheduling module manages the inventory scheduling between multiple shelves, optimizes the overall inventory distribution, ensures that the inventory levels and resource utilization efficiency of each shelf reach the optimal, and intelligently inductively stores the mobile phone films to be stored; Inventory detection module: The inventory detection module is mainly used to obtain the storage situation in the shelf in real time and monitor the inventory status of the shelf in real time, including the current storage quantity, the available storage quantity, and the quantity to be stored.

9. The fully automatic mobile phone film inventory management system based on big data according to claim 7, characterized in that The multi-shelf scheduling module specifically includes: Adopt a distributed storage architecture, so that each shelf acts as a distributed node to manage the inventory. Each shelf node stores its own inventory information and communicates with other shelf nodes to achieve real-time data synchronization; For each storage shelf, obtain its current storage capacity C i , the maximum storage capacity M i , and calculate to obtain the remaining available storage space R i ; According to the remaining space of each shelf, initially allocate the quantity of mobile phone films to be stored. The formula is: Detect whether each shelf meets the maximum storage constraint. If the result exceeds the remaining available space of the shelf, adjust its quantity to be stored; Randomly generate multiple initial solutions based on the genetic algorithm, and each solution represents the storage allocation quantity T of a shelf i ; Calculate the fitness F of each storage allocation quantity T to be stored through the objective function. The formula is as follows: i ​ where α and β are weight coefficients, C i and C j are the shelf storage capacities, and D i is the storage cost; Through crossover operations and mutation operations, obtain the optimal solution for shelf scheduling; Establish an optimization objective function and optimize the overall inventory distribution by adjusting the storage quantity T to be stored i If there are n shelves, the cost formula to be minimized is as follows:

10. The fully automatic mobile phone film inventory management system based on big data according to claim 7, characterized in that, The inventory detection module specifically includes: Obtain the current storage quantity and the quantity to be stored of each shelf in real time; Process and analyze the collected data to form a comprehensive and real-time inventory status report, providing functions such as inventory warning, inventory distribution analysis, and dynamic inbound and outbound.

Images