A fully automatic intelligent fresh-keeping refrigerator system and its operation method

Abstract

The present invention provides a fully automatic intelligent fresh-keeping refrigerator system and an operation method, which relates to the technical field of smart home. The system includes a high-resolution camera, an image processing chip and an edge computing device module, an RFID tag and an RFID reader module, a cloud server and an interaction interface. The present invention combines a deep learning model with Internet of Things technology to achieve real-time monitoring and accurate prediction of the freshness of food in the refrigerator. Its advantages are that it can automatically identify the state of food, timely warn of expired food, effectively reduce food waste, and continuously optimize the prediction accuracy through adaptive learning, provide personalized diet suggestions and health management for users, and greatly improve the convenience and health of smart home life.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home, and particularly to a fully automatic intelligent fresh-keeping refrigerator system and an operation method thereof. Background Art

[0002] As an important branch of artificial intelligence, computer vision technology simulates and enhances human visual capabilities through digital image processing and deep learning models, covering image acquisition, preprocessing, and advanced image analysis and understanding, showing great application potential in fields such as smart home. The embedded system, through the deep integration of network technology and communication technology, has improved the communication intelligence and flexibility of devices. Especially in the field of image processing, it provides powerful technical support for smart home devices.

[0003] However, existing smart refrigerators still have deficiencies in meeting the specific needs of the elderly, such as food management problems caused by memory decline, cognitive lag, and saving habits, which require further attention and improvement. Summary of the Invention

[0004] For this reason, the embodiments of the present invention provide a fully automatic intelligent fresh-keeping refrigerator system and an operation method thereof, which are used to solve the problem that existing smart refrigerators in the prior art are difficult to meet specific healthy diet needs.

[0005] To solve the above problems, the embodiments of the present invention provide a fully automatic intelligent fresh-keeping refrigerator system, which includes:

[0006] A high-resolution camera for real-time acquisition of image information of foods in the refrigerator;

[0007] An image processing chip and an edge computing device module, connected to the high-resolution camera, including an image processing chip and an edge computing device, for preprocessing the acquired image information;

[0008] An RFID tag and an RFID reader module, connected to the image processing chip and the edge computing device module, the RFID tag and the RFID reader module including an RFID tag and an RFID reader, for embedding the RFID tag into each dinner plate and reading the RFID tag information through the RFID reader;

[0009] A cloud server, connected to the RFID tag and the RFID reader module, including a deep learning module, a database management system, an image recognition module, and an anomaly detection and alarm system, for receiving the RFID tag information read by the RFID reader, and using the read RFID tag information for adaptive learning, database matching algorithm processing, and generating feedback information, including food freshness prediction results, personalized diet suggestions, and alarm information;

[0010] The interaction interface, including the refrigerator touch screen, voice prompt module and mobile phone APP, is used to receive the feedback information generated by the cloud server and provide real-time monitoring and alarm reminder services to users;

[0011] Among them, the deep learning module includes a cuisine prediction and recommendation system, and the cuisine prediction and recommendation system adopts a strategy of classifying by cuisine for deep learning training, specifically including:

[0012] Define the main function, and the input parameters include prediction results, initial confidence, user feedback, seasonal factors, user preferences and market trends;

[0013] Find the index with the highest probability from the prediction results, obtain the corresponding cuisine from the cuisine list according to this index, and use the highest probability value as the confidence score;

[0014] Create a feedback dictionary containing the predicted cuisine and the confidence score;

[0015] Update the confidence according to user feedback;

[0016] Traverse the cuisine list, calculate the weight for each cuisine, and this weight is based on seasonal factors, user preferences and market trends, and adjust the updated confidence according to the weight;

[0017] If there are cooking methods related to non-cuisine in the user feedback, make adjustments, that is, adjust the updated confidence according to the non-cuisine feedback; otherwise, execute the next step;

[0018] Calculate the freshness score of the dish according to the updated confidence, user preferences, seasonal factors and market trends;

[0019] Add the updated confidence and freshness score to the feedback dictionary and return the feedback dictionary.

[0020] Preferably, the image processing chip and the edge computing device module include an image processing chip and an edge computing device. The image processing chip includes filtering, denoising and feature extraction modules for preprocessing the image; the edge computing device includes a data compression and encryption module for compressing the data before transmission to reduce the data volume and encrypting it to ensure the security of data transmission.

[0021] Preferably, the updating of the confidence according to user feedback specifically includes:

[0022] Obtain the index of the predicted cuisine in the cuisine list;

[0023] Obtain the actual cuisine in the user feedback and its index in the cuisine list;

[0024] Adjust the initial confidence according to whether the prediction is accurate:

[0025] If the prediction is correct, increase the confidence level of the predicted cuisine type;

[0026] If the prediction is incorrect, decrease the confidence level of the predicted cuisine type and increase the confidence level of the actual cuisine type;

[0027] Normalize the updated confidence levels.

[0028] Preferably, the updated confidence level is expressed as:

[0029] ;

[0030] In the formula, is the updated confidence level, is the initial confidence level, is the weight, is the adjustment factor.

[0031] Preferably, the freshness score of the dish is expressed as:

[0032] ;

[0033] In the formula, is the freshness score, is the set of all cuisine types, is the updated confidence level, is the weight of user preference, is the weight of seasonal factors, is the weight of market trends.

[0034] Preferably, the anomaly detection and alarm system includes a preset threshold judgment module for comparing the real-time monitored data with the preset threshold and triggering an alarm once the preset threshold is exceeded.

[0035] Preferably, the refrigerator touch screen is used to display the real-time food freshness monitoring results, personalized dietary suggestions, and alarm information, and supports users to perform interactive operations through the touch screen.

[0036] Preferably, the voice prompt module can, according to the instructions of the cloud server, broadcast the food freshness status, personalized dietary suggestions, and alarm information to the user by voice.

[0037] Preferably, the mobile APP can remotely receive the food freshness data and alarm information sent by the cloud server and supports users to perform remote control and parameter settings through the mobile APP.

[0038] The embodiment of the present invention also provides an operation method for a full-automatic intelligent fresh-keeping refrigerator system. This method uses the above-mentioned full-automatic intelligent fresh-keeping refrigerator system and specifically includes:

[0039] Use a high-resolution camera to collect image information of the food in the refrigerator in real time;

[0040] Use an image processing chip and an edge computing device module to preprocess the collected image information;

[0041] Use RFID tags and an RFID reader module to embed RFID tags into each dinner plate, and read the RFID tag information through the RFID reader;

[0042] Use a cloud server to receive the RFID tag information read by the RFID reader, and use the read RFID tag information for adaptive learning, database matching algorithm processing, and generate feedback information, including food freshness prediction results, personalized diet suggestions, and alarm information;

[0043] Receive the feedback information generated by the cloud server through an interaction interface and provide real-time monitoring and alarm reminder services to users.

[0044] It can be seen from the above technical solutions that the present invention application has the following beneficial effects:

[0045] (1) Real-time monitoring and accurate prediction: The present invention integrates a high-resolution camera, image processing technology, RFID identification, and cloud deep learning, realizing instant monitoring of the food status in the refrigerator and accurate prediction of freshness, effectively reducing food waste and promoting healthy eating.

[0046] (2) Multimodal data fusion and intelligent decision-making: The present invention improves the comprehensiveness and accuracy of data processing through the fusion processing of multimodal information such as images and RFID tag data, combined with cloud intelligent analysis. At the same time, the adaptive learning mechanism continuously optimizes the model, enhancing the intelligent decision-making ability and prediction accuracy of the system.

[0047] (3) Continuous self-optimization ability: The system of the present invention can automatically verify the recognition accuracy and adjust the model parameters for retraining when necessary to cope with changes in different environments and food states. This self-optimization mechanism ensures the stability and adaptability of the system during long-term operation.

[0048] (4) Personalized user experience and service: The present invention provides multiple interaction methods such as touch screens, voice prompts, and mobile phone APPs, facilitating users to view the refrigerator status in real time and receive alarm reminders. At the same time, based on the user's eating habits and feedback, customized diet suggestions and services are provided to enhance the user experience and improve the product competitiveness.

[0049] (5)Deep learning optimization driven by cuisine recognition: The present invention adopts a scenario-based training method classified by cuisine, reducing the internal data variability and enabling the model to focus more on the feature learning of specific cuisines. This strategy significantly improves the accuracy of the model in dish recognition and prediction, and enhances the effectiveness and value of the model in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following briefly describes the drawings required in the embodiments. By referring to the drawings, the features and advantages of the present invention will be more clearly understood. The drawings are schematic and should not be construed as limiting the present invention in any way. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Among them:

[0051] Figure 1 It is a structural block diagram of a full-automatic intelligent fresh-keeping refrigerator system provided in the embodiment;

[0052] Figure 2 It is a flowchart of an operation method of a full-automatic intelligent fresh-keeping refrigerator system provided in the embodiment; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following clearly and completely describes the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention. Embodiment 1

[0054] To solve the problem that existing intelligent refrigerators in the prior art are difficult to meet specific healthy diet requirements. As Figure 1 shown, the embodiment of the present invention proposes a full-automatic intelligent fresh-keeping refrigerator system, which includes:

[0055] A high-resolution camera for real-time acquisition of image information of foods in the refrigerator;

[0056] An image processing chip and an edge computing device module, connected to the high-resolution camera, including an image processing chip and an edge computing device, for preprocessing the acquired image information;

[0057] The RFID tag and RFID reader module are connected to the image processing chip and the edge computing device module. The RFID tag and RFID reader module includes an RFID tag and an RFID reader, which are used to embed the RFID tag into each dinner plate and read the RFID tag information through the RFID reader.

[0058] The cloud server, connected to the RFID tag and RFID reader module, includes a deep learning module, a database management system, an image recognition module, and an anomaly detection and alarm system. It is used to receive the RFID tag information read by the RFID reader, perform adaptive learning, database matching algorithm processing using the read RFID tag information, and generate feedback information, including food freshness prediction results, personalized diet suggestions, and alarm information.

[0059] The interaction interface includes a refrigerator touch screen, a voice prompt module, and a mobile phone APP, which are used to receive the feedback information generated by the cloud server and provide real-time monitoring and alarm reminder services to the user.

[0060] From the above technical solutions, the present invention proposes a fully automatic intelligent fresh-keeping refrigerator system. By integrating advanced devices such as a high-resolution camera, an image processing chip, an edge computing device module, an RFID tag and an RFID reader module, and a cloud server, and combining a deep learning model with Internet of Things technology, it realizes real-time monitoring and accurate prediction of the freshness of food in the refrigerator. Its advantages are that it can automatically identify the food status, timely warn of expired food, effectively reduce food waste, and continuously optimize the prediction accuracy through adaptive learning, providing personalized diet suggestions and health management for users, greatly improving the convenience and health of smart home life.

[0061] This embodiment details how to construct a fully automatic intelligent fresh-keeping refrigerator system, which integrates advanced computer vision technology, Internet of Things (IoT) technology, and artificial intelligence algorithms, aiming to provide users with an intelligent food management experience.

[0062] 1. Hardware module integration:

[0063] High-resolution camera: Install a high-resolution (such as 4K) camera at the top inside the refrigerator to ensure that it can clearly capture the image information of each layer of food in the refrigerator, such as color, shape, and texture. The camera is connected to the image processing chip through a USB interface.

[0064] Image processing chip: Select a dedicated image processing chip integrated with filtering, denoising, and feature extraction modules, and install it in the control board on the back of the refrigerator. This chip is responsible for receiving the image data transmitted by the camera and performing preliminary image preprocessing, including removing noise, enhancing contrast, and extracting key features (such as color, shape, etc.).

[0065] Edge computing device: Deploy an edge computing server inside the refrigerator, including a data compression and encryption module, which is used to compress the data before transmission to reduce the data volume and encrypt it to ensure the security of data transmission.

[0066] RFID tag and RFID reader module: Embed RFID tags into each dinner plate to achieve fast identification under the scanning of the RFID reader. In addition, the RFID tag and RFID reader module also includes a Bluetooth module, which is used to transmit the RFID tag information read by the RFID reader to the cloud server.

[0067] In the smart refrigerator system, the integrated RFID tags and high-resolution cameras work closely together to achieve instant and accurate monitoring of the food status in the refrigerator. The RFID tags quickly identify the items on each dinner plate through the built-in RFID reader in the refrigerator, while the high-definition camera captures the appearance details of the food, such as color, shape, and texture, and analyzes them using advanced image processing algorithms. This process not only quickly determines the type, storage time, and current status of the food, but also constructs a detailed food information database.

[0068] 2. Software modules and cloud services:

[0069] Cloud server: Deploy a high-performance server in the cloud, including a deep learning module, a database management system, an image recognition module, and an anomaly detection and alarm system. The server receives the RFID tag information read by the RFID reader through the Bluetooth module, and uses the read RFID tag information for adaptive learning, database matching algorithm processing, and generates feedback information, including food freshness prediction results, personalized diet suggestions, and alarm information.

[0070] Deep learning model: Use a large amount of food images and freshness data to train the deep learning model so that it can accurately identify food types and predict their freshness. The model regularly learns from user feedback, continuously optimizes the parameters, and improves the prediction accuracy.

[0071] Anomaly detection and alarm system: The system has preset thresholds built in, such as the remaining days of food shelf life, temperature range, etc. Once the real-time monitored data exceeds the threshold, the system will automatically trigger the alarm mechanism.

[0072] Specifically, the system uses the deep learning model to continuously optimize its recognition ability. By regularly comparing and verifying the newly collected data with the previous recognition results, the system can evaluate the recognition accuracy in real time, and when it finds that the accuracy is lower than the preset standard, it will automatically adjust the model parameters for retraining. This adaptive learning mechanism ensures that the system can flexibly adapt to changes in different environments and food statuses, thus maintaining a high level of recognition accuracy.

[0073] To further improve the recognition effect and application value of deep learning models, the present invention adopts a strategy of classifying by cuisine to train deep learning models. This method effectively reduces the differences between data samples, enabling the model to focus on the unique flavor characteristics and cooking techniques of each cuisine. Specifically, by carefully classifying mainstream cuisines such as Sichuan cuisine, Cantonese cuisine, and Shandong cuisine, and collecting and annotating rich image data, the deep learning model can deeply learn and reflect the characteristics of each cuisine during the training process, thereby improving the accuracy of dish recognition and prediction. The following is a specific description of the algorithm:

[0074] S1: Define the main function, and the input parameters include prediction results, initial confidence, user feedback, seasonal factors, user preferences, and market trends.

[0075] S2: Find the index with the highest probability from the prediction results, and obtain the corresponding cuisine from the cuisine list according to this index, and use the highest probability value as the confidence score.

[0076] S3: Create a feedback dictionary containing the predicted cuisine and the confidence score.

[0077] S4: Update the confidence according to user feedback, including the following steps:

[0078] S41: Obtain the index of the predicted cuisine in the cuisine list;

[0079] S42: Obtain the actual cuisine in the user feedback and its index in the cuisine list;

[0080] S43: Adjust the initial confidence according to whether the prediction is accurate:

[0081] If the prediction is correct, increase the confidence of the predicted cuisine;

[0082] If the prediction is incorrect, decrease the confidence of the predicted cuisine and increase the confidence of the actual cuisine;

[0083] S44: Normalize the updated confidence.

[0084] S5: Traverse the cuisine list, calculate the weight for each cuisine, which is based on seasonal factors, user preferences, and market trends, and adjust the updated confidence according to the weight, where the updated confidence is expressed as:

[0085] ;

[0086] In the formula, is the updated confidence, is the initial confidence, is the weight, is the adjustment factor.

[0087] S6: If there are cooking methods not related to cuisine types in the user feedback, make adjustments, that is, adjust the updated confidence level according to the non-cuisine feedback; otherwise, execute step S7.

[0088] S7: Calculate the freshness score of the dish based on the updated confidence level, user preferences, seasonal factors, and market trends, where the freshness score of the dish is expressed as:

[0089] ;

[0090] In the formula, FS is the freshness score, C is the set of all cuisine types, is the weight of user preferences, is the weight of seasonal factors, is the weight of market trends.

[0091] In practical applications, a high-resolution camera uses image recognition technology to accurately analyze the dishes on the plate, identify their types and ingredients, and comprehensively evaluate the freshness and nutritional value of the food by combining historical data and the user's dietary preferences. Based on this, the system generates personalized dietary suggestions for the user, such as suggesting replacements or adjustments for foods that are about to expire or do not meet health requirements. At the same time, the system can also customize exclusive dietary plans according to the user's specific health conditions (such as diabetes, hypertension, etc.) to achieve precise health management. In addition, by continuously collecting user feedback and food status data, the system can dynamically adjust its recommendation strategies and prediction models. Every choice and preference of the user is detailedly recorded and used as valuable training data to improve the recognition and prediction capabilities of the deep learning model. This mechanism ensures that the system can evolve with the changes in user needs and continuously provide dietary suggestions and health management services that suit the individual's actual situation.

[0092] 3. Interaction Interface Implementation:

[0093] Refrigerator Touch Screen: Install a high-resolution touch screen on the outside of the refrigerator door to display real-time food freshness monitoring results, personalized dietary suggestions, alarm information, and the user operation interface. Users can view the food status, receive alarm reminders, and perform simple interaction operations (such as setting reminders, adjusting parameters, etc.) through the touch screen.

[0094] Voice Prompt Module: Integrate an intelligent voice assistant (such as technology based on Amazon Alexa or Google Assistant). When it detects that the food is about to expire or the temperature is abnormal, it broadcasts relevant information to the user by voice, providing great convenience especially for users with poor eyesight or limited mobility.

[0095] Mobile APP: Develop a smartphone application that supports iOS and Android systems. Users can remotely receive food freshness data and alarm information sent by the cloud server through the mobile APP, and perform remote control (such as adjusting the refrigerator temperature, setting reminders, etc.) and personalized parameter settings.

[0096] Through the above specific embodiments, the present invention constructs a fully functional and intelligent fully automatic intelligent fresh-keeping refrigerator system, effectively improving the user's food management experience, reducing food waste, and ensuring a healthy diet. Embodiment 2

[0097] As Figure 2 shown, the present invention provides an operation method for a fully automatic intelligent fresh-keeping refrigerator system, which uses the fully automatic intelligent fresh-keeping refrigerator system of the above Embodiment 1, and specifically includes:

[0098] Use a high-resolution camera to collect image information of the food in the refrigerator in real time;

[0099] Use an image processing chip and an edge computing device module to preprocess the collected image information;

[0100] Use RFID tags and RFID readers to embed RFID tags into each dinner plate, and read the RFID tag information through the RFID reader;

[0101] Use the cloud server to receive the RFID tag information read by the RFID reader, and use the read RFID tag information for adaptive learning and database matching algorithm processing, and generate feedback information, including food freshness prediction results, personalized diet suggestions, and alarm information;

[0102] Receive the feedback information generated by the cloud server through the interaction interface and provide real-time monitoring and alarm reminder services to users.

[0103] In this embodiment, the high-resolution camera scans the inside of the refrigerator regularly to collect food image information; the RFID reader reads the information in the RFID tag in real time. The image processing chip preprocesses the image, and the edge computing device further processes and compresses and encrypts the data, and then transmits it to the cloud server through the Bluetooth module. After receiving the data, the cloud server uses a deep learning model to predict the food freshness and verifies the result through a database matching algorithm. Subsequently, feedback information is generated, including freshness prediction results, personalized diet suggestions, and alarm information. The feedback information is displayed to the user through the refrigerator touch screen, voice prompt module, and mobile APP to provide real-time monitoring and alarm reminder services.

[0104] A method for operating a fully automatic intelligent fresh-keeping refrigerator system according to this embodiment uses the aforementioned fully automatic intelligent fresh-keeping refrigerator system. Therefore, the specific implementation manners in the method for operating the fully automatic intelligent fresh-keeping refrigerator system can be seen in the embodiment part of the fully automatic intelligent fresh-keeping refrigerator system described above. Therefore, its specific implementation manners can be referred to the descriptions of the corresponding various part embodiments. To avoid redundancy, they will not be elaborated here.

[0105] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0107] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0108] Obviously, the above embodiments are merely examples for clear illustration and not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all the implementation manners here. And the obvious changes or modifications derived therefrom still fall within the protection scope of the present invention.

Claims

1. A fully automatic intelligent fresh refrigerator system, characterized in that: include: High-resolution camera, used to collect image information of food in the refrigerator in real time; An image processing chip and edge computing device module, connected to the high-resolution camera, includes an image processing chip and an edge computing device, and is used to pre-process the collected image information; An RFID tag and RFID reader module is connected to the image processing chip and edge computing device module, wherein the RFID tag and RFID reader module includes an RFID tag and an RFID reader, and is used to embed the RFID tag into each plate and read the RFID tag information through the RFID reader; A cloud server, connected to the RFID tag and RFID reader modules, includes a deep learning module, a database management system, an image recognition module, and an anomaly detection and alarm system, for receiving RFID tag information read by the RFID reader, and using the read RFID tag information to perform adaptive learning, database matching algorithm processing, and generate feedback information, including food freshness prediction results, personalized dietary recommendations, and alarm information; The interactive interface includes a refrigerator touch screen, a voice prompt module and a mobile phone APP, which is used to receive feedback information generated by the cloud server and provide users with real-time monitoring and alarm reminder services; The deep learning module includes a cuisine prediction and recommendation system, which uses a cuisine classification strategy to perform deep learning training, specifically including: Define the main function, and the input parameters include prediction results, initial confidence, user feedback, seasonal factors, user preferences and market trends; Find the index with the highest probability from the prediction results, obtain the corresponding cuisine from the cuisine list based on the index, and use the highest probability value as the confidence score; Create a feedback dictionary containing the predicted cuisine and confidence score; Update confidence based on user feedback; Traverse the list of dishes and calculate the weight of each dish based on seasonal factors, user preferences, and market trends. Adjust the updated confidence according to the weight. If there are non-cuisine related cooking methods in user feedback, then make adjustments, i.e., adjust the updated confidence according to the non-cuisine feedback; otherwise, proceed to the next step; Calculate the freshness score of dishes based on updated confidence, user preferences, seasonal factors and market trends; The updated confidence is expressed as: In the formula, is the updated confidence, is the initial confidence, w c is the weight, ΔC is the adjustment factor; The freshness score of the dish is expressed as: Where FS is the freshness score, C is the set of all cuisines, and W is user is the weight of user preference, W seasonal is the weight of seasonal factors, W market is the weight of the market trend; Add the updated confidence and freshness scores to the feedback dictionary and return the feedback dictionary.

2. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The image processing chip and edge computing device module include an image processing chip and an edge computing device. The image processing chip includes filtering, denoising and feature extraction modules for pre-processing images; the edge computing device includes a data compression and encryption module for compressing data before transmission to reduce the amount of data, and encrypting to ensure the security of data transmission.

3. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The updating of the confidence level according to the user feedback specifically includes: Get the index of the predicted cuisine in the cuisine list; Get the actual cuisine in the user feedback and its index in the cuisine list; Adjust the initial confidence level based on whether the prediction is accurate: If the prediction is correct, increase the confidence of the predicted cuisine; If the prediction is wrong, reduce the confidence of the predicted cuisine and increase the confidence of the actual cuisine; normalize the updated confidence.

4. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The abnormality detection and alarm system includes a preset threshold judgment module, which is used to compare the real-time monitored data with the preset threshold and trigger an alarm once the preset threshold is exceeded.

5. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The refrigerator touch screen is used to display real-time food freshness monitoring results, personalized diet suggestions and alarm information, and supports users to perform interactive operations through the touch screen.

6. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The voice prompt module can broadcast food freshness status, personalized dietary suggestions and alarm information to users through voice according to the instructions of the cloud server.

7. The fully automatic intelligent fresh refrigerator system according to claim 1 is characterized in that: The mobile phone APP can remotely receive food freshness data and alarm information sent by the cloud server, and supports users to perform remote control and parameter setting through the mobile phone APP.

8. A method for operating a fully automatic intelligent fresh refrigerator system, characterized in that: The method uses the fully automatic intelligent fresh refrigerator system according to any one of claims 1 to 7, and specifically comprises: Use a high-resolution camera to collect image information of food in the refrigerator in real time; Use image processing chips and edge computing device modules to pre-process the collected image information; The RFID tag is embedded into each plate using the RFID tag and RFID reader module, and the RFID tag information is read by the RFID reader; The cloud server receives the RFID tag information read by the RFID reader, and uses the read RFID tag information to perform adaptive learning and database matching algorithm processing, and generate feedback information, including food freshness prediction results, personalized dietary recommendations and alarm information; Receive feedback information generated by the cloud server through the interactive interface and provide users with real-time monitoring and alarm reminder services.

Images