Intelligent fresh food sorting method, system and storage medium
Through the intelligent fresh food sorting system, machine learning and computer vision technology are used to solve the problems of low efficiency, poor accuracy and insufficient traceability in fresh food sorting management, and efficient and accurate sorting and distribution are achieved. Improve operational efficiency and customer satisfaction.
Patent Information
- Application Number
- CN202310904540.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-21
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2043-07-21
AI Technical Summary
The existing fresh food sorting management methods have problems such as low order processing efficiency, low sorting accuracy, cumbersome manual operations, poor traceability, time-consuming and error-prone, especially under the requirements of large-scale and high-efficiency, the results are limited.
Adopt intelligent fresh food sorting system, including order management system, MRP intelligent computing system and intelligent sorting system, through machine learning prediction models and computer vision technology, long-term early warning and real-time analysis, automated sorting and distribution, and reduce manual intervention , improve accuracy and efficiency.
It significantly reduces the error rate of fresh food sorting, reduces fresh food loss, improves operational efficiency and customer satisfaction, realizes accurate sorting and packaging, and improves the traceability of the system.
Smart Images

Figure CN117455368B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of fresh produce, and in particular to an intelligent fresh produce sorting and delivery method, system and storage medium. Background Art
[0002] Traditional fresh produce sorting management methods mainly rely on manual operation and management combined with electronic equipment. First, manual labor is used to sort fresh products. Workers manually take products out of warehouses or cold storage according to order requirements and product characteristics, and classify, sort and pack them as required. Subsequently, each product is weighed manually using a weighing device to ensure that the exact weight of the product meets the order requirements. This is common in scenarios where sorting by weight is required, such as fresh products such as meat and fish. The quality and compliance of the product are ensured by visually inspecting the appearance, color, freshness and other characteristics of the product. This is a common quality control method, but there may be problems with subjectivity and human differences. After sorting, workers manually pack the products according to order requirements, such as boxing and labeling. This requires certain packaging skills and experience to ensure the safety and integrity of the product. Finally, a paper or manual recording system is used to record and manage relevant data and information during the sorting process. This includes order information, sorting quantity, product quality, etc. for subsequent tracking and management. Traditional fresh produce sorting management methods still have certain applications.
[0003] In addition, a certain degree of automation and information technology has also been applied to the field of sorting management to improve efficiency, accuracy and traceability.
[0004] The existing fresh food management system still has some limitations in the application of fresh food, such as low order processing efficiency, low sorting accuracy, cumbersome manual operation, poor traceability, time-consuming, error-prone, and limited effect under large-scale and high-efficiency requirements. Therefore, a comprehensive management solution is needed to solve these problems and improve the operational efficiency and customer satisfaction of fresh food enterprises. Summary of the invention
[0005] In view of this, the present application provides an intelligent fresh food sorting method, system and storage medium to solve the pain points and problems in the field of fresh food sorting, such as high error rate, excessive fresh food loss, and time-consuming and labor-intensive manual work.
[0006] In a first aspect, an intelligent fresh produce sorting system includes the following components:
[0007] An order management system, used to generate demand information instructions and send the demand information instructions to the MRP intelligent computing system;
[0008] The MRP intelligent operation system is used to obtain the demand information instructions of the order management system, and / or perform intelligent analysis based on the stored historical data and inventory information to achieve at least one of the following operations:
[0009] 1) Long-term warning: According to the immediate demand information instructions or periodic demand information instructions of the order management system, based on the historical data of a certain period, the future sales volume of product A is predicted; and according to the inventory information, decision analysis is performed. When the inventory quantity and / or weight of product A is lower than the sales volume threshold for the next n days, an alarm is issued for the inventory of product A, and a purchase order plan is issued, where n is an integer greater than zero;
[0010] 2) Real-time analysis: Receive the sales demand instruction for product A from the order system, call the inventory information, and conduct real-time sales demand analysis. When the inventory quantity and / or weight of product A is greater than the sales demand quantity and / or weight of product A, push the sorting instruction to the intelligent sorting system; when the inventory quantity and / or weight of product A is less than the sales demand quantity and / or weight of product A, feedback an alarm prompt to the order management system, and the order management system sends an instruction to the client, requesting the client to terminate, suspend or delay delivery;
[0011] The intelligent sorting system sorts product A from the storage area according to the demand information instruction of the order management system and / or the analysis result of the MRP intelligent operation system, and obtains the required weight and / or quantity of product A according to the demand information;
[0012] Delivery service system, including but not limited to:
[0013] 1) An electronic code sheet printing system is used to generate a corresponding electronic code sheet for product A obtained in the sorting task, including information such as product name, quantity, weight, etc., and issue the product delivery information;
[0014] 2) An intelligent distribution unit, which is used to receive the electronic code sheet and perform corresponding distribution operations.
[0015] Preferably, the long-term warning of the MRP intelligent computing system adopts a machine learning prediction model for prediction, which establishes a prediction model through intelligent analysis and learning of historical sales data, weather data, seasonal data, including but not limited to historical sales data, weather data, and seasonal data, predicts future demand trends, and makes threshold judgments based on these prediction results, and issues purchase order plans, thereby more accurately meeting the order needs of the next n days.
[0016] Preferably, the intelligent sorting system uses computer vision technology and / or object recognition algorithms to analyze product features including but not limited to appearance, size and weight, and automatically identify and sort different products; the intelligent sorting system can efficiently allocate products to corresponding orders, reducing manual intervention and error rates.
[0017] In a second aspect, an intelligent fresh produce sorting method is provided, characterized in that it comprises:
[0018] Generate demand information instructions through the order management system and send the demand information to the MRP intelligent calculation system;
[0019] According to the acquired demand information instruction, as well as the stored historical data and inventory information, the MRP intelligent computing system performs intelligent analysis to implement at least one of the following operations:
[0020] 1) Long-term warning: According to the immediate demand information instructions or periodic demand information instructions of the order management system, based on the historical data of a certain period, the future sales of product A are predicted, and decision analysis is performed according to the inventory information. When the inventory quantity and / or weight of product A is lower than the threshold of the sales situation in the next n days, an alarm is issued for the inventory of product A, and a purchase order plan is issued, where n is an integer greater than zero;
[0021] 2) Real-time analysis: Receive the sales demand instruction for product A from the order system, call the inventory information, and conduct real-time sales demand analysis. When the inventory quantity and / or weight of product A is greater than the sales demand quantity and / or weight of product A, push the sorting instruction to the intelligent sorting system; when the inventory quantity and / or weight of product A is less than the sales demand quantity and / or weight of product A, feedback an alarm prompt to the order management system, and the order management system sends an instruction to the client, requesting the client to terminate, suspend or delay delivery;
[0022] According to the demand information instruction of the order management system and / or the analysis result of the MRP intelligent operation system, product A is sorted from the storage area, and the required weight and / or quantity of product A is obtained according to the demand information;
[0023] Generate a corresponding electronic code sheet for product A obtained in the sorting task, including information such as product name, quantity, weight, etc., and send the product delivery information to the intelligent delivery unit, which performs corresponding delivery allocation operations.
[0024] Preferably, the long-term warning adopts a machine learning prediction model for prediction, which establishes a prediction model through intelligent analysis and learning of historical sales data, weather data, and seasonal data, predicts future demand trends, and makes threshold judgments based on these prediction results, and issues purchase order plans, so as to more accurately meet the order needs of the next n days.
[0025] Preferably, the intelligent sorting uses computer vision technology and / or object recognition algorithms to analyze product features including but not limited to appearance, size and weight, and automatically identify and sort different products; the intelligent sorting system can efficiently allocate products to corresponding orders, reducing manual intervention and error rates.
[0026] In a third aspect, the present application further provides a computer-readable storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, the fresh food sorting method as described in any embodiment of the present application is implemented.
[0027] The present application relates to an intelligent fresh produce sorting method, system and storage medium, including an order management system, an MRP intelligent computing system, an intelligent sorting system and a distribution service system. The order management system of the present application can realize the rapid receipt, processing and management of orders, automatically process a large number of orders, and realize the rapid processing and distribution of orders.
[0028] The MRP intelligent computing system conducts long-term early warning and real-time analysis based on order demand information and historical data to predict sales and trigger alarm prompts or issue purchase order plans.
[0029] The intelligent sorting system integrates weighing equipment and sorting systems to realize automatic weighing and intelligent sorting operations. By connecting with the order management system and distribution service system, it associates order information with weighing data to achieve accurate sorting and packaging.
[0030] The system and method of the present application utilize advanced technical means to solve the pain points and technical problems in the field of fresh produce sorting, such as high error rate, excessive fresh produce loss, and time-consuming and labor-intensive manual labor, thereby improving operational efficiency and customer satisfaction.
[0031] Based on the above ideal embodiments of the present application, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, but must be determined according to the scope of the claims. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a structural schematic diagram of an intelligent fresh produce sorting system according to a preferred embodiment of the present invention;
[0033] Figure 2 This is a schematic diagram of the structure of an intelligent fresh produce sorting system according to a second preferred embodiment of the present invention;
[0034] Figure 3 It is a flow chart of the intelligent fresh produce sorting method according to the third preferred embodiment of the present invention;
[0035] Figure 4 It is a structural schematic diagram of an electronic device provided by a fourth preferred embodiment of the present invention.
[0036] Embodiments of the invention
[0037] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described through implementation methods with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In the following embodiments, each embodiment provides optional features and examples at the same time. The various features recorded in the embodiments can be combined to form multiple optional solutions. Each numbered embodiment should not be regarded as only one technical solution.
[0038] Embodiment 1:
[0039] Figure 1 An intelligent fresh food sorting system provided in the first embodiment of the present invention specifically includes the following components:
[0040] The order management system 1 is used to generate demand information instructions and send the demand information instructions to the MRP intelligent operation system;
[0041] The MRP intelligent computing system 2 is used to obtain demand information instructions from the order management system 1, and / or perform intelligent analysis based on stored historical data and inventory information to achieve at least one of the following operations:
[0042] 1) Long-term warning: According to the immediate demand information instructions or periodic demand information instructions of the order management system 1, such as automatically issuing a long-term warning demand information instruction every 7 days, based on a certain period of historical data (this embodiment is set to the sales information of the past quarter), predict the future sales of product A. According to the inventory information, when the inventory quantity and / or weight of product A (such as frozen lamb legs) is lower than the preset threshold value (such as 20 pieces) of the sales in the next 7 days, an alarm is issued for the inventory of product A (frozen lamb legs) and a purchase order plan is issued.
[0043] 2) In addition, the system can perform instant analysis: receive sales demand instructions for product A (frozen lamb legs) from the order system, call inventory information, and perform instant sales demand analysis; when the inventory quantity and / or weight of product A (frozen lamb legs) is greater than the sales demand quantity and / or weight of product A (frozen lamb legs) (such as 7 pieces), push the sorting instruction to the intelligent sorting system; when the inventory quantity and / or weight of frozen lamb legs is less than the sales demand quantity and / or weight of product A, feedback an alarm to the order management system 1, which sends instructions to the client, requesting the client to terminate, suspend or delay delivery, and perform the next step based on the client's feedback. If delivery is delayed, an immediate purchase order will be issued; if suspended, the information on active arrival will be pushed to the client after the order under long-term warning is completed;
[0044] The intelligent sorting system 3 sorts product A (frozen lamb legs) from the storage area according to the demand information instruction of the order management system 1 and / or the analysis result of the MRP intelligent operation system 2, and obtains the required weight and / or quantity of product A (frozen lamb legs) according to the demand information;
[0045] Delivery service system 4, including but not limited to:
[0046] 1) An electronic code sheet printing system 41 is used to generate a corresponding electronic code sheet for product A (frozen lamb leg) obtained in the sorting task, including information such as product name, quantity, weight, etc., and issue the product delivery information;
[0047] 2) An intelligent distribution unit 42, which is used to receive the electronic code sheet and perform corresponding distribution operations.
[0048] In an embodiment of the present invention, the long-term warning of the MRP intelligent computing system 22 adopts a machine learning prediction model for prediction. It establishes a prediction model by intelligently analyzing and learning historical sales data, weather data, and seasonal data, predicting future demand trends, and based on these prediction results, performing threshold judgments and issuing purchase order plans, thereby more accurately meeting order demands for the next n days (such as 7 days).
[0049] In an embodiment of the present invention, the intelligent sorting system 3 uses computer vision technology and / or object recognition algorithm to analyze the characteristics of the product, including but not limited to appearance, size and weight, and automatically identify and sort different products; the intelligent sorting system 3 can efficiently allocate products to corresponding orders, reducing manual intervention and error rate.
[0050] Embodiment 2
[0051] Figure 2It is a structural diagram of an intelligent fresh food sorting system provided in the second embodiment of the present invention. This embodiment provides a specific machine learning prediction model, computer vision technology and / or object recognition algorithm, which can be executed by the intelligent fresh food sorting system provided in the embodiment of the present invention. The training device is implemented in software and / or hardware, and is preferably configured in an electronic device, such as a server of the intelligent fresh food sorting system. The embodiment of the present invention has the same inventive concept as the above embodiment, and the technical details not described in this embodiment can be found in the above embodiment.
[0052] In the embodiment of the present invention, the machine learning prediction model meets the following conditions:
[0053] 1) Collect and preprocess data: collect historical data from the past period of time, including but not limited to sales data Q, weather data T and seasonal data J, and represent the sales data, weather data and seasonal data as vectors Q, T and J respectively; process the historical data, including but not limited to data cleaning, missing value filling and feature engineering; obtain a training data set D and a test data set Dtest;
[0054] 2) Performing model training: constructing an LSTM neural network model f for long-term warning of fresh food sorting, including an input layer, an LSTM layer, an output layer, etc.; the LSTM neural network model f includes an LSTM layer, a fully connected layer, and an activation function; using a training data set D to train the model, and using a back propagation algorithm to optimize the model parameters so as to minimize the prediction error of the model; using methods such as cross-validation to select and tune the model to select the best model parameters and hyperparameters;
[0055] 3) Performing model evaluation: Using the test data set Dtest to evaluate the trained model, including but not limited to calculating the prediction error and / or evaluation index of the model;
[0056] 4) Sales volume forecast: Set the forecast time length to n days (e.g. 7 days), input future weather conditions, seasonal data, and historical sales data, use the trained model to make forecasts, and output the forecast value Qv of sales volume in the next n days (7 days). The LSTM neural network forecast model formula can be expressed as:
[0057] Qv=f(Q,T,J)=σ(Wq×Q+WT×T+Wj×J+b)
[0058] Among them, Q represents historical sales data, T represents weather condition data, and J represents seasonal data; they are represented as vectors Q, T and J respectively.
[0059] Wq, WT, Wj are weight matrices of historical sales data, weather conditions, and seasonal data, respectively, and b is a bias vector. The parameters are learned through training;
[0060] σ is the activation function, usually ReLU, tanh and other functions are used.
[0061] f(Q,T,J) represents the future sales forecast output of the LSTM neural network, that is, the predicted value Qv of sales in the next n days, which is obtained by inputting the input data into the LSTM neural network and calculating the output.
[0062] Vector Q: represents historical sales data, which can be a vector containing sales data at multiple time points. For example, if the sales data of the past 30 days is considered, then vector Q can be represented as a vector of length 30, where each element represents the sales volume of a certain day in the past 30 days.
[0063] Vector T: represents weather data, which can be a vector containing weather data at multiple time points. For example, if the weather conditions in the next 7 days are considered, then vector T can be represented as a vector of length 7, where each element represents the weather conditions of a certain day in the next 7 days. Specific weather data can be selected according to actual needs, such as temperature, humidity, rainfall, etc.
[0064] Vector J: represents seasonal data, which can be a vector containing seasonal data at multiple time points. For example, if the seasonal data for the next 12 months is considered, then vector J can be represented as a vector of length 12, where each element represents the seasonal data for a certain month in the next 12 months. The specific seasonal data can be selected according to actual needs, such as month, quarter, etc.
[0065] Weight matrices Wq, WT, and Wj: weight matrices corresponding to historical sales data, weather data, and seasonal data, respectively. These weight matrices are learned through the training process and are used to adjust the impact of different data on the prediction.
[0066] Bias vector b: used to adjust the offset of the prediction function to improve the prediction effect by increasing or decreasing the constant value of the predicted value.
[0067] Activation function σ: It is used to perform nonlinear transformation on the output of the LSTM neural network to increase the expressiveness of the model. Common activation functions include ReLU (Rectified Linear Unit), tanh (Hyperbolic Tangent), sigmoid, etc.
[0068] Prediction value Qv: represents the output of the LSTM neural network, that is, the predicted value of sales in the next n days. It is obtained by applying the prediction model to the input data and is the model's prediction result of future sales volume.
[0069] The specific values of the above parameters need to be learned through the training process, and the weight matrix and bias vector are adjusted by minimizing the prediction error so that the model can obtain more accurate prediction results.
[0070] For example, to predict the sales volume of a product in the next 7 days, the dimensions of historical sales data, weather conditions, and seasonal conditions are 30 days, 5 weather types, and 3 seasonal types, respectively. Then the corresponding weight matrix can be as follows:
[0071] Wq: A matrix of size (30 x 7) used to perform weight transformation on the historical sales data. For example, if the historical sales data contains 30 days of sales data, and we want to predict the sales for the next 7 days, then Wq will contain a weight matrix with 30 rows and 7 columns.
[0072] WT: A matrix of size (5 x 7) used to transform the weights of weather conditions. For example, if there are 5 types of weather conditions (sunny, cloudy, rainy, snowy, and foggy), and we want to predict the sales volume for the next 7 days, then WT will contain a weight matrix with 5 rows and 7 columns.
[0073] Wj: A matrix of size (3 x 7) used to transform the weights of seasonal conditions. For example, if there are 3 types of seasonal conditions (cold season, warm season, and hot season), and we want to predict the sales volume for the next 7 days, then Wj will contain a weight matrix with 3 rows and 7 columns.
[0074] These weight matrices will be learned during the training process of the LSTM neural network to optimally weight historical sales data, weather conditions, and seasonal conditions, thereby obtaining more accurate sales forecast results.
[0075] In the embodiment of the present invention, the historical data further includes comprehensive additional information, including but not limited to holiday characteristics, promotion activity characteristics, product attribute characteristics and other information; Z represents the comprehensive additional information, which is a vector; Wz is the weight matrix of the comprehensive additional information, and the model for future sales prediction considering the comprehensive additional information Z obtained through training and learning satisfies the following conditions:
[0076] Qv=f(Q,T,J,Z)=σ(Wq×Q+WT×T+Wj×J+Wz×Z+b).
[0077] In the embodiment of the present invention, the data of the training data set D satisfies: D = {(Q_1, T_1, J_1, Z_1), (Q_2, T_2, J_2, Z_2), ..., (Q_m, T_m, J_m, Z_m)}; wherein each sample data is composed of historical sales data Q, weather data T, seasonal data J and comprehensive additional information Z, and m represents the number of samples in the training data set D;
[0078] The data of the test data set Dtest satisfies: Dtest = {(Q'_1, T'_1, J'_1, Z'_1), (Q'_2, T'_2, J'_2, Z'_2), ..., (Q'_n, T'_n, J'_n, Z'_n)}, wherein each sample data consists of historical sales data Q', weather condition data T', seasonal data J' and comprehensive additional information Z', and n represents the number of samples in the test data set Dtest.
[0079] In the embodiment of the present invention, the back propagation algorithm meets the following conditions:
[0080] 1) Input a sample (Q, T, J, Z) in the training data set D into the model f; obtain the output H of the hidden layer through calculations of the input layer and the LSTM layer; obtain the predicted value Qv through calculations of the output layer for the output H of the hidden layer;
[0081] 2) Compare the predicted value Qv with the corresponding actual sales data, calculate the value of the loss function, and measure the prediction error of the model;
[0082] 3) According to the loss function, calculate the gradient of each parameter (weight matrix and bias vector) in the model, that is, find the partial derivative of each parameter in the loss function;
[0083] 4) Use the gradient descent algorithm or other optimization algorithms to update the parameters in the model according to the setting of the learning rate and reduce the value of the loss function.
[0084] In the embodiment of the present invention, the cross-validation method satisfies the following conditions:
[0085] 1) Divide the training dataset D into K non-overlapping subsets (usually called folds), denoted as D_1, D_2, ..., D_K;
[0086] 2) For each fold D_i, use it as a validation set and the other K-1 folds as training sets. Use the training set to train the model and make predictions on the validation set. Calculate the evaluation indicators on the validation set.
[0087] 3) Repeat the above step 2) to divide each fold D_i into a validation set and a training set, and obtain the results of K times of training and validation;
[0088] 4) Calculate the average of K validation results as the final evaluation index of the model;
[0089] 5) Select the best model parameters and hyperparameter combinations based on the final evaluation indicators.
[0090] Specific formulas and calculation processes, such as the specific forms of loss functions and evaluation indicators, as well as specific algorithms for parameter updating and model selection, need to be determined based on the actual application scenario and the choice of specific models.
[0091] In the embodiment of the present invention, the computer vision technology and / or object recognition meets the following conditions:
[0092] 1) Input data: object visual features V, represented as feature vectors, including appearance features and size features, with a dimension of (D1,1); weight features W, represented as the weight of the object, with a dimension of (1);
[0093] 2) Establish an object visual recognition model: input V, establish an object visual recognition function: f_v(V), output the estimated value of the object classification label, expressed as L_hat, that is, L_hat=f_v(V); input W, establish a weight recognition function: f_w(W), output the estimated value of the object weight, expressed as W_hat, that is, W_hat=f_w(W);
[0094] 3) Set allocation rules and optimization algorithm: input L_hat and W_hat, set the allocation rule function: allocate(L_hat, W_hat), and output the optimized allocation result, expressed as A, that is, A=allocate(L_hat, W_hat).
[0095] The detailed implementation example illustrates the specific function calculation process:
[0096] 1) For object visual recognition model:
[0097] Input: V, represented as a feature vector, including appearance features and size features. Assume that the dimension of V is (D1,1).
[0098] Object visual recognition function: f_v(V), for example, using a convolutional neural network (CNN) model.
[0099] Output: The estimated classification label of the object, denoted as L_hat.
[0100] Object visual recognition function calculation process:
[0101] First, the input feature vector V is passed into the model according to the weights and biases of the trained CNN model.
[0102] Through the calculation of convolution, pooling and full connection layers, the feature vector V is converted into the estimated probability distribution of object classification, expressed as P(L|V).
[0103] Finally, according to the probability distribution P(L|V), the classification label with the highest probability is selected as the classification label estimate L_hat of the object.
[0104] 2) Weight recognition model:
[0105] Input: W, which is the weight of the object.
[0106] Weight identification function: f_w(W), such as a linear regression model.
[0107] Output: An estimate of the object’s weight, denoted as W_hat.
[0108] Weight recognition function calculation process:
[0109] The linear regression model parameters (weights and biases) obtained through training are used to pass the input weight value W to the model for calculation.
[0110] According to the calculation formula of the linear regression model, the estimated value W_hat of the object weight is calculated.
[0111] Allocation rules and optimization algorithms:
[0112] Input: L_hat and W_hat, which represent the estimated classification label and weight of the object respectively.
[0113] Allocation rule function: allocate(L_hat,W_hat), for example, rule-based allocation method.
[0114] Output: The optimal allocation result, denoted as A.
[0115] 3) Allocation rule function calculation process:
[0116] Based on the values of L_hat and W_hat, pre-defined allocation rules are used, such as determining the target order or container for allocation based on the classification label and weight threshold.
[0117] According to the calculation process of the allocation rule, the estimated classification labels and weight values are converted into the optimized allocation result A.
[0118] The function calculation process in the above example is a description of an embodiment. The present application may be based on a specific object visual recognition model, a weight recognition model and an allocation rule function. The specific calculation process and model parameters may vary.
[0119] In the embodiment of the present invention, it further comprises a sensor or RFID disposed in the storage area to monitor the storage and retrieval information of the materials in real time, and to feed back the relevant data to the MRP intelligent computing system 2 and / or the intelligent sorting system 3, so as to update the bill of materials information in a timely manner;
[0120] In the embodiment of the present invention, the sensor or RFID tag set on product A is used to monitor the changes in the inventory of product A in real time, and transmit the data to the MRP intelligent computing system 2 and / or the intelligent sorting system 3 to ensure the accuracy of the inventory information and avoid out-of-stock or overstock situations.
[0121] In the embodiment of the present invention, the plan for issuing the purchase order is one of the following situations:
[0122] 1) The purchase order plan issued is a production demand order plan, which is pushed to the production workshop for order plan production;
[0123] 2) The purchase order plan issued is a commissioned processing or purchase order plan, which is pushed to the procurement unit for commissioned processing or procurement.
[0124] In an embodiment of the present invention, the intelligent sorting system 3 includes an intelligent weighing module, which is used to weigh product A through an intelligent networked electronic scale platform, and associate the weight information of product A with the order information of the order management system 1 and / or the MRP intelligent computing system 2, and automatically calculate the weight and quantity of the sorting task based on the weight and quantity of product A in the order information.
[0125] In the embodiment of the present invention, the code sheet printing system generates a corresponding electronic code sheet according to the sorting task, including information such as product name, quantity, weight, etc. The electronic code sheet can be directly used for delivery management and customer online signing.
[0126] In the embodiment of the present invention, the intelligent delivery unit 42 can intelligently arrange delivery routes, vehicles, personnel, etc. according to the code list information, optimize and issue delivery tasks, and can also track the delivery progress and location information of product A in real time.
[0127] In the embodiment of the present invention, a background workshop visual monitoring system is further included, which is used to display the progress, quantity, weight and other information of each sorting task in real time, and to alarm and prompt abnormal situations, so as to adjust the production plan and delivery route in time.
[0128] Embodiment 3
[0129] Figure 3 It is a flow chart of the intelligent fresh produce sorting method according to the third preferred embodiment of the present invention;
[0130] See also Figure 3, provides an intelligent fresh food sorting method, characterized by comprising:
[0131] S1: Generate demand information instructions through the order management system 1 and send the demand information to the MRP intelligent calculation system 2;
[0132] S2: Based on the acquired demand information instructions, as well as the stored historical data and inventory information, the MRP intelligent computing system 2 performs intelligent analysis and implements at least one of the following operations:
[0133] 1) S201: Long-term warning. According to the immediate demand information instruction or the periodic demand information instruction of the order management system 1, based on the historical data of a certain period, the future sales of product A are predicted. According to the inventory information, when the inventory quantity and / or weight of product A is lower than the threshold of the sales in the next n days, an alarm is issued for the inventory of product A, and a purchase order plan is issued, where n is an integer greater than zero;
[0134] 2) S202: Real-time analysis. Receive the sales demand instruction for product A from the order system, call the inventory information, and conduct real-time sales demand analysis. When the inventory quantity and / or weight of product A is greater than the sales demand quantity and / or weight of product A, push the sorting instruction to the intelligent sorting system 3; when the inventory quantity and / or weight of product A is less than the sales demand quantity and / or weight of product A, feedback an alarm prompt to the order management system 1, and the order management system 1 sends an instruction to the client, requesting the client to terminate, suspend or delay delivery;
[0135] S3: sorting product A from the storage area according to the demand information instruction of the order management system 1 and / or the analysis result of the MRP intelligent operation system 2, and obtaining the required weight and / or quantity of product A according to the demand information;
[0136] S4: Generate a corresponding electronic code sheet for the product obtained by the sorting task, including information such as product name, quantity, weight, etc., and send the product delivery information to the intelligent delivery unit 42, which performs corresponding delivery allocation operations.
[0137] In an embodiment of the present invention, the long-term warning adopts a machine learning prediction model for prediction, which establishes a prediction model through intelligent analysis and learning of historical sales data, weather data, seasonal data, etc., predicts future demand trends, and makes threshold judgments based on these prediction results, and issues purchase order plans, so as to more accurately meet the order needs of the next n days.
[0138] In an embodiment of the present invention, the intelligent sorting uses computer vision technology and / or object recognition algorithm to analyze the characteristics of the product, including but not limited to appearance, size and weight, and automatically identify and sort different products; the intelligent sorting system 3 can efficiently allocate products to corresponding orders, reducing manual intervention and error rate.
[0139] Embodiment 4
[0140] Figure 4 It is a structural schematic diagram of an electronic device provided in Example 4 of the present invention. Figure 4 A block diagram of an exemplary electronic device 12 suitable for use in implementing embodiments of the present invention is shown.
[0141] Figure 4 The electronic device 12 shown is only an example and should not bring any limitation to the functions and application scope of the embodiments of the present invention. The device 12 is typically an electronic device that undertakes the intelligent fresh food sorting method model training or sorting function.
[0142] like Figure 4 As shown, the electronic device 12 is in the form of a general purpose computing device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a memory 28, and a bus 18 connecting the various components (including the memory 28 and the processing unit 16).
[0143] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
[0144] The electronic device 12 typically includes a variety of computer readable media. These media can be any available media that can be accessed by the electronic device 12, including volatile and non-volatile media, removable and non-removable media.
[0145] The memory 28 may include computer device readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a read-only optical disk (Compact Disc-Read Only Memory, CD-ROM), a digital video disk (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product 40, which has a set of program modules 42, which are configured to perform the functions of various embodiments of the present invention. The program product 40 may be stored in, for example, the memory 28, and such program modules 42 include, but are not limited to, one or more application programs, other program modules, and program data, each of which or some combination may include the implementation of a network environment. The program module 42 generally performs the functions and / or methods in the embodiments described in the present invention.
[0146] The electronic device 12 may also communicate with one or more external devices 14 (e.g., keyboard, mouse, camera, etc., and display), may also communicate with one or more devices that enable a user to interact with the electronic device 12, and / or communicate with any device that enables the electronic device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. In addition, the electronic device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN) and / or a public network, such as the Internet) through a network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device 12 through a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, disk arrays (RAID) devices, tape drives, and data backup storage devices.
[0147] The processor 16 executes various functional applications and data processing by running the program stored in the memory 28, for example, implementing the intelligent fresh food sorting method provided in the above embodiment of the present invention, which includes:
[0148] The order management system 1 generates a demand information instruction, and sends the demand information to the MRP intelligent operation system 2; based on the acquired demand information, as well as the stored historical data and inventory information, the MRP intelligent operation system 2 performs intelligent analysis to achieve at least one of long-term warning or instant analysis operations. The long-term warning is to predict the future sales of product A based on the immediate demand information instruction or periodic demand information instruction of the order management system 1 and the historical data of a certain period, and according to the inventory information, when the inventory quantity and / or weight of product A is lower than the threshold of the sales in the next n days, the inventory of product A is warned and a purchase order plan is issued, where n is an integer greater than zero. The instant analysis is to receive the sales demand instruction for product A from the order system, call the inventory information, and perform an instant sales demand analysis. When the inventory quantity and / or weight of product A is greater than the quantity and / or weight of the sales demand of product A, push the sorting instruction to the intelligent sorting system 3; when the inventory quantity and / or weight of product A is less than the quantity and / or weight of the sales demand of product A, feedback an alarm prompt to the order management system 1, and the order management system 1 sends an instruction to the client to request the client to terminate, suspend or delay delivery. According to the demand information instruction of the order management system 1 and / or the analysis results of the MRP intelligent operation system 2, sort product A from the storage area, and obtain the required weight and / or quantity of product A according to the demand information. Generate a corresponding electronic code sheet for the product obtained by the sorting task, including information such as product name, quantity, weight, etc., and send the product delivery information to the intelligent delivery unit 42, and the intelligent delivery unit 42 performs the corresponding delivery distribution operation.
[0149] Alternatively, to implement the intelligent fresh food sorting method provided in the above-mentioned embodiment of the present invention, the method further includes: the long-term warning is predicted by using a machine learning prediction model, which establishes a prediction model by intelligently analyzing and learning historical sales data, weather data, and seasonal data, predicting future demand trends, and based on these prediction results, performing threshold judgments and issuing purchase order plans, thereby more accurately meeting order demands for the next n days.
[0150] Alternatively, to implement the intelligent fresh food sorting method provided in the above-mentioned embodiment of the present invention, the method also includes: the intelligent sorting uses computer vision technology and / or object recognition algorithm to analyze the characteristics of the product including but not limited to appearance, size and weight, and automatically identifies and sorts different products; the intelligent sorting system 3 can efficiently allocate products to corresponding orders, reducing manual intervention and error rate.
[0151] Of course, those skilled in the art will appreciate that the processor may also implement the technical solution of the intelligent fresh produce sorting method provided in any embodiment of the present invention.
[0152] Embodiment 5
[0153] A computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the fresh produce sorting method as described in any embodiment of the present application is implemented.
[0154] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor device, device or component, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution device, device or component.
[0155] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution device, apparatus, or device.
[0156] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0157] Computer program code for performing the operations of the present invention may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).
[0158] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims
1. An intelligent fresh food sorting system, comprising the following components: An order management system, used to generate demand information instructions and send the demand information instructions to the MRP intelligent computing system; The MRP intelligent computing system is used to obtain the demand information instructions sent by the order management system and perform intelligent analysis based on the stored historical data and inventory information to achieve the following operations: 1) Long-term warning: According to the instant demand information instruction or the periodic demand information instruction of the order management system, based on the historical data of a certain period, the future sales volume of product A is predicted, and decision analysis is performed according to the inventory information. When the inventory quantity and / or weight of product A is lower than the preset threshold of the sales volume in the next n days, an alarm is issued for the inventory of product A, and a purchase order plan is issued, where n is an integer greater than zero; 2) Real-time analysis: receiving the sales demand instruction for product A from the order management system, calling the inventory information, and performing real-time sales demand analysis. When the inventory quantity and / or weight of product A is greater than the sales demand quantity and / or weight of product A, a sorting instruction is pushed to the intelligent sorting system; when the inventory quantity and / or weight of product A is less than the sales demand quantity and / or weight of product A, an alarm is fed back to the order management system, and the order management system sends an instruction to the client, requesting the client to terminate, suspend or delay delivery; An intelligent sorting system, which sorts product A from the storage area according to the demand information instruction of the order management system and / or the analysis result of the MRP intelligent operation system, and obtains the required weight and / or quantity of product A according to the demand information instruction; a distribution service system, including but not limited to: 1) An electronic code sheet printing system generates a corresponding electronic code sheet for product A obtained by the sorting task according to the analysis results of the intelligent sorting system, including product name, quantity, and weight information, and issues the product delivery information; 2) An intelligent distribution unit, which is used to receive the electronic code sheet and perform corresponding distribution operations; The long-term warning adopts a machine learning prediction model for prediction, which establishes a prediction model through intelligent analysis and learning of historical sales data, weather data, and seasonal data, predicts future demand trends, and makes threshold judgments based on these prediction results, and issues purchase order plans, so as to more accurately meet the order needs of the next n days; The machine learning prediction model meets the following conditions: set the prediction time length to n days, input future weather data T, seasonal data J, and historical sales data Q, use the trained model to make predictions, and output the predicted sales value Qv for the next n days. The LSTM neural network prediction model formula can be expressed as: Qv=f(Q,T,J)=σ(Wq×Q+WT×T+Wj×J+b) Among them, Q represents historical sales data, T represents weather data, and J represents seasonal data; they are represented as vectors Q, T, and J respectively; Wq, WT, Wj are weight matrices of historical sales data, weather data, and seasonal data, respectively, and b is a bias vector. The parameters are learned through training; σ is the activation function, usually ReLU or tanh function; f(Q,T,J) represents the future sales forecast output of the LSTM neural network, that is, the predicted value Qv of sales in the next n days, which is obtained by inputting the input data into the LSTM neural network and calculating the output; The intelligent sorting system includes an intelligent weighing module, which is used to weigh product A through an intelligent networked electronic scale platform, and associate the weight information of product A with the order information of the order management system and / or the MRP intelligent computing system, and automatically calculate the weight and quantity of the sorting task based on the weight and quantity of product A in the order information.
2. The fresh food sorting system according to claim 1, characterized in that: The machine learning prediction model also meets the following conditions: 1) Collect and preprocess data: collect historical data from the past period of time, including but not limited to sales data Q, weather data T and seasonal data J, and represent the sales data, weather data and seasonal data as vectors Q, T and J respectively; process the historical data, including but not limited to data cleaning, missing value filling and feature engineering; Get the training data set D and the test data set Dtest; 2) Model training: constructing an LSTM neural network model f for long-term warning of fresh food sorting, including an input layer, an LSTM layer, and an output layer; the LSTM neural network model f includes an LSTM layer, a fully connected layer, and an activation function; using a training data set D to train the model, and using a back propagation algorithm to optimize the model parameters so that the prediction error of the model is minimized; using a cross-validation method for model selection and tuning; 3) Performing model evaluation: Using the test data set Dtest to evaluate the trained model, including but not limited to calculating the prediction error and / or evaluation index of the model.
3. The fresh food sorting system according to claim 2, characterized in that: The historical data also includes comprehensive additional information, including but not limited to holiday characteristics, promotional activity characteristics, and product attribute characteristic information; Z represents the comprehensive additional information, which is a vector; Wz is the weight matrix of the comprehensive additional information, obtained through training and learning, and the model for future sales prediction considering the comprehensive additional information Z satisfies the following conditions: Qv=f(Q,T,J,Z)=σ(Wq×Q+WT×T+Wj×J+Wz×Z+b).
4. The fresh food sorting system according to claim 3, characterized in that: The data of the training data set D satisfies: D = {(Q_1, T_1, J_1, Z_1), (Q_2, T_2, J_2, Z_2), ..., (Q_m, T_m, J_m, Z_m)}; wherein each sample data is composed of historical sales data Q, weather data T, seasonal data J and comprehensive additional information Z, and m represents the number of samples in the training data set D; The data of the test data set Dtest satisfies: Dtest = {(Q'_1, T'_1, J'_1, Z'_1), (Q'_2, T'_2, J'_2, Z'_2), ..., (Q'_n, T'_n, J'_n, Z'_n)}, wherein each sample data consists of historical sales data Q', weather condition data T', seasonal data J' and comprehensive additional information Z', and n represents the number of samples in the test data set Dtest.
5. The fresh food sorting system according to claim 4, characterized in that: The back propagation algorithm meets the following conditions: 1) Input a sample (Q, T, J, Z) in the training data set D into the model f; obtain the output H of the hidden layer through calculations of the input layer and the LSTM layer; obtain the predicted value Qv through calculations of the output layer for the output H of the hidden layer; 2) Compare the predicted value Qv with the corresponding actual sales data, calculate the value of the loss function, and measure the prediction error of the model; 3) According to the loss function, calculate the gradient of the weight matrix and bias vector in the model, that is, find the partial derivative of each parameter in the loss function; 4) Use the gradient descent algorithm or other optimization algorithms to update the parameters in the model according to the setting of the learning rate and reduce the value of the loss function.
6. The fresh food sorting system according to claim 4, characterized in that: The cross validation method meets the following conditions: 1) Divide the training dataset D into K non-overlapping subsets, denoted as D_1, D_2, ..., D_K; 2) For each fold D_i, use it as a validation set and the other K-1 folds as training sets. Use the training set to train the model and make predictions on the validation set. Calculate the evaluation indicators on the validation set. 3) Repeat the above step 2) to divide each fold D_i into a validation set and a training set, and obtain the results of K times of training and validation; 4) Calculate the average of K validation results as the final evaluation index of the model; 5) Select the best model parameters and hyperparameter combinations based on the final evaluation indicators.
7. The fresh food sorting system according to claim 1, characterized in that: The intelligent sorting system uses computer vision technology and / or object recognition algorithms to analyze the product's features including but not limited to appearance, size and weight, and automatically identifies and sorts different products.
8. The fresh food sorting system according to claim 7, characterized in that: The computer vision technology and / or object recognition meet the following conditions: 1) Input data: object visual features V, represented as feature vectors, including appearance features and size features, with a dimension of (D1,1); weight features W, represented as the weight of the object, with a dimension of (1); 2) Establish an object visual recognition model: input V, establish an object visual recognition function: f_v(V), output the estimated value of the object classification label, expressed as L_hat, that is, L_hat=f_v(V); input W, establish a weight recognition function: f_w(W), output the estimated value of the object weight, expressed as W_hat, that is, W_hat=f_w(W); 3) Set allocation rules and optimization algorithm: input L_hat and W_hat, set the allocation rule function: allocate(L_hat, W_hat), and output the optimized allocation result, expressed as A, that is, A=allocate(L_hat, W_hat).
9. The fresh food sorting system according to claim 1, characterized in that: It further includes sensors or RFIDs arranged in the storage area to monitor the storage and retrieval information of materials in real time and feed back the relevant data to the MRP intelligent computing system and / or the intelligent sorting system.
10. The fresh food sorting system according to claim 1, characterized in that: The sensor or RFID tag installed on product A is used to monitor the inventory changes of product A in real time and transmit the data to the MRP intelligent computing system and / or the intelligent sorting system.
11. The fresh food sorting system according to claim 1, characterized in that: The purchase order plan is issued in one of the following situations: 1) The purchase order plan issued is a production demand order plan, which is pushed to the production workshop for order plan production; 2) The purchase order plan issued is a commissioned processing or purchase order plan, which is pushed to the procurement unit for commissioned processing or procurement.
12. The fresh food sorting system according to claim 1, characterized in that: The code sheet printing system generates the corresponding electronic code sheet according to the sorting task, including product name, quantity, and weight information. The electronic code sheet is used for delivery management and customer online signing.
13. The fresh food sorting system according to claim 1, characterized in that: The intelligent distribution unit intelligently arranges the distribution route, vehicles, and personnel according to the code list information, issues and optimizes the distribution tasks, and tracks the distribution progress and the location information of product A in real time.
14. The fresh food sorting system according to claim 1, characterized in that: It further includes a background visual monitoring system, which is used to display the progress, quantity, and weight information of each sorting task in real time, and to issue alarms and prompts for abnormal situations, so as to adjust the production plan and delivery route in time.
15. An intelligent fresh food sorting method, characterized in that: include: Generate demand information instructions through the order management system, and send the demand information instructions to the MRP intelligent calculation system; The MRP intelligent computing system performs intelligent analysis based on the acquired demand information instructions, as well as the stored historical data and inventory information, to achieve the following operations: 1) Long-term warning: According to the instant demand information instruction or the periodic demand information instruction of the order management system, based on the historical data of a certain period, the future sales volume of product A is predicted; And based on the inventory information, a decision analysis is performed. When the inventory quantity and / or weight of product A is lower than the preset threshold of the sales volume in the next n days, an alarm is issued for the inventory of product A, and a purchase order plan is issued, where n is an integer greater than zero; 2) Real-time analysis: receiving the sales demand instruction for product A from the order system, calling the inventory information, and performing real-time sales demand analysis. When the inventory quantity and / or weight of product A is greater than the sales demand quantity and / or weight of product A, the sorting instruction is pushed to the intelligent sorting system; when the inventory quantity and / or weight of product A is less than the sales demand quantity and / or weight of product A, an alarm is fed back to the order management system, and the order management system sends an instruction to the client, requesting the client to terminate, suspend or delay delivery; According to the demand information instruction of the order management system and / or the analysis result of the MRP intelligent operation system, product A is sorted from the storage area, and the required weight and / or quantity of product A is obtained according to the demand information; Generate a corresponding electronic code sheet for product A obtained in the sorting task, including product name, quantity, and weight information, and send the product delivery information to the intelligent delivery unit, which performs corresponding delivery allocation operations; The long-term warning adopts a machine learning prediction model for prediction, which establishes a prediction model through intelligent analysis and learning of historical sales data, weather data, and seasonal data, predicts future demand trends, and makes threshold judgments based on these prediction results, and issues purchase order plans, so as to more accurately meet the order needs of the next n days; The machine learning prediction model meets the following conditions: set the prediction time length to n days, input future weather data T, seasonal data J, and historical sales data Q, use the trained model to make predictions, and output the predicted sales value Qv for the next n days. The LSTM neural network prediction model formula can be expressed as: Qv=f(Q,T,J)=σ(Wq×Q+WT×T+Wj×J+b) Among them, Q represents historical sales data, T represents weather data, and J represents seasonal data; they are represented as vectors Q, T, and J respectively; Wq, WT, Wj are weight matrices of historical sales data, weather data, and seasonal data, respectively, and b is a bias vector. The parameters are learned through training; σ is the activation function, usually ReLU or tanh function; f(Q,T,J) represents the future sales forecast output of the LSTM neural network, that is, the predicted value Qv of sales in the next n days, which is obtained by inputting the input data into the LSTM neural network and calculating the output; The intelligent sorting system includes an intelligent weighing module, which is used to weigh product A through an intelligent networked electronic scale platform, and associate the weight information of product A with the order information of the order management system and / or the MRP intelligent computing system, and automatically calculate the weight and quantity of the sorting task based on the weight and quantity of product A in the order information.
16. The fresh food sorting method according to claim 15, characterized in that: The intelligent sorting uses computer vision technology and / or object recognition algorithms to analyze the product's features including but not limited to appearance, size and weight, and automatically identifies and sorts different products.
17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the fresh food sorting method as described in any one of claims 15-16 is implemented.
18. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the fresh food sorting method as described in any one of claims 15-16 is implemented.
