A fresh food storage control method based on industrial Internet

Abstract

This application provides a method for fresh food storage control based on the Industrial Internet. This method acquires fresh food storage information. The fresh food storage information is sent to an environmental condition model to obtain suitable storage environment data for the corresponding fresh food. This suitable storage environment data includes pressure conditions, temperature conditions, and humidity conditions. A fresh food identification model is used to identify images of the fresh food to determine if surface defects exist. This fresh food identification model is trained using images of several different types of fresh food samples. Based on the suitable storage environment data and the output of the fresh food identification model, it is determined whether the current storage environment of the fresh food matches the corresponding suitable storage environment data. The current storage environment is obtained through an active identification carrier set in the fresh food's storage environment. If the current storage environment of the fresh food does not match the corresponding suitable storage environment data, environmental control information is generated.

Description

Technical Field

[0001] This application relates to the field of industrial internet technology, and in particular to a method for fresh food storage control based on industrial internet. Background Art

[0002] With the improvement of people's living standards and the rapid development of the logistics industry, the consumption of fresh produce is increasing day by day, but problems such as fresh produce storage environment and chaotic logistics and transportation management still exist.

[0003] Fresh produce, such as frozen fish, is susceptible to damage and spoilage when stored in unstable or unsuitable environments, such as warehouses or transport vehicles. This can lead to the fish thawing and spoiling, and the spoilage of one item can easily trigger spoilage in adjacent items, causing significant losses for businesses. Furthermore, unsuitable storage conditions can prevent businesses from promptly detecting spoiled produce, potentially leading to disputes and a negative consumer experience. Summary of the Invention

[0004] This application provides a method for fresh food storage control based on the Industrial Internet, which is used to ensure the stability of the storage environment for fresh products and improve consumers' fresh food purchasing experience.

[0005] On the one hand, embodiments of this application provide a method for fresh food storage control based on the Industrial Internet, the method comprising:

[0006] Acquire fresh produce storage information. This information is sent to an environmental condition model to obtain suitable storage environment data for the corresponding fresh produce. This suitable storage environment data includes pressure, temperature, and humidity conditions. A fresh produce identification model is used to identify images of the fresh produce to determine if surface defects exist. This model is trained using images of several different types of fresh produce samples. Based on the suitable storage environment data and the output of the fresh produce identification model, it is determined whether the current storage environment of the fresh produce matches the corresponding suitable storage environment data. The current storage environment is obtained through an active identification carrier set in the fresh produce's storage environment. If the current storage environment of the fresh produce does not match the corresponding suitable storage environment data, environmental control information is generated.

[0007] In one implementation of this application, an environmental condition model is used to determine the fresh produce type in the fresh produce storage information. This fresh produce storage information includes at least one or more of the following: fresh produce image and fresh produce name. The fresh produce type is determined based on the fresh produce's storage conditions. The model also determines whether any fresh produce of a particular type has incompatible storage conditions. If no incompatible fresh produce exists, the environmental condition model determines suitable storage environment data for the fresh produce. If incompatible fresh produce exists, fresh produce storage prompts are generated, and the suitable storage environment data for each fresh produce type output by the environmental condition model is determined. These fresh produce storage prompts are used to instruct operators to properly classify and store the fresh produce.

[0008] In one implementation of this application, several different types of fresh food sample images are input into a fresh food recognition model. The fresh food sample images include surface defect information. The fresh food sample images include at least seafood products. The surface features and contour features of each fresh food sample image are extracted using the fresh food recognition model for model training. If the loss function value of the fresh food recognition model is less than a preset threshold, the trained fresh food recognition model is obtained.

[0009] In one implementation of this application, the name of the seafood product is determined. The seafood product name is then matched against a pre-stored list of fresh seafood names. This pre-stored list includes at least: hairy crab, king crab, and lobster. If the seafood product name matches one item in the pre-stored list, the surface and contour features of each fresh seafood sample image are extracted using a fresh seafood recognition model corresponding to the matched name. The images corresponding to these surface and contour features are then sent to a pre-set comparison image database to determine the fresh seafood's viability.

[0010] In one implementation of this application, a fresh produce identification model is used to determine the quantity of fresh produce corresponding to its storage information when surface defects are identified. The ratio of the quantity of defective fresh produce to the total quantity of fresh produce is then checked against a preset value. If not, an image of the surface defects is sent to an expert terminal to determine whether the surface defects constitute an inedible condition.

[0011] In one implementation of this application, identification data and current storage environment data from each active identification carrier are received. Based on the environmental data, corresponding environmental condition curves for each identification within a preset time period are generated. These environmental condition curves include at least: a pressure condition curve, a humidity condition curve, and a temperature condition curve. The coordinate values ​​of each environmental condition curve at each time point are determined to match the threshold ranges of the suitable storage environment data. If the coordinate values ​​of each environmental condition curve at each time point match the threshold ranges of the suitable storage environment data, the current storage environment of the fresh produce is determined to match the corresponding suitable storage environment data based on the output of the fresh produce identification model. If the coordinate values ​​of each environmental condition curve at each time point do not match the threshold ranges of the suitable storage environment data, the current storage environment of the fresh produce is determined to be mismatched with the corresponding suitable storage environment data.

[0012] In one implementation of this application, if the output result indicates that the fresh produce has surface defects, it is determined whether the surface defects constitute an inedible condition. If the surface defects are inedible, it is determined that the current storage environment of the fresh produce does not match the corresponding suitable storage environment data. If the surface defects are not inedible, a defect warning message is generated and sent to the management terminal corresponding to the current storage environment, so that the management terminal can acquire images of the fresh produce in real time at preset time intervals.

[0013] In one implementation of this application, the system receives location information from an active identification carrier and determines the current environmental control location of the storage environment using an electronic map. A preset number of management terminals are determined, centered on the current environmental control location and with a preset distance as the radius. Control information is sent to each management terminal to determine if the operator at each terminal is available. "Available" means the operator arrives at the current environmental control location within a preset time. At least one management terminal is selected from the available terminals of the operators as the control management terminal, and environmental control information is sent to the control management terminal.

[0014] In one implementation of this application, operational information from each management terminal is obtained. This operational information includes at least current location information, current control completion time, and planned control information. Based on the operational information, it is determined whether operators can reach the environmental control location of the current storage environment within a preset time, thereby determining whether the operators at each management terminal are available.

[0015] In one implementation of this application, the arrival time of the operator at the management terminal to the environmental control location is calculated based on the current location information of the management terminal, the current control completion time, and the planned control information. The corresponding target control time is determined based on the transmission time of the control information and a preset time. Based on the arrival times and the target control time, a control management terminal is selected from at least one management terminal, and environmental control information is sent to the control management terminal. The operator at the control management terminal must at least meet the target control time.

[0016] Through the above-described solution, this application can regulate the storage environment of fresh produce based on surface defects and current storage environment data, thereby ensuring the stability of the storage environment, guaranteeing the quality of fresh produce, providing consumers with safe and reliable fresh produce, and improving consumers' fresh produce purchasing experience. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a schematic diagram of a fresh food storage control method based on the Industrial Internet in an embodiment of this application;

[0019] Figure 2 This is another flowchart illustrating a fresh food storage control method based on the Industrial Internet, as described in this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] Fresh produce is typically stored at room temperature or low temperature during transportation and storage. If the environment is unsuitable, it can easily become damaged or spoil. Furthermore, some fresh produce cannot withstand pressure from heavy objects; such pressure can cause damage and lead to oxidation and spoilage in the air. Therefore, the storage of fresh produce requires extreme care, with constant monitoring of the environment to ensure its stability.

[0022] Having dedicated personnel monitor the environment of fresh produce is a waste of manpower, and the environment may be at a low temperature, making it unsuitable for humans to stay for long.

[0023] Based on this, this application provides a fresh food storage control method based on the Industrial Internet to ensure the stability of the storage environment for fresh products and improve the fresh food purchasing experience for consumers.

[0024] The various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0025] This application provides a method for fresh food storage control based on the Industrial Internet, such as... Figure 1 As shown, the method may include steps S101-S105:

[0026] S101, the server obtains fresh food storage information.

[0027] In this embodiment, the fresh produce storage information can be the name and image of the fresh produce collected by the user through a handheld terminal or desktop computer during storage. The terminal that collects the fresh produce storage information can send the information to the server based on the user's actions, or the server can automatically retrieve the information from the terminal at fixed intervals. The server in this application can be a single server or a server cluster; this application does not specifically limit its scope.

[0028] It should be noted that the server, as the executing entity of the fresh food storage control method based on the Industrial Internet, exists only as an example, and the executing entity is not limited to the server. This application does not make any specific limitation in this regard.

[0029] S102, the server sends the fresh food storage information to the environmental condition model to obtain the suitable environmental data for the storage of the corresponding fresh food.

[0030] The data on suitable storage environments include: pressure conditions, temperature conditions, and humidity conditions.

[0031] In this embodiment, the server sends fresh produce storage information to an environmental condition model to obtain suitable storage environment data for the corresponding fresh produce, specifically including:

[0032] First, the server uses an environmental condition model to determine the type of fresh produce in the fresh produce storage information.

[0033] The fresh produce storage information includes at least one or more of the following: fresh produce image, fresh produce name. The fresh produce type is determined based on the fresh produce's storage conditions.

[0034] The environmental condition model can be a neural network model or other image recognition models. If the environmental condition model is a neural network model, then when the server inputs an image or name of fresh produce containing storage information into the neural network model, the model will output the fresh produce type. If the environmental condition model is an image recognition model, the model obtains the fresh produce type by recognizing the fresh produce features in the image. The fresh produce type can be obtained based on the fresh produce's storage conditions. Specifically, fresh produce storage conditions can include room temperature storage, -1 degree Celsius storage, -18 degrees Celsius storage, etc. Specific storage conditions can be set during actual use, and this application does not impose specific limitations on them.

[0035] Due to the diversity of storage conditions for fresh produce, there are also different types of fresh produce. In order to facilitate the differentiation of storage environments for different types of fresh produce, we can ensure the proper storage of fresh produce by determining its type.

[0036] Then, the server determines whether there are any fresh produce items among the fresh produce types that do not meet the corresponding storage conditions.

[0037] In this embodiment of the application, the storage conditions of the fresh food types are mismatched, that is, the storage conditions are inconsistent in each fresh food type. For example, fresh food A is stored at room temperature and fresh food B is stored at -1 degree Celsius. The server can determine that the two fresh food types are mismatched.

[0038] In the absence of fresh produce with mismatched storage conditions, the server uses an environmental condition model to determine suitable storage environment data for the fresh produce.

[0039] When the storage conditions of all fresh produce in the fresh produce storage information are consistent, the server can output suitable storage environment data for fresh produce through an environmental condition model. This suitable storage environment data includes at least: pressure conditions, temperature conditions, humidity conditions, and gas filling conditions.

[0040] When there are fresh produce items whose storage conditions do not match, the server generates fresh produce storage prompts and determines the suitable storage environment data for each type of fresh produce output by the environmental condition model.

[0041] Among them, the fresh food storage reminder information is used to remind operators of fresh food storage to classify and store fresh food.

[0042] When the server determines that some fresh produce has inconsistent storage conditions among its storage information, it can generate a storage warning message, such as: "Fresh produce A and fresh produce B have inconsistent storage conditions; please take note." Simultaneously, the server generates suitable storage environment data for each fresh produce. This storage warning message and the corresponding suitable storage environment data are then sent to the handheld terminal of the fresh produce storage operator, enabling the operator to adjust the storage environment accordingly.

[0043] S103, the server uses a fresh produce identification model to identify images of fresh produce in order to determine whether the fresh produce has surface defects.

[0044] The fresh produce identification model was trained using images of several different types of fresh produce samples.

[0045] In this embodiment, the image of fresh produce can be obtained by an image acquisition device installed in a warehouse, a storage box for fresh produce, or the transport compartment of a transport vehicle. The image acquisition device can be, for example, a camera. The image of fresh produce can also be an image of fresh produce from fresh produce storage information. Through the fresh produce identification model, physical and physiological defects on the surface of the fresh produce can be determined. For example, physical defects include at least: slight deformation caused by surface compression and surface scratches. Physiological defects include at least: traces of bacterial erosion and decay.

[0046] In one embodiment of this application, when the server identifies images of fresh produce using a fresh produce identification model to determine whether the fresh produce has surface defects, it needs to perform model training and identification of whether the fresh produce is alive, including the following steps:

[0047] S201, the server inputs several different types of fresh produce sample images into the fresh produce recognition model.

[0048] Images of fresh produce samples include information on surface defects. Images of fresh produce samples must include at least: seafood products.

[0049] The fresh produce sample images can be obtained by crawling from the internet. These images contain surface defect information, including both physical and physiological defects. The fresh produce identification model can be a convolutional neural network model.

[0050] S202, the server uses a fresh produce identification model to extract surface and contour features from the images of each fresh produce sample for model training.

[0051] The fresh produce identification model can acquire the surface and contour features of fresh produce in the images of each fresh produce sample, identify fresh produce and the features of defects in fresh produce, and continuously train the model.

[0052] S203, the server obtains the trained fresh food identification model when the loss function value of the fresh food identification model is less than a preset threshold.

[0053] The server determines that the fresh produce identification model has completed training when the loss function value obtained by the model is less than a preset threshold. Loss functions can include 0-1 loss, absolute value loss, log-log loss, etc. The specific loss function used can be selected during actual use.

[0054] S204, The server determines the name of the seafood product.

[0055] After the server completes the fresh produce identification model, it can also obtain the fresh produce name from the fresh produce storage information.

[0056] S205, the server matches the seafood product names with the pre-stored fresh food name table.

[0057] The list of pre-stored fresh food names must include at least: hairy crab, king crab, and lobster.

[0058] The server will match the obtained fresh produce names with a pre-stored fresh produce name table, which may contain fresh produce whose life status needs to be monitored, or fresh produce that needs to undergo other treatments during storage.

[0059] S206, when the fresh produce name matches the pre-stored fresh produce name table, the server extracts the surface features and contour features of each fresh produce sample image through the fresh produce recognition model corresponding to the matched fresh produce name, and then sends the corresponding images of the surface features and contour features to the preset comparison image database to determine the survival information of the fresh produce.

[0060] For example, if the seafood is named "hairy crab," it's crucial to ensure the crabs exhibit vital signs during storage (transportation). The server can obtain a corresponding seafood identification model based on the product's name, such as a model for identifying the survival of hairy crabs, or another model for identifying lobsters. This model extracts surface and contour features from the seafood sample image and sends them to a database for identifying seafood survival. The database then compares these features with feature images of several hairy crabs exhibiting vital signs. By comparing the feature images with the surface and contour features, the survival information of the seafood can be obtained. For instance, the leg hairs on a hairy crab can be used to identify its survival; by comparing the state of the leg hairs, the freshness of the crab can be determined.

[0061] S207, the server uses a fresh produce identification model to determine the quantity of fresh produce corresponding to the fresh produce storage information if the fresh produce has surface defects.

[0062] After the server can identify surface defects on the surface of fresh produce using a fresh produce identification model, it can first determine the quantity of fresh produce from the fresh produce storage information, which is the total number of fresh produce, such as 10 hairy crabs.

[0063] S208, the server determines whether the ratio of the number of defective fresh produce with surface defects to the total number of fresh produce is greater than a preset value.

[0064] The server can determine the quantity of fresh produce with surface defects and calculate the ratio using the following method:

[0065]

[0066] Where T is the ratio of the number of defective fresh produce with surface defects to the total number of fresh produce, s is the number of defective fresh produce, and S is the total number of fresh produce. The preset value can be set during actual use, for example, to 0.4 or 0.3. By setting the preset value, the probability of fresh produce spoilage during storage can be minimized.

[0067] S209, if the ratio of the number of defective fresh produce with surface defects to the total number of fresh produce is not greater than a preset value, the server sends the defect image of the surface defects to the expert terminal to determine whether the surface defects are inedible defects.

[0068] The server can allow expert terminals to identify the type of surface defects when the ratio is not greater than a preset value. The expert terminal can be one of the nodes in a pre-built blockchain network, ensuring the authenticity, reliability, and traceability of the surface defect information certified by the expert terminal. The expert terminal can either machine-identify whether a surface defect image is inedible, or send the defect image to several user terminals for scoring. If the score is lower than a preset value, the expert terminal determines that the surface defect in the image is inedible.

[0069] In addition, if the score is successfully recognized by the expert terminal after the user terminal completes the scoring of the surface defect image, the user terminal will receive corresponding rewards, such as bonus points or coupons.

[0070] S104, the server determines whether the current storage environment of the fresh produce matches the corresponding storage environment data of the fresh produce based on the data on suitable storage environment and the output of the fresh produce identification model.

[0071] The current storage environment is obtained through active labeling carriers set up in the storage environment of fresh produce.

[0072] In this embodiment of the application, based on the data on suitable storage environments and the output results of the fresh produce identification model, it is determined whether the current storage environment of the fresh produce matches the corresponding suitable storage environment data, specifically including:

[0073] First, the server receives the identifiers from each active identifier carrier and the environmental data of the current storage environment.

[0074] In this embodiment, the fresh food storage environment, such as a storage warehouse, transport vehicle, or fresh food storage box, is equipped with an active identification carrier. This active identification carrier has an identifier that can be sent to secondary nodes, and it can acquire data from various sensors in the current storage environment, such as temperature sensors, humidity sensors, gas sensors, and pressure sensors. The active identification carrier sends the identifier and the environmental data collected by the sensors to a server. The active identification carrier can send environmental data to the server in real time, or it can send environmental data for a certain period of time to the server at intervals.

[0075] Then, based on the environmental data, the server generates corresponding environmental condition curves for each identifier within a preset time period.

[0076] Among them, the environmental condition curves include at least: pressure condition curve, humidity condition curve, and temperature condition curve.

[0077] After receiving environmental data, the server can generate environmental condition curves for a preset time period based on the identifiers. For example, identifier x represents the pressure, humidity, and temperature condition curves, while identifier o represents the temperature condition curve. The horizontal axis of the environmental condition curves represents time, and the vertical axis represents the corresponding curve values, such as pressure, humidity, and temperature values.

[0078] Next, the server determines whether the coordinate values ​​of each environmental condition curve at each time point match the threshold ranges of each condition for storing suitable environmental data.

[0079] The server can match the environmental condition curve values ​​at various time points with various condition threshold intervals. In this embodiment, each environmental condition curve has a condition threshold interval for storing suitable environmental data. For example, the pressure condition curve for fresh produce A has a condition threshold interval of [m1, n1], the temperature condition curve has a condition threshold interval of [m2, n2], and the pressure condition curve for fresh produce B has a condition threshold interval of [m3, n3]. These condition threshold intervals vary for different fresh produce or different regions, and the specific condition threshold intervals need to be adjusted during actual use.

[0080] Once the coordinate values ​​of each environmental condition curve at each time point are determined and matched with the threshold ranges of each condition in the data on suitable storage environment, the server determines whether the current storage environment of the fresh produce matches the corresponding data on suitable storage environment based on the output of the fresh produce identification model.

[0081] In this embodiment, after the server determines that the environmental condition curve data matches the suitable storage environment data, the server will also determine the output result of the fresh food identification model, that is, determine whether the fresh food surface identified by the fresh food identification model is free of surface defects, or whether the surface defects are edible. Furthermore, it determines that the current storage environment matches the suitable storage environment data.

[0082] If the coordinate values ​​of each environmental condition curve at each time point do not match the threshold ranges of each condition in the data on suitable storage environment, the server determines that the current storage environment of the fresh produce does not match the corresponding data on suitable storage environment for the fresh produce.

[0083] In one embodiment of this application, the server determines whether the current storage environment of the fresh produce matches the corresponding suitable storage environment data based on the output of the fresh produce identification model, specifically including:

[0084] When the server outputs that the fresh produce has surface defects, it determines whether the surface defects are inedible defects.

[0085] When the surface defect is classified as an inedible defect, the server determines that the current storage environment of the fresh produce does not match the corresponding suitable storage environment data.

[0086] When the surface defect is edible, the server generates a defect warning message and sends it to the corresponding management terminal in the current storage environment, so that the management terminal can acquire images of the fresh produce in real time at preset time intervals.

[0087] When the surface defect is an edible defect, a defect alert message can be sent to the management terminal so that the personnel at the management terminal can pay attention to the fresh food with the edible defect and prevent the surface defect from developing into an inedible defect.

[0088] S105, when the current storage environment of fresh produce does not match the corresponding suitable storage environment data, the server generates environmental control information.

[0089] In this embodiment of the application, when the server generates environmental control information in the event that the current storage environment of the fresh produce does not match the corresponding suitable storage environment data for the fresh produce, the following steps are included:

[0090] First, the server receives the identification and positioning information sent by the active identification carrier and determines the environmental control location of the current storage environment through an electronic map.

[0091] The active identification carrier can send location information to the server. The server can locate the active identification carrier's position through an electronic map and obtain the terminal position of the active identification carrier for environmental control. This terminal position is the environmental control position.

[0092] Then, the server uses the current storage environment's environmental control location as the center and a preset distance as the radius to determine a preset number of management terminals.

[0093] Next, the server sends control information to each management terminal to determine whether the operators of each management terminal are available.

[0094] Idle time refers to the time within which operators can reach the environmental control position of the current storage environment.

[0095] Specifically, the server sends control information to each management terminal to determine whether the operators at each management terminal are available, including:

[0096] The server can obtain job information from each management terminal.

[0097] The operational information includes at least the current location information, the current control completion time, and the planned control information.

[0098] The management terminal can be a terminal that is currently regulating other storage environments, or a terminal that is not regulating other environments.

[0099] Based on the job information, the server determines whether the operator can reach the environmental control location of the current storage environment within a preset time, in order to determine whether the operators of each management terminal are available.

[0100] The server will be able to reach the terminal at the current storage environment location within a preset time and designate it as an idle terminal.

[0101] Finally, the server selects at least one management terminal from the idle management terminals of each operator as the control management terminal and sends environmental control information to the control management terminal.

[0102] Specifically, the server selects at least one management terminal from among the idle management terminals of each operator as the control management terminal, and sends environmental control information to the control management terminal, including:

[0103] The server calculates the arrival time of the operator at the environmental control location based on the current location information of the management terminal, the current control completion time, and the planned control information.

[0104] The server determines the corresponding target control time based on the sending time of the control information and the preset time.

[0105] Generally, the server can calculate the destination adjustment time T3 = T1 + T2 from the sending time (e.g., T1) and the preset time T2. In actual use, the destination adjustment time can also be obtained in other ways.

[0106] Based on the arrival time and destination control time, the server determines the control management terminal from at least one management terminal and sends environmental control information to the control management terminal.

[0107] Operators of the control and management terminal must at least meet the target control time.

[0108] In this embodiment of the application, the control and management terminal selected by the server needs to arrive at the environmental control location within the target control time.

[0109] Environmental control information can include data on suitable storage environments for fresh produce, enabling operators at the control and management terminal to adjust the current storage environment data to meet the requirements of suitable storage conditions.

[0110] In addition, to make it easier to calculate the time to reach the environmental control location, the moving speed of the management terminal can be obtained, thereby accurately calculating the arrival time.

[0111] This application, through the above-mentioned solution, can collect fresh food environment data through the Industrial Internet and adjust the storage environment based on the fresh food environment data, thereby ensuring the stability of the storage environment for fresh products. Providing a stable storage environment for fresh products can guarantee their freshness, taste, and appearance, and improve the fresh food purchasing experience for consumers.

[0112] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0113] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0114] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for fresh food storage control based on the Industrial Internet, characterized in that, The method includes: Obtain information on fresh food storage; The fresh food storage information is sent to an environmental condition model to obtain suitable storage environment data for the corresponding fresh food; wherein, the suitable storage environment data includes: pressure conditions, temperature conditions, and humidity conditions; The fresh produce identification model is used to identify images of fresh produce in order to determine whether the fresh produce has surface defects; wherein, the fresh produce identification model is trained using images of several different types of fresh produce samples. Based on the storage suitability environment data and the output of the fresh produce identification model, it is determined whether the current storage environment of the fresh produce matches the corresponding storage suitability environment data; wherein, the current storage environment is obtained through an active identification carrier set in the storage environment of the fresh produce; When the current storage environment of the fresh produce does not match the corresponding suitable storage environment data for the fresh produce, environmental control information is generated. Before identifying the image of the fresh produce using a fresh produce identification model to determine whether the fresh produce has surface defects, the method further includes: Several different types of fresh produce sample images are input into the fresh produce recognition model; the fresh produce sample images include surface defect information; the fresh produce sample images include at least: seafood products; The surface and contour features of each fresh produce sample image are extracted using the fresh produce identification model for model training. If the loss function value of the fresh produce identification model is less than a preset threshold, the trained fresh produce identification model is obtained. The method further includes, after identifying the image of the fresh produce using a fresh produce identification model to determine whether the fresh produce has surface defects: Using the fresh produce identification model, if the fresh produce has surface defects, the corresponding quantity of fresh produce is determined based on the fresh produce storage information. Determine whether the ratio of the number of defective fresh produce with the surface defects to the total number of fresh produce is greater than a preset value; If not, the defect image of the surface defect is sent to an expert terminal to determine whether the surface defect is an inedible defect.

2. The method according to claim 1, characterized in that, The fresh food storage information is sent to an environmental condition model to obtain suitable environmental data for the corresponding fresh food storage, specifically including: The environmental condition model is used to determine the type of fresh produce in the fresh produce storage information; wherein, the fresh produce storage information includes at least one or more of the following: fresh produce image, fresh produce name; the fresh produce type is obtained based on the fresh produce storage conditions; and Determine whether any of the fresh produce types have corresponding storage conditions that do not match; In the absence of fresh produce with mismatched storage conditions, the suitable storage environment data for the fresh produce is determined using the environmental condition model. In the event that the fresh produce has mismatched storage conditions, fresh produce storage prompt information is generated, and the suitable storage environment data for each type of fresh produce output by the environmental condition model is determined; wherein, the fresh produce storage prompt information is used to prompt the fresh produce storage operator to classify and store the fresh produce.

3. The method according to claim 1, characterized in that, The method further includes: Determine the name of the seafood product; The seafood product names are matched with a pre-stored list of fresh seafood names; the pre-stored list of fresh seafood names includes at least: hairy crab, king crab, and lobster; If the seafood product name matches one of the items in the pre-stored fresh food name table, after extracting the surface features and contour features of each fresh food sample image using the fresh food identification model corresponding to the matched fresh food name, the images corresponding to the surface features and contour features are sent to a preset comparison image database to determine the survival information of the fresh food.

4. The method according to claim 1, characterized in that, Based on the storage suitability environment data and the output of the fresh produce identification model, it is determined whether the current storage environment of the fresh produce matches the corresponding storage suitability environment data, specifically including: Receive identifiers from each of the active identifier carriers and environmental data of the current storage environment; Based on the environmental data, corresponding environmental condition curves are generated for each of the aforementioned identifiers within a preset time period; wherein, the environmental condition curves include at least: pressure condition curve, humidity condition curve, and temperature condition curve; Determine whether the coordinate values ​​of each environmental condition curve at each time point match the threshold ranges of each condition for storing suitable environmental data. If a match is found, the current storage environment of the fresh produce is determined to match the corresponding suitable storage environment data of the fresh produce, based on the output of the fresh produce identification model. If they do not match, it is determined that the current storage environment of the fresh produce does not match the corresponding suitable storage environment data for the fresh produce.

5. The method according to claim 4, characterized in that, Based on the output of the fresh produce identification model, determine whether the current storage environment of the fresh produce matches the corresponding suitable storage environment data, specifically including: If the output result indicates that the fresh produce has surface defects, determine whether the surface defects are inedible defects. If so, it is determined that the current storage environment of the fresh produce does not match the corresponding suitable storage environment data for the fresh produce; If not, a defect warning message is generated and sent to the corresponding management terminal of the current storage environment, so that the management terminal can acquire images of the fresh produce in real time at preset time intervals.

6. The method according to claim 1, characterized in that, When the current storage environment of the fresh produce does not match the corresponding suitable storage environment data for the fresh produce, after generating environmental control information, the method further includes: Receive the identification and positioning information sent by the active identification carrier, and determine the environmental control location of the current storage environment through an electronic map; Using the environmental control location of the current storage environment as the center and a preset distance as the radius, a preset number of management terminals are determined; Control information is sent to each of the management terminals to determine whether the operators of each management terminal are idle; wherein, idle means that the operators arrive at the environmental control position of the current storage environment within a preset time. Among the management terminals of the idle operators, at least one management terminal is selected as the control management terminal, and the environmental control information is sent to the control management terminal.

7. The method according to claim 6, characterized in that, Sending control information to each of the management terminals to determine whether the operators of each management terminal are available, specifically including: Obtain the operation information of each of the management terminals; wherein, the operation information includes at least the current location information, the current control completion time, and the planned control information; Based on the work information, it is determined whether the operator can reach the environmental control location of the current storage environment within the preset time, so as to determine whether the operators of each management terminal are idle.

8. The method according to claim 7, characterized in that, Among the management terminals of the idle operators, at least one management terminal is selected as the control management terminal, and the environmental control information is sent to the control management terminal, specifically including: Based on the current location information of the management terminal, the current control completion time, and the planned control information, calculate the arrival time of the operator of the management terminal to the environmental control location; The corresponding target control time is determined based on the sending time of the control information and the preset time. Based on the arrival time and the target control time, the control management terminal is determined from the at least one management terminal, and the environmental control information is sent to the control management terminal; the operator of the control management terminal must at least meet the target control time.

Images